Protein scaffolds

Abstract

The present invention provides protein scaffolds and methods for preparing, screening, engineering and using protein scaffolds.

Description

[Technical Field]

[0001] FIELD OF THE INVENTION The present invention provides protein scaffolds and methods for preparing, screening, engineering and using protein scaffolds. [Background technology]

[0002] Background of the Invention The present invention relates to protein scaffolds that are useful, for example, for generating products with novel binding characteristics.

[0003] Interactions between molecules, such as proteins, and ligands are essential for multiple functions in living organisms. The ability to obtain protein molecules with binding properties to targets of interest is important in biological science and medicine. For example, the ability to diagnose disease can be facilitated by the ability to detect the presence of a target of interest associated with that pathology. In another example, it is known that modulating interactions between molecules in the body can have a therapeutic effect, and many drugs have been developed using molecules that bind to ligands, receptors, enzymes, and other targets of therapeutic interest.

[0004] Antibodies are known to generally have higher target specificity than small molecule drugs due to their relatively large and complex binding surfaces, and in therapeutic applications, antibodies are known to have a lower probability of inducing toxicity from indiscriminate binding. However, the use of antibodies is known to have drawbacks, such as the typical need for mammalian cell production to obtain full-length antibodies for therapeutic use and the generally lower tissue penetration of full-length antibodies compared to smaller molecules.

[0005] The use of antibody fragments can overcome some of these drawbacks, but antibody fragments tend to aggregate more easily and lack stability compared to full-length antibodies. For example, due to the instability of scFv molecules, time-consuming stability maturation may be necessary for some applications (Honegger A. et al., 2009), and lack of thermal stability may render scFv molecules ineffective in vivo (Willuda J. et al., 1999). In some situations, the instability of scFvs may hinder their use in engineering bispecific and multispecific constructs (Miller B. et al. 2010, Xu L. et al., 2013). Therefore, there is interest in engineering non-immunoglobulin protein molecules to overcome some of these drawbacks.

[0006] Attempts have been made to develop non-immunoglobulin protein molecules by randomizing protein surfaces to generate libraries of novel binding proteins (e.g., Binz H. et al., 2003; Vogt M., Skerra A., 2004). However, in some cases, the difficulties inherent in the randomization process may result in scaffold library members with stability only slightly better than that of antibody fragments. In addition, differences in the structure of individual scaffold proteins and the topography of the scaffold-binding surface may generally result in bias in the types of epitopes that each scaffold efficiently recognizes (Gilbreth R., Koide S., 2012). For example, the rigid, concave binding surface of DARPins may limit the structural diversity of epitopes that can be recognized by these scaffolds (Schilling J. et al., 2014; Gilbreth R., Koide S., 2012). In a related example, loop DARPin scaffolds replace the recessed binding surface of DARPins with protrusions along the middle, predicting that they bind to epitopes with different shapes than DARPins (Schilling J. et al., 2014). In another example, the cage-like structure of anticalin scaffolds tends to sandwich their binding targets, while affibodies have planar binding site architectures that tend to recognize similarly flat surfaces on their targets (Gilbreth R., Koide S., 2012). The topography of the scaffold binding surface generally correlates with the type of epitope recognized with high affinity. Summary of the Invention [Problem to be solved by the invention]

[0007] Therefore, there is a need to develop stable, small, artificial antibody-like molecules for a variety of therapeutic, diagnostic, and industrial applications. [Means for solving the problem]

[0008] Citation or discussion of a reference herein shall not be construed as an admission that such is prior art to the present invention.

[0009] Summary of the Invention The present invention provides recombinant non-naturally occurring protein scaffolds that can be used to obtain binding activity to compounds of interest. In particular, the scaffolds described herein can be used to display defined loops analogous to the complementarity determining regions ("CDRs") of antibody variable regions. These loops can be subjected to randomization or restricted evolution to generate the diversity required to bind to various target compounds.

[0010] The present invention provides a recombinant non-naturally occurring polypeptide scaffold comprising multiple alpha helices and beta strands connected by multiple loop regions and three 10 Recombinant CheB containing a helix (engineered two-fold α / β sandwich fold) c A recombinant non-naturally occurring polypeptide scaffold is provided that comprises a domain, wherein at least one loop region is a non-naturally occurring variant of the cognate loop region.

[0011] In specific embodiments, the recombinant scaffold protein (hereinafter known as the "scaffold of the invention") is a recombinant CheB having at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity outside the loop regions to SEQ ID NO:1. c domain, wherein at least one loop region is a non-naturally occurring variant of the cognate loop region of SEQ ID NO:1.

[0012] In another embodiment, the present invention also provides a polypeptide display library comprising a plurality of scaffolds of the present invention. The libraries of the present invention are useful for capturing and identifying target-binding scaffolds of the present invention.

[0013] In another embodiment, the present invention also provides isolated nucleic acid molecules encoding the scaffolds and libraries of the present invention.

[0014] In another embodiment, the present invention also provides methods of making, using, screening, optimizing and engineering the scaffolds and libraries of the present invention.

[0015] In yet another embodiment, the present invention also provides a pharmaceutical composition comprising a scaffold of the present invention.

[0016] In another embodiment, the present invention also provides methods for treating, preventing, ameliorating, detecting, diagnosing or monitoring a disease or a symptom thereof in a patient by administering a therapeutically effective amount of a scaffold of the present invention or a pharmaceutical composition comprising a scaffold of the present invention.

[0017] Brief description of the diagram For the purpose of illustrating the invention, there is shown in the drawings certain embodiments of the invention. However, the invention is not limited to the precise arrangements and instrumentalities of the embodiments shown in the drawings. [Brief explanation of the drawings]

[0018] [Figure 1] Schematic representation of the structure of a polypeptide containing the wild-type CheBc domain (PDB ID: 3SFT) (SEQ ID NO: 80). [Figure 2] Structural comparison of the wild-type CheBc domain with an example scaffold of the invention. (A) Schematic representation of the structure of a polypeptide comprising the wild-type CheBc domain (PDB ID: 3SFT) (SEQ ID NO: 80). (B) Schematic representation of a model of a test loop-graft construct (SEQ ID NO: 11) of a scaffold of the invention with three grafted artificial loops. [Figure 3]Schematic representation of an artificially cut scaffold of the present invention derived from a model of a test loop graft construct of a scaffold of the present invention with three grafted artificial loops. (A) Structural representation of the artificial loop (comprising three test loop grafts). The drawing shows individual artificial loops at (i) position 1, (ii) position 2, and (iii) position 3, consisting of SEQ ID NO: 85, SEQ ID NO: 86, and SEQ ID NO: 87, respectively. (B) Structural representation of the base of the scaffold. (C) Structural representation of a test loop graft construct of a scaffold of the present invention, consisting of a combination of three test loop grafts and a scaffold base, comprising SEQ ID NO: 11. [Figure 4] Schematic representation of the polypeptide sequence of the CheBc domain (SEQ ID NO: 1) showing a schematic representation of regions of secondary structure including alpha helices, beta strands, and 310 helices connected by loop regions (based on the annotation in Cho K. et. al, 2011). Candidate loop regions for randomization include the amino acid residues underlined in the drawing. [Figure 5] Schematic representation of the polypeptide sequence of the CheBc domain (SEQ ID NO: 1) with positions selected for test loop grafting underlined. [Figure 6] Schematic representation of aligned polypeptide sequences of test loop graft constructs of the scaffold of the invention. Sequence identity is shown to the right of their respective sequences in the alignment scheme. The alignment scheme shows constructs with test loop grafts at positions 2 and 3 (SEQ ID NO:8), test loop grafts at positions 1 and 2 (SEQ ID NO:9), test loop grafts at positions 1 and 3 (SEQ ID NO:10), and test loop grafts at positions 1, 2, and 3 (SEQ ID NO:11) aligned with the polypeptide sequence of a CheBc domain (SEQ ID NO:1) that does not contain a test loop graft. The positions of the test loop grafts in the example of SEQ ID NO:11 are underlined and labeled (1), (2), and (3) in the drawing to indicate test loop graft positions 1, 2, and 3, respectively. [Figure 7]SDS-PAGE of purified test loop graft constructs of the scaffold of the invention. Lanes are labeled at the top of the figure. Lane M: Precision Plus Protein Standard (Bio-Rad); Lane 1: CheBc domain (SEQ ID NO: 1); Lane 2: construct with test loop grafts at positions 2 and 3 (SEQ ID NO: 8); Lane 3: construct with test loop grafts at positions 1 and 2 (SEQ ID NO: 9); Lane 4: construct with test loop grafts at positions 1 and 3 (SEQ ID NO: 10); Lane 5: construct with test loop grafts at positions 1, 2, and 3 (SEQ ID NO: 11). [Figure 8] Size-exclusion chromatography profile of a purified test loop graft construct (SEQ ID NO: 11) of a scaffold of the invention with test loop grafts at positions 1, 2, and 3. The labeled arrows in the figure indicate the SEC chromatogram-derived size estimates of protein species and their relative abundance. (A) The 22.6 kDa protein peak species (corresponding to the monomer fraction) was present at 96.4%. (B) The 47.7 kDa protein peak species (corresponding to the dimer) was present at 3.6%. [Figure 9] First derivative curves of fluorescence intensity obtained by differential scanning fluorometry (DSF) of purified test loop graft constructs of scaffolds of the invention. (A) The construct with test loop grafts at positions 2 and 3 (SEQ ID NO: 8) had a Tm of 89.9°C. (B) The construct with test loop grafts at positions 1 and 2 (SEQ ID NO: 9) had a Tm of 92.0°C. (C) The construct with test loop grafts at positions 1 and 3 (SEQ ID NO: 10) had a Tm of 91.3°C. (D) The construct with test loop grafts at positions 1, 2, and 3 (SEQ ID NO: 11) had a Tm of 89.4°C. [Figure 10]Schematic representation of primers and DNA fragments used to generate assembled DNA fragments comprising the coding regions of the scaffold randomized library of the present invention. Arrows in the figure indicate the annealing position and direction of the PCR primers listed in Table 1 used to amplify and assemble the individual DNA fragments. Rectangles in the figure represent DNA fragments, and labels within the rectangles indicate the identifiers of the frameworks and loops comprising the polypeptide regions encoded by each DNA fragment. The overlapping regions of the rectangles represent overlapping complementary nucleotide sequences that allow joining of the DNA fragments by PCR. The external primers EcoRIF and AscIR contain restriction enzyme sites for EcoRI and AscI, respectively. [Figure 11] Schematic representation of overlap extension PCR assembly of DNA fragments containing sequences encoding the framework and loop regions of a randomized library of scaffolds of the present invention. Arrows in the figure indicate the annealing position and direction of the PCR primers listed in Table 1 used to amplify and assemble the individual DNA fragments. Rectangles in the figure represent DNA fragments, and labels within the rectangles indicate the identities of the framework and loop regions comprising the polypeptide regions encoded by each DNA fragment. The overlapping regions of the rectangles represent overlapping complementary nucleotide sequences that allow joining of the DNA fragments by PCR. The external primers EcoRIF and AscIR contain restriction enzyme sites for EcoRI and AscI, respectively. (A) Representation of the first PCR assembly of DNA fragments containing the coding regions of FR1+L1, FR2+L2, and FR3+L3+FR4. (B) Representation of the second PCR assembly showing the subsequent assembly of DNA fragments containing the coding region of FR1-L1+FR2-L2+FR3-L3-FR4. (C) Representation of the assembled products showing the resulting DNA fragments containing the coding regions of the randomized scaffold library. [Figure 12A]ELISA screening of individual clones from the selection output against the target PD-L1. The bar graph shows the ELISA signals obtained from clones derived from the output of the third round of phage display selection performed with the PD-L1 target at (i) a concentration of 5 nM and (ii) a concentration of 0.5 nM (plate 1 and plate 2, respectively). [Figure 12B] ELISA screening of individual clones from the selection output against the target HER2. The bar graph shows the ELISA signals obtained from clones derived from the output of the third round of phage display selection performed with the HER2 target at (i) a concentration of 5 nM and (ii) a concentration of 0.5 nM (plates 3 and 4, respectively). [Figure 13]Denaturing SDS-PAGE (4-12%) analysis of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. Proteins were purified under native conditions on cobalt agarose beads from the soluble fraction of lysates of E. coli XL1-Blue cells containing the inducible expression construct. The amount of purified protein loaded in each gel lane is equivalent to that derived from a 100 microliter overnight 2xYT broth shake flask culture. Proteins were visualized by Coomassie blue staining. Arrows indicate the approximate predicted migration position of the scaffold proteins based on molecular weight calculations. Sample lanes are labeled at the top of the figure. Lane M: Precision Plus Protein Standard (Bio-Rad); lane 1: 1-12D (SEQ ID NO: 52); lane 2: 1-12B (SEQ ID NO: 53); lane 3: 2-8B (SEQ ID NO: 54); lane 4: 1-2A (SEQ ID NO: 55); lane 5: 1-3E (SEQ ID NO: 56); lane 6: 1-12C (SEQ ID NO: 57); lane 7: 1-10B (SEQ ID NO: 58); lane 8: 1-1E (SEQ ID NO: 59); lane 9: 2-3H (SEQ ID NO: 60); lane 10: 2-6D (SEQ ID NO: 61) Number 61); Lane 11: 3-1D (SEQ ID NO: 62); Lane 12: 3-3A (SEQ ID NO: 63); Lane 13: 3-5A (SEQ ID NO: 64); Lane 14: 3-7E (SEQ ID NO: 65); Lane 15: 3-8A (SEQ ID NO: 66); Lane 16: 3-10H (SEQ ID NO: 67); Lane 17: 4-3E (SEQ ID NO: 68); Lane 18: 4-7D (SEQ ID NO: 69); Lane 19: 4-9B (SEQ ID NO: 70); Lane 20: 4-12B (SEQ ID NO: 71). [Figure 14]Denaturing SDS-PAGE (4-12%) analysis of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. Proteins were purified under native conditions on cobalt agarose beads from the soluble fraction of lysates of E. coli XL1-Blue cells containing the inducible expression construct. The amount of purified protein loaded in each gel lane is equivalent to that derived from a 100 microliter overnight 2xYT broth shake flask culture. Proteins were visualized by Coomassie blue staining. Arrows indicate the approximate predicted migration position of the scaffold proteins based on molecular weight calculations. Sample lanes are labeled at the top of the figure. Lane M: Precision Plus Protein Standard (Bio-Rad); Lane 1: 1-12B (SEQ ID NO: 53); Lane 2: 1-2A (SEQ ID NO: 55); Lane 3: 1-3E (SEQ ID NO: 56); Lane 4: 1-12C (SEQ ID NO: 57); Lane 5: 1-1E (SEQ ID NO: 59); Lane 6: 2-3H (SEQ ID NO: 60); Lane 7: 1-8E (SEQ ID NO: 72); Lane 8: 1-11C (SEQ ID NO: 73); Lane 9: 1-1B (SEQ ID NO: 74); Lane 10: 1-12E (SEQ ID NO: 75); Lane 11: 1-1F (SEQ ID NO: 76); Lane 12: 2-5H (SEQ ID NO: 77); Lane 13: 2-5C (SEQ ID NO: 78); Lane 14: 2-7B (SEQ ID NO: 79). [Figure 15]Denaturing SDS-PAGE (4-12%) analysis of purified proteins from selected ELISA-positive target-binding proteins of the scaffold of the present invention and a test loop-graft construct (SEQ ID NO: 49) derived from the chemotaxis protein CheY of Fervidobacterium pennivorans. Proteins were purified under native conditions on cobalt agarose beads from the soluble fraction of lysates of E. coli XL1-Blue cells containing the inducible expression construct. The amount of purified protein loaded in each gel lane is equivalent to that from a 100 microliter overnight 2xYT broth shake-flask culture. Proteins were visualized by Coomassie blue staining. Arrows indicate the approximate predicted migration positions of the scaffold proteins based on molecular weight calculations. Lanes are labeled at the top of the figure. Lane M: Precision Plus Protein Standard (Bio-Rad); Lane 1: 3-3A (SEQ ID NO: 63); Lane 2: 3-7E (SEQ ID NO: 65); Lane 3: 3-8A (SEQ ID NO: 66); Lane 4: 4-7D (SEQ ID NO: 69); Lane 5: Test loop graft construct derived from the chemotaxis protein CheY of Fervidobacterium pennivorans (SEQ ID NO: 49). [Figure 16A] Figure 1 shows first derivative curves of fluorescence intensity obtained by differential scanning fluorimetry (DSF) of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. The melting temperatures of each protein are indicated in the figure: (i) 1-12C: 84.5°C; (ii) 1-2A: 84.6°C; (iii) 1-3E: 85.6°C. The sequence identifiers of the proteins are SEQ ID NO:57, SEQ ID NO:55, and SEQ ID NO:56, respectively. [Figure 16B] Figure 1 shows first derivative curves of fluorescence intensity obtained by differential scanning fluorimetry (DSF) of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. The melting temperatures of each protein are indicated in the figure: (i) 1-1B: 81.3°C; (ii) 1-8E: 82.8°C; (iii) 1-12E: 88.0°C. The sequence identifiers of the proteins are SEQ ID NO:74, SEQ ID NO:72, and SEQ ID NO:75, respectively. [Figure 16C]Figure 1 shows first derivative curves of fluorescence intensity obtained by differential scanning fluorimetry (DSF) of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. The melting temperatures of each protein are indicated in the figure: (i) 1-11C: 81.0°C; (ii) 2-7B: 84.3°C. The sequence identifiers of the proteins are SEQ ID NO: 73 and SEQ ID NO: 79, respectively. [Figure 16D] Figure 1 shows first derivative curves of fluorescence intensity obtained by differential scanning fluorimetry (DSF) of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. The melting temperatures of each protein are indicated in the figure: (i) 3-7E: 85.9°C; (ii) 3-8A: 81.9°C. The sequence identifiers of the proteins are SEQ ID NO: 65 and SEQ ID NO: 66, respectively. [Figure 16E] Figure 1 shows first derivative curves of fluorescence intensity obtained by differential scanning fluorimetry (DSF) of selected purified ELISA-positive target-binding proteins from the scaffolds of the present invention. The melting temperatures of each protein are indicated in the figure: (i) 4-7D: 82.4°C; (ii) 3-3A: 81.6°C. The sequence identifiers of the proteins are SEQ ID NO: 69 and SEQ ID NO: 63, respectively. [Figure 17A] Affinity determination of selected purified PD-L1 binding proteins on scaffolds of the invention. EC50 was determined by ELISA. EC50 is shown in the figures as follows: (i) 1-12E: 165 nM; (ii) 1-1B: 108 nM; (iii) 1-12C: 91 nM. The sequence identifiers of the proteins are SEQ ID NO:75, SEQ ID NO:74, and SEQ ID NO:57, respectively. [Figure 17B] Affinity determination of selected purified HER2-binding proteins from the scaffolds of the present invention; EC50 was determined by ELISA. EC50 is shown in the figures as follows: (i) 3-8A: 4 nM; (ii) 4-7D: 65 nM; (iii) 3-7E: 22 nM. The sequence identifiers of the proteins are SEQ ID NO: 66, SEQ ID NO: 69, and SEQ ID NO: 65, respectively. [Figure 18]Schematic representation of a portion of the computer-generated output of a polypeptide sequence alignment between the CheBc domain SEQ ID NO:1 (query) and the orthologous domain SEQ ID NO:48 (subject) in the chemotaxis protein CheY of Fervidobacterium pennivorans, amino acid residues 147–337, GenBank ID: ANE42371.1. Positions selected for test loop grafting in the query and subject polypeptides are boxed. A homology of 78% amino acid residue identity was observed between the homologous regions of the two proteins. Sequence alignment was performed using the blastp algorithm on the NCBI (National Center for Biotechnology Information) website. Residue numbering for proteins from Fervidobacterium sp. corresponds to the GenBank ID numbering scheme. [Figure 19] Schematic representation of test loop grafting of protein domains from Fervidobacterium pennivorans. (A) Schematic representation of the polypeptide sequence (SEQ ID NO: 48) of amino acid residues 147-337 of the chemotaxis protein CheY from Fervidobacterium pennivorans GenBank ID: ANE42371.1, with the positions selected for test loop grafting underlined. Residue numbering in the figure corresponds to SEQ ID NO: 48. (B) Schematic representation of the polypeptide sequence (SEQ ID NO: 49) of a test loop graft construct derived from amino acid residues 147-337 of the chemotaxis protein CheY from Fervidobacterium pennivorans GenBank ID: ANE42371.1, with the artificial test loop graft underlined. Residue numbering in the figure corresponds to SEQ ID NO: 49. [Figure 20]Schematic representation of a portion of the computer-generated output of a polypeptide sequence alignment between the wild-type polypeptide sequence of the chemotaxis protein CheY from Fervidobacterium pennivorans (GenBank ID: ANE42371.1, amino acid residues 147-337 (SEQ ID NO: 48) (subject) and the corresponding test loop-graft construct (SEQ ID NO: 49) (query). Sequence alignment was performed using the blastp algorithm on the NCBI (National Center for Biotechnology Information) website. Residue numbering in the drawing of the wild-type CheY polypeptide sequence (subject) corresponds to the GenBank ID numbering scheme, while residue numbering in the drawing of the test loop-graft construct (query) corresponds to SEQ ID NO: 49. [Figure 21] Fluorescence intensity first derivative curve obtained by differential scanning fluorimetry (DSF) of purified protein from a test loop graft construct (SEQ ID NO: 49) derived from the chemotaxis protein CheY of Fervidobacterium pennivorans. The melting temperature of the protein was 78.0°C. DETAILED DESCRIPTION OF THE INVENTION

[0019] table Brief description of the table Table 1. Nucleotide sequences of PCR primers used for amplification, assembly, and cloning of scaffold framework DNA fragments, test loop graft DNA fragments, and randomized loop region DNA fragments of the scaffolds of the present invention.

[0020] Table 2. Purification yields and melting temperatures of test loop graft constructs of scaffolds of the invention having test loop grafts at positions 2 and 3 (SEQ ID NO: 8), test loop grafts at positions 1 and 2 (SEQ ID NO: 9), test loop grafts at positions 1 and 3 (SEQ ID NO: 10), and test loop grafts at positions 1, 2, and 3 (SEQ ID NO: 11). Proteins were purified from 50 mL E. coli shake flask cultures in 2xYT medium.

[0021] [Table 1]

[0022] [Table 2]

[0023] Detailed Description The protein scaffolds described herein were designed to be superior to both antibody-derived fragments and non-antibody domains. The primary advantage of the scaffolds of the present invention over antibody fragments is structural. The scaffolds are derived from structurally conserved, stable, and soluble protein domains found in a wide range of prokaryotes. As a result, the scaffolds exhibit superior folding and thermostability properties compared to antibody fragments. The creation of antibody fragments involves removing part of the antibody's native fold, often exposing amino acid residues that are buried in a hydrophobic environment in intact antibodies. These hydrophobic residues include those at the interface between the variable and constant domains. Exposure of these hydrophobic residues to solvent increases the likelihood of aggregation.

[0024] Furthermore, the scaffolds of the present invention provide the functional advantages of antibody molecules. In particular, despite the fact that the scaffolds of the present invention are not immunoglobulins, the engineered binding surface bears some designed similarity to that of the variable region of an IgG heavy chain, and includes solvent-exposed variable loops in a manner similar to antibody CDRs. Due to this structure, the scaffolds of the present invention have antigen-binding properties that are virtually similar to those of antibodies. As a result, loop randomization and shuffling strategies similar to in vivo affinity maturation processes can be used in vitro.

[0025] The scaffold of the present invention comprises the CheB methylesterase C-terminal catalytic domain (CheB), which plays an important role in chemotaxis and is therefore found in many prokaryotes. c Based on the structure of Thermotoga maritima CheB c The domain was found to be thermostable, soluble, and easy to produce, properties that facilitate the generation of a diverse collection of variants of the scaffold of the present invention that can bind to specific targets. c Analysis of the structural data of the domain (Figure 1) revealed the location of several exposed surface residues within secondary structure elements and connecting loops. These exposed surface residues are attractive candidates for introducing structural variations and generating a diverse pool of scaffold molecules with artificial binding surfaces. In the present invention, some of these residues were tested to assess their suitability for randomization.

[0026] In the present invention, surprisingly, CheB cThe domain was found to be remarkably tolerant to the randomized design of the present invention, including the grafting of unstructured and artificially long loop regions (Figures 2 and 3). As reported herein, the scaffold of the present invention was found to be monomeric and thermostable using experimental test loop grafting (Figures 8 and 9). This is surprising, since the insertion of an unnaturally long unstructured loop is generally predicted to destabilize a given domain (Schilling J. et al., 2014; Nagi A., Regan L. 1997; Regan L. 1999). This is even more surprising, since the experimental test loop in the present invention did not contain a stabilizing loop stem region (e.g., as in Schilling J. et al., 2014) or other artificially engineered loop stabilizing features. Furthermore, the thermal stability of the scaffolds of the present invention was not unreasonably affected by the grafting of two or even three unstructured test loops at various grafting positions in the recombinant test protein constructs evaluated in a randomized design (Figure 6), all of which were similarly stable (Figure 9).

[0027] Therefore, in the present invention, CheB c It has been found that the domains can be effectively exploited for the purpose of engineering highly randomized libraries of the scaffolds of the invention for the isolation of binding proteins.

[0028] The randomized design and random screening approach used herein has been found to provide a facile and efficient means of obtaining specific binders to targets of interest. Thus, the scaffolds of the present invention are expected to be particularly useful, for example but not limited to, in the development of a variety of therapeutic, diagnostic, and detection reagents against multiple targets.

[0029] As a result of the above, the present invention provides a recombinant scaffold protein comprising multiple alpha helices and beta strands connected by multiple loop regions and three 10A recombinant CheB comprising a helix (engineered two-turn α / β sandwich fold) (FIG. 4) and having at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to SEQ ID NO: 1 outside the loop regions. c domain; and at least one loop region is a non-naturally occurring variant of the cognate loop region of SEQ ID NO:1.

[0030] 3 10 The helix consists of only a few amino acid residues and can be easily inserted, substituted, or deleted using routine experiments to create a 3 10 It will be apparent to one skilled in the art that variants of the scaffolds of the invention can be generated that lack helices. Thus, one embodiment of the present invention is a scaffold comprising 3 10 This includes variants of the scaffolds of the invention that lack helices. Similarly, some of the other secondary structure elements, such as β9 (Figure 4), are small, and one of skill in the art can derive variants of the scaffolds of the invention that lack one or more of those individual secondary structure elements with minimal effort and a reasonable expectation of success.

[0031] In another specific embodiment, the scaffold of the invention is linked to secondary structure elements corresponding to non-loop regions of SEQ ID NO: 1, including 16-18 (inclusive), 29-37 (inclusive), 43-47 (inclusive), 60-61 (inclusive), 66-75 (inclusive), 80-84 (inclusive), 92-93 (inclusive), 103-107 (inclusive), 124-125 (inclusive), 135-137 (inclusive), 149-150 (inclusive), 160-162 (inclusive) of SEQ ID NO: 1. (inclusive), 173-176 (inclusive), and 180-181 (inclusive) amino acid residue positions; at least one of said loop regions is a non-naturally occurring variant of the cognate loop region of SEQ ID NO: 1; and has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to SEQ ID NO: 1 outside the non-naturally occurring variant loop region.

[0032] In another specific embodiment, a scaffold of the invention comprises four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43) connected by loop regions, wherein the framework region polypeptides have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to the cognate regions of SEQ ID NO:1; and at least one loop region is a non-naturally occurring variant of the cognate loop region of SEQ ID NO:1.

[0033] In another embodiment, the present invention also relates to nucleic acids encoding individual collective library members of randomized variants of the scaffolds of the present invention. There are various methods for introducing variation in nucleic acids encoding polypeptide sequences, including, but not limited to, incorporation of DNA fragments containing degenerate codons or mixtures of coupled trinucleotides, use of error-prone PCR, DNA fragment shuffling, and various other methods and combinations of methods, which are well known and readily available to those skilled in the art.

[0034] In a specific embodiment, coupling of trinucleotide mixtures is a well-known method that allows for increased control over the relative frequency and type of codons incorporated into randomized DNA fragments. However, due to imperfect accuracy of this method, artificial codons are also incorporated, and random deletions or insertions of trinucleotides also occur. These events provide an additional source of variation that may allow the isolation of additional scaffold variants with useful properties. Thus, scaffold variants resulting from this well-known additional source of variation constitute one embodiment of the present invention. Similarly, methods using oligonucleotides encoding degenerate NNK codons also result in various well-known artifacts. Thus, these two methods allow for the generation of diverse DNA fragments that encode both controlled and accidental variations in polypeptide sequence. As a result of the above, in a specific embodiment of the present invention, polynucleotides encoding framework region polypeptides of the scaffolds of the present invention are linked to oligonucleotides encoding variant loop region polypeptides randomized by either trinucleotide coupling or degenerate NNK codons, or a combination thereof.

[0035] In another specific embodiment, oligonucleotides encoding variant loop region polypeptides can be randomized with various degenerate codons, such as, but not limited to, NNK, NNS, NHK, VNK, NNN, or combinations thereof. In other embodiments, oligonucleotides encoding variant loop region polypeptides can be randomized, for example, with error-prone polymerases in error-prone PCR, with mutator strains of cultured cells or microorganisms, or by various other means of random or targeted mutagenesis known to those of skill in the art.

[0036] In a specific embodiment of the present invention, DNA fragments containing sequences encoding randomized loop regions of scaffolds of the present invention (e.g., those set forth in SEQ ID NOS: 16-20) are linked to DNA fragments containing sequences encoding framework regions of scaffolds of the present invention (e.g., those set forth in SEQ ID NOS: 81-84), such that the framework region polypeptides have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity with the cognate regions of SEQ ID NOS: 1. In a specific embodiment, the fragments are linked by overlap extension PCR, such as by the means depicted in the schemes shown in Figures 10 and 11. Of course, other suitable methods for generating recombinant DNA molecules can be substituted and are well known to those of skill in the art. Furthermore, numerous variations in loop length and encoded randomized amino acid composition can be experimentally tested by those of skill in the art to generate suitably randomized scaffold variants. Furthermore, it is well known to those of skill in the art that numerous possible nucleic acid sequences using different codes can be utilized to encode the same polypeptide. One of skill in the art can select codons known to be utilized at varying frequencies in different organisms, for example, as a means of optimizing production yields of the scaffolds of the invention. Thus, the nucleic acid sequences of the invention are not limited to the representative examples set forth herein.

[0037] In a specific embodiment, a scaffold of the invention comprises four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43), where the framework region polypeptide has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity with the cognate region of SEQ ID NO:1; and three loop regions L1 (SEQ ID NO:44), L2 (SEQ ID NO:44), and L3 (SEQ ID NO:47), where Xaa represents any amino acid; and L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4 to form a continuous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4. In another specific embodiment, in the L1, L2, and L3 loop regions, Xaa represents an amino acid selected from the group consisting of serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, and tyrosine. In one embodiment, a scaffold of the invention comprises a polypeptide represented, for example, by SEQ ID NO: 2. In another embodiment, it is encoded by a polynucleotide comprising, for example, the polynucleotide represented by SEQ ID NO: 5. One skilled in the art can substitute other polynucleotides with minimal effort to obtain a polynucleotide comprising the coding region of the polypeptide represented by SEQ ID NO: 2.

[0038] In another specific embodiment, a scaffold of the invention comprises four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43), where the framework region polypeptide has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to the cognate region of SEQ ID NO:1; and three loop regions L1 (SEQ ID NO:45), L2 (SEQ ID NO:46), and L3 (SEQ ID NO:47), where Xaa represents any amino acid; and L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4 to form a continuous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4. In another specific embodiment, in the L1 and L2 loop regions, Xaa represents any amino acid; for the L3 loop region, Xaa represents an amino acid selected from the group consisting of serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, and tyrosine. In one embodiment, a scaffold of the invention comprises a polypeptide represented by, for example, SEQ ID NO: 3. In another embodiment, it is encoded by a polynucleotide comprising, for example, the polynucleotide represented by SEQ ID NO: 6. One skilled in the art can substitute other polynucleotides with minimal effort to obtain a polynucleotide comprising the coding region of the polypeptide represented by SEQ ID NO: 3.

[0039] In some embodiments, the scaffolds of the present invention can be engineered to bind to targets by grafting loop regions from other binding molecules, such as, but not limited to, CDRs of antibodies or loop regions from other polypeptides with known binding activity. In other embodiments, peptides with known activity can be grafted into the scaffold, such as antimicrobial peptides, cell membrane penetrating peptides, platelet aggregation inhibitory peptides, metastasis inhibitory peptides, immunomodulatory peptides, and other peptides with known activity.

[0040] In other embodiments, affinity maturation can be performed on the scaffolds of the invention to obtain binders with stronger or weaker binding affinities or biological activities than the parent clones. Many methods for introducing sequence variations for affinity maturation purposes are well known to those skilled in the art, including, but not limited to, loop randomization, error-prone PCR, sexual PCR, and other methods. Such methods can also be used to obtain binders with altered biophysical, physiological, or other properties.

[0041] In other embodiments, the scaffolds of the invention can be randomized, for example, with a variety of different loop lengths, loop grafting positions, loop amino acid compositions, and number of grafted loops. c Recognizing the findings herein that domains are remarkably tolerant of the randomized designs of the present invention, including the grafting of unstructured and artificially long loop regions, one of skill in the art can reasonably expect to be able to readily discover other grafting solutions with minimal effort. It is therefore within the skill of the art to identify alternative randomized loop grafting schemes with a reasonable expectation of success using routine testing.

[0042] In some embodiments, variations can be introduced into one or more structural regions of the scaffolds of the present invention outside the loop regions. Recognizing the findings disclosed herein and using, for example, freely available structural data, one skilled in the art can simply by routine trial and error identify and test regions of the scaffold suitable for mutation and randomization. Thus, in one embodiment, these non-loop regions can be used to introduce structural variations and generate a diverse pool of scaffold molecules with artificial binding surfaces.

[0043] In another embodiment, randomization methods can be used to generate a diverse pool of scaffold molecules with artificial binding surfaces that include variants of loop regions, non-loop regions, and combinations thereof.

[0044] In some embodiments, variations can be introduced into previously non-randomized regions of the scaffolds of the invention to generate additional randomized libraries of the invention. Such variants can include, for example, but are not limited to, variants of previously non-randomized loop regions or scaffold framework regions to generate binders for targets with higher or lower affinity, or with altered biophysical, physiological, or other properties.

[0045] In other embodiments, shortened or extended versions of the scaffolds of the invention can be readily generated. For example, the first four amino acid residues of SEQ ID NO: 1 are identical to those of wild-type CheB. cIt is known that the four residues are not resolved in the crystal structure of a polypeptide containing the domain (PDB ID: 3SFT) (SEQ ID NO: 80). Therefore, it will be apparent to one skilled in the art that these four residues are not essential to the structure of the scaffolds of the present invention, and that one, two, three, four or more of these N-terminal residues may be freely substituted or deleted. In other embodiments, recognizing the high thermal stability of the scaffolds of the present invention disclosed herein, it will be apparent to one skilled in the art that by routine experimentation and utilizing the structural and sequence data, various N-terminal, C-terminal or internally truncated or extended versions of the scaffolds of the present invention will be generated with a reasonable expectation of success.

[0046] In another embodiment, a circularly permuted version of a scaffold of the invention can be formed by linking the N- and C-termini of the scaffold molecule and introducing a new terminus at another position. It will be clear and obvious to one skilled in the art to recognize that the N- and C-termini of the scaffold are close to each other, recognize the stability of the scaffold, and use available structural data of the scaffold to identify suitable positions for engineering an alternative terminus to generate a circularly permuted scaffold of the invention. Such constructs can be easily engineered and tested for stability without undue effort using simple routine experimentation.

[0047] In another embodiment, cyclized molecules of the scaffolds of the present invention that do not have termini can be constructed using, for example, intein-mediated trans-splicing cyclization, disulfide bond formation, isopeptide bond formation, or various chemical or molecular biological techniques known to those skilled in the art. Cyclized proteins are generally known to have improved conformational stability and resistance to exopeptidases and thermal degradation.

[0048] In another embodiment, scaffolds of the invention with improved stability can be generated by various means, such as the introduction of intramolecular disulfide bonds, intramolecular chemical cross-linking, isopeptide bond formation, and other well-known means. In another embodiment, well-known stability maturation techniques, such as those involving the generation of libraries of mutant scaffold variants and well-known selection methods based on improved stability or production yield, can be performed. In another embodiment, rational design of improved stability variants can be performed. In yet another embodiment, fusions of scaffolds of the invention with proteins known to have high solubility or stability can be used to improve the overall solubility or stability of molecules comprising scaffolds of the invention. In another embodiment, resistance to aggregation, or resistance to proteolytic degradation, or resistance to chemical degradation can be improved by these or other well-known stability enhancement and selection techniques.

[0049] CheB c Although the amino acid sequences of the domains are highly diverse, CheB from various species c It is known that there is a high structural conservation in the domain (Cho K., et al., 2011). Therefore, by using the known structural conservation, for example, by identifying orthologous loop regions corresponding to those reported in the present invention, it is possible to identify the CheB domain of Thermotoga maritima. c It is obvious to one skilled in the art to develop a randomized protein scaffold from a protein containing a domain and an orthologous domain.

[0050] In another embodiment, polypeptide or polynucleotide sequence homology searches can be used to identify proteins that can be utilized in the same manner as the present invention. For example, a routine polypeptide homology search identifies several proteins, such as the domain of the chemotaxis protein CheY of Fervidobacterium pennivorans (GenBank ID: ANE42371.1 amino acid residues 147-337) (SEQ ID NO: 48), as CheB. c It can be determined that the Fervidobacterium sp. protein domain shows homology to the nucleotide sequence of ... By further utilizing the details of the randomization strategy disclosed in this invention, artificially long test loop regions can be easily grafted (Figures 19 and 20). In this example, the resulting artificially test loop grafted Fervidobacterium sp. protein (SEQ ID NO: 49) exhibits a similar affinity to the wild-type Fervidobacterium sp. protein (SEQ ID NO: 48) and CheB, despite having three unnaturally long unstructured test loops. cAlthough the sequence identity with the domain (SEQ ID NO: 1) was less than 80%, it was easily produced and thermostable ( FIG. 21 ). Furthermore, individual loop-grafted variants of the Fervidobacterium sp. protein are predicted to show even lower sequence homology with the scaffold of the present invention. Using these findings, and knowing the results of the library construction and screening described in this invention, constructing a randomized library of, for example, the above-described Fervidobacterium sp. protein domain and obtaining a protein molecule with binding properties to a desired target with a reasonable expectation of success requires no more than routine work using techniques well known to those skilled in the art. Therefore, knowing the disclosure and teachings herein, it will require no more than basic skill and routine experimentation for those skilled in the art to readily identify and use other homologous proteins with exploitable properties similar to those of the scaffold of the present invention.

[0051] A further embodiment of the present invention is directed to means of screening the randomized libraries of the present invention for specific binding to a target molecule.

[0052] One embodiment of the present invention includes a method for obtaining a polypeptide scaffold that binds to a target, the method comprising: (a) contacting a target ligand with a randomized library under conditions that allow the formation of a scaffold:target ligand complex; and (b) obtaining from the complex a scaffold that binds to the target ligand.

[0053] The fundamental principle of all selection techniques is the physical association of the phenotype (i.e., the protein to be displayed) and the genotype (i.e., the nucleic acid encoding the protein to be displayed). Various selection techniques achieve this association using different strategies and are well known to those skilled in the art. Thus, in some embodiments, examples of such techniques include, but are not limited to, viral display, bacterial display, yeast display, mammalian cell display, mRNA display, ribosome display, cDNA display, or phage display. For example, filamentous phage display has been observed to survive extreme selection conditions, such as heat (Dudgeon K., et al., 2013) and in vivo selection in live animals (Du B., et al., 2010).

[0054] In a specific embodiment of the present invention, phage display is the selection technique of choice.

[0055] Most phage display methods use filamentous phage, although lambdoid, T4 and T7 phage display systems are also known.

[0056] In a specific embodiment of the present invention, phage display is performed using a scaffold of the present invention fused to the full-length pIII protein of filamentous M13 phage. However, phage display of the scaffold of the present invention is not limited to this example. Filamentous phage display can be performed by fusing a protein of interest to the N-terminus of the full-length pIII minor coat protein or its truncated or recombinant derivatives, which are well known to those skilled in the art and can be easily substituted.

[0057] Those skilled in the art are aware that the display of a protein of interest on the surface of phage typically relies on translocation of the protein of interest into the bacterial periplasm. In some embodiments, a fusion protein containing the protein of interest fused to a phage coat protein is translocated. In other embodiments, the protein of interest is allowed to form a disulfide bond with the phage coat protein after translocation into the periplasm. Various signal sequences suitable for translocating various proteins of interest into the periplasm have been described and are well known to those skilled in the art. Additional signal sequences can be easily identified from proteins already known to be efficiently and abundantly transported into the periplasm (Schmidt A. et al., 2015), and these clearly selected signal sequences can be easily fused to proteins of interest using established techniques known to those skilled in the art. Well-known assays, such as ELISA, can easily be used to monitor the display of proteins of interest on phage particles when fused to various signal sequences (Zhao N. et al., 2016). Thus, it will be apparent to one of skill in the art, using routine experimentation and without undue experimentation, to construct a variety of suitable phage display vectors that exploit the diversity of signal sequences for the purposes of phage display of the scaffolds of the present invention with a reasonable expectation of success.

[0058] In a specific embodiment of the invention, a modified pADL-10b phagemid vector (Antibody Design Labs) containing EcoRI and AscI restriction enzyme sites corresponding to the restriction sites of the library insert is used to construct libraries of the invention and generate fusions of scaffolds of the invention with phage pIII proteins for display on phage particles. DNA fragments encoding randomized scaffolds of the invention are cloned into the vector via the EcoRI and AscI sites to generate a large number of different recombinant phagemids encoding different randomized scaffold library members. Many other suitable restriction enzymes or other suitable methods of generating recombinant DNA are known to those of skill in the art and can be substituted.

[0059] In a specific embodiment, these recombinant phagemids are then transformed into E. coli strain XL1-Blue to generate a large number of clones that collectively encode a large number of different randomized library members of the scaffolds of the invention. Of course, many other suitable E. coli strains, such as TG1, can be readily substituted and are well known to those of skill in the art. Typically, 1 x 10 10 Library complexities on the order of 10 members can be obtained by this method.

[0060] In a specific embodiment of the invention, the library is then superinfected with M13-helper phage, e.g., VCSM13, in liquid culture according to known methods. Other helper phage strains, such as M13KO7, can be easily substituted and are well known to those skilled in the art. These helper phage strains often contain mutated DNA sequences that favor packaging of phagemids (containing the individual randomized library genes encoding the scaffolds of the invention) into mature phage particles, thus creating a physical association between the individual phage-displayed randomized library members of the scaffold and the genes that encode them.

[0061] In a specific embodiment, after this infection, the incubation temperature of the culture is reduced to produce phage particles displaying randomized library members of the scaffold of the present invention. A specific incubation temperature is a temperature known to efficiently produce fusion proteins between the scaffold of the present invention and a phage coat protein, such as 26°C. In a specific embodiment of the present invention, expression of the gene for the pIII fusion protein with the scaffold of the present invention is induced in bacterial cells from the phagemid lac promoter by adding up to 0.5 mM IPTG. Induction conditions are selected so that a significant fraction of the phages produced represent at least one randomized scaffold of the present invention. Of course, those skilled in the art can easily select other suitable experimental conditions, such as the use of other phagemid promoters, induction conditions, etc., through simple experimentation.

[0062] In another specific embodiment, the resulting mixture of recombinant phages is isolated after, for example, a 16-hour culture incubation step. Various methods for isolating phage mixtures from cultures are known, such as precipitation from bacterial culture supernatants using concentrated solutions of polyethylene glycol and NaCl. The isolated phage mixture, displaying a large number of randomized library members of the scaffolds of the invention, is then resuspended in a suitable buffer, for example, PBS with 20% (v / v) glycerol, and aliquoted for storage at -80°C. Other suitable storage buffers and storage conditions are well known to those skilled in the art and can be substituted. Typically, the phage titer obtained by this method is 10 per milliliter. 13 It is on the order of phage particles.

[0063] In another embodiment, these phage library stocks containing a large number of individual randomized scaffolds of the present invention displayed on each phage particle are used as a source to obtain high-affinity binders to desired targets by selection methods well known to those skilled in the art. There are many possible variations of this method of selecting binders to a target, such as using cells overexpressing the desired target molecule on their surface (to obtain binders to protein complexes), or selecting against bacteria or viral particles (to obtain therapeutic candidates against infectious agents), or in vivo selection in live animals (to obtain tumor or tumor-specific binders), or selection against components derived from the above. In some embodiments, these methods include the feasibility of immobilizing the target molecule on a solid support, incubating it with the phage library for a predetermined period of time, washing away unbound phage library members, and eluting phage library members that bind to the desired target molecule using an elution buffer (e.g., an acidic buffer, such as a buffer containing 100 mM glycine pH 2.2). There are many other elution methods well known to those skilled in the art, such as using buffers with a basic pH, using proteases such as trypsin, high salt buffers, competing with unlabeled target to release the binder, competing with other molecules known to bind to the target, using conditions that alter the structure of the target, and other techniques that can be readily employed.

[0064] In one embodiment, the eluted phage library members are then used to infect a suitable strain of E. coli to generate multiple copies of the enriched phage library members, which are then used in subsequent selection cycles to obtain further enrichment of binding clones.

[0065] In another embodiment of the invention, the diversity of the selection output obtained at various stages of the enrichment process can be further increased by recombining the selection output with a collection of variant loop regions to generate a population of enriched library member variants. In other embodiments, such variants can be introduced, for example, by replacing loop regions with randomized variants, randomizing additional loop regions, or generating variants of the scaffold framework. In one embodiment, DNA obtained from the selection output can be recombined using PCR with DNA fragments encoding variant loop regions, which can also be used to generate phage that display enriched library member variants. Further selection cycles using some or all of these types of enriched library member variants can be repeated to obtain a greater variety of target-binding clones with desired properties, such as increased affinity. Thus, in one embodiment, library members can be obtained that contain a mixture of loop variants generated using trinucleotide coupling or degenerate codons, as well as variations introduced by numerous other well-known methods.

[0066] In a specific embodiment of the present invention, targets are labeled with biotin, followed by subsequent capture of the biotinylated target on a surface coated with streptavidin, neutravidin, or similar biotin-binding molecules known to those of skill in the art. In some embodiments, paramagnetic beads coated with a biotin-binding surface can be used. In this method, the concentration of the target molecule can be precisely controlled (e.g., 500 nM to 50 pM or less), which facilitates the selection of high-affinity binding library members. Of course, there are numerous variations in target presentation and selection conditions that are well known and available to those of skill in the art.

[0067] In another embodiment, after multiple selection cycles, a population of phage library clones enriched for binding to the desired target is obtained. Individual phagemid clones encoding proteins of the scaffolds of the present invention with binding activity are contained within this population. Genes encoding these binders can be obtained by phagemid DNA purification or PCR amplification or various other methods known to those skilled in the art, and polypeptide sequences can be deduced from their DNA sequences, which can be easily obtained by DNA sequencing techniques well known to those skilled in the art. In another embodiment, after subcloning into an appropriate expression vector, individual scaffolds of the present invention of interest can be purified using various purification procedures from various host cells or in vitro translation systems well known to those skilled in the art. The binding affinity and specificity of individual binders can be characterized using techniques such as ELISA and surface plasmon resonance, or various other techniques well known to those skilled in the art.

[0068] Further embodiments of the present invention relate to polynucleotides encoding binding proteins or fusion proteins of the scaffolds of the present invention, vectors comprising said polynucleotides, and host cells comprising said polynucleotides and / or said vectors. The polynucleotides can be DNA, RNA, or any other analog thereof. There are many vectors and host cells known to those skilled in the art that can be used to serve multiple purposes. Such purposes may include (but are not limited to) protein production or gene therapy, or the production of viral particles that display or encode a protein of interest. Those skilled in the art can select polynucleotides, vectors, and host cells from a wide variety of well-known sources and routinely confirm their suitability.

[0069] In another embodiment of the invention, a polynucleotide comprising a coding region for a polypeptide comprising a scaffold of the invention can be used for the in vivo production of said polypeptide by administration of said polynucleotide, for example for purposes of treating disease, hi one embodiment, nucleoside-modified RNA encoding said polypeptide can be administered intravenously in a polymer- or lipid-based formulation to allow translation of the nucleic acid and production of the polypeptide in the patient's body.

[0070] In other embodiments, the present invention relates to the expression and purification of the scaffolds of the present invention and fusion proteins derived therefrom.

[0071] In one embodiment, this involves (a) isolating a nucleic acid molecule encoding a scaffold that binds to a target ligand, (b) operably linking the nucleic acid to an expression vector, and (c) expressing the nucleic acid operably linked to the expression vector in a cell.

[0072] Those skilled in the art are well aware that numerous host organisms, such as E. coli and other bacterial strains, yeast and other eukaryotic cells, including mammalian and insect cells and multicellular organisms, as well as cell-free expression systems, can be used for recombinant protein production. Furthermore, a choice between numerous expression vectors and expression methods is contemplated. The scaffolds of the present invention can be produced and purified by numerous established methods well known to those skilled in the art. The suitability of a method depends on the host cell used, the expression vector and expression strategy used, and other factors known to those skilled in the art. Thus, in some embodiments, these well-known methods of recombinant protein production can be readily employed by those skilled in the art.

[0073] In specific embodiments, purification of the scaffolds of the present invention can be simplified by fusing affinity tag peptide sequences with known affinities for certain materials. For example, certain tags, such as polyhistidine tags, FLAG tags, Strep tags, glutathione S-transferase tags, and numerous other tags, are well known to those skilled in the art and can be used in numerous affinity purification schemes. For example, these tags can be conveniently fused to recombinant proteins of interest and used to selectively capture the recombinant proteins from complex mixtures with their respective affinity partners immobilized, for example, on a resin or in a column. In another embodiment, the binding targets (or variants of the binding targets) of the scaffolds of the present invention themselves can be used by those skilled in the art in affinity purification schemes. In further specific embodiments of the present invention, such affinity tags can be removed from the recombinant binding proteins of the scaffolds of the present invention by engineering a protease cleavage site between the affinity tag and the scaffold. Numerous protease sites, such as those for tobacco etch virus (TEV) protease, thrombin, factor Xa, and numerous other protease sites, are well known to those skilled in the art and can be freely selected.

[0074] In another embodiment, the resulting scaffolds of the invention can be used unmodified or further modified by constructing various fusion proteins, such as bispecific or polyspecific binding molecules, or fusions to various other components. The fusions, and those described in the following embodiments, can be formed by, for example, but not limited to, dimerization domains, covalent isopeptide bonds, chemical crosslinks, disulfide bonds, amino acid linkers, or other means known to those of skill in the art. In one specific embodiment, the amino acid linker comprises a soluble, flexible polypeptide linker containing small and / or hydrophilic amino acids, such as glycine, serine, alanine, and threonine residues, although one of skill in the art can use numerous other amino acid combinations to generate linkers with desired properties.

[0075] Thus, in another embodiment, the present invention relates to a fusion protein comprising at least two scaffolds of the present invention to generate bispecific or bivalent fusion molecules. In another embodiment, scaffolds of the present invention can also be fused to generate multispecific and / or multivalent fusion molecules.

[0076] In another embodiment, the present invention also relates to fusion proteins comprising one or more scaffolds of the invention fused to additional binding domains, such as scFvs or other domains with binding activity, to generate multispecific and / or multivalent target binding proteins.

[0077] In additional embodiments, the present invention relates to fusions of the scaffolds of the present invention with one or more proteins that covalently or non-covalently associate to form multiprotein complexes, thus generating protein complexes with multivalent and / or multispecific binding activity, said fusions being formed by, for example, but not limited to, dimerization domains, chemical crosslinks, disulfide bonds, isopeptide bonds, amino acid linkers, or other means known to those skilled in the art.

[0078] In additional embodiments, the present invention relates to fusion proteins comprising one or more scaffolds of the invention fused to a functional Fc domain, in some specific embodiments, a human Fc domain. This may include N- or C-terminal Fc fusions or fusions to internal regions of the Fc domain, or combinations thereof. Furthermore, the resulting fusion protein may comprise different binding scaffolds of the invention with specificity for different ligand targets, thus generating bispecific or multispecific ligand-binding fusion proteins. In another embodiment, one or more scaffolds of the invention may be fused to an existing antibody to generate improved functionality, such as multispecific binding. In yet another embodiment, the Fc domain can be used to target or redirect an organism's immune response to the specific binding site of the binding protein of the invention.

[0079] In further embodiments, monovalent, bispecific, or multispecific constructs employing one or more scaffolds of the invention can be used in immunotherapeutic applications, such as the development of CAR-T cell-like therapies. Other examples include (but are not limited to) the recruitment of T cells either locally around cancer cells or systemically, or the inhibition of immune checkpoints. Those skilled in the art will recognize the numerous biological targets and immune system mechanisms that can be effectively exploited to achieve this goal.

[0080] In another embodiment, the present invention relates to fusion proteins comprising one or more scaffolds of the present invention fused to a pharmaceutically and / or diagnostically active component. Fusion proteins of the scaffolds of the present invention may include non-polypeptide components, such as non-peptide linkers, non-peptide ligands, or therapeutically or diagnostically relevant radionuclides. In specific embodiments, such pharmaceutically and / or diagnostically active components may be selected from the group comprising molecules such as cytokines, toxic compounds, chemokines, ligands, receptors, fluorescent dyes, photosensitizers, procoagulants, anticoagulants, enzymes for prodrug activation, and radionuclides. Numerous other pharmaceutically and / or diagnostically active components are known to those skilled in the art, and the present invention is not limited to the representative examples listed herein.

[0081] In another embodiment, the present invention relates to fusion proteins comprising one or more scaffolds of the present invention fused to a serum half-life modulating moiety, such as, but not limited to, polyethylene glycol (PEG), immunoglobulins, and albumin-binding peptides. One skilled in the art can select additional macromolecules or binding domains suitable as fusion partners that are also suitable for serum half-life extension purposes.

[0082] In a specific embodiment of the present invention, recombinant proteins comprising the scaffolds of the present invention essentially do not induce an immunogenic response in mammals, such as mice, rats, monkeys, or humans. Thus, one embodiment of the present invention relates to the generation of derivatives of the scaffolds of the present invention with reduced immunogenicity. Naturally, the immunogenicity of derivatives of the scaffolds of the present invention depends not only on the scaffold-derived portions but also on the randomized regions and other portions of the fusion protein. Various software and databases are available for in silico prediction of peptide binding to MHC molecules, and those skilled in the art can use such software or databases to generate derivatives of the recombinant scaffolds of the present invention with reduced immunogenicity risk, as well as fusion constructs comprising the recombinant scaffolds of the present invention. In one specific embodiment, the binding of wild-type Thermotoga maritima CheB to wild-type Thermotoga maritima CheB was predicted by searching a freely available database of peptides predicted to bind to MHC class II molecules. c It was found that proteins containing the domain (SEQ ID NO: 80) contain multiple potential T cell epitopes. c Through repeated database interrogations of domain sequence variants, it was found that modifying SEQ ID NO: 80 by incorporating the amino acid residue substitutions Met53Gln and Ser125Glu allows for a reduction in the predicted immunogenicity of the scaffold of the invention. It will be apparent to those skilled in the art that various other amino acid sequence variants can be generated to reduce the immunogenicity of the scaffold or of individual binding molecules. In other embodiments, the immunogenic risk of individual variants can be assessed using standard techniques, such as administering the recombinant protein of interest to a mammal and appropriate analysis of the immune response, which are well known to those skilled in the art.

[0083] In some embodiments, the scaffolds of the present invention comprise polypeptide sequence variants with improved developability. Such variants can include, for example, variants lacking cysteine ​​residues, variants lacking predicted N-glycosylation sites, and variants with reduced predicted degradation risks, such as predicted deamidation, isomerization, oxidation, fragmentation, and aggregation. In one specific embodiment, modification of SEQ ID NO: 80 by incorporating the amino acid residue substitution Cys161Ser allows for the generation of cysteine-free scaffolds. It will be obvious to those skilled in the art to generate additional scaffold sequence variants with improved quality using well-known techniques.

[0084] In other embodiments, the libraries of the present invention are expected to generate highly stable and soluble target-binding candidates with high specificity and affinity, making them particularly well suited for therapeutic and / or diagnostic applications. Accordingly, a highly relevant embodiment of the present invention relates to the use of the scaffolds of the present invention or fusion derivatives thereof for preparing pharmaceuticals or diagnostic tools.

[0085] In a specific embodiment, one or more scaffolds of the invention or fusion derivatives thereof are used in the preparation of a medicament or diagnostic tool for the treatment or diagnosis of a disease, in another specific embodiment, for the diagnosis or treatment of a cancer, cardiovascular, infectious or inflammatory disease.

[0086] In one specific embodiment, one or more scaffolds of the invention or fusion derivatives thereof are used in the preparation of diagnostic tools, including devices that utilize surface plasmon resonance for the detection of binding complexes, e.g., for the diagnosis of cancer or cardiovascular, infectious or inflammatory diseases.

[0087] Another embodiment of the present invention relates to pharmaceutical or diagnostic compositions comprising one or more scaffolds of the present invention or fusion derivatives thereof and suitable pharmaceutically acceptable excipients and / or carriers. Those skilled in the art can select suitable excipients and carriers from the abundant prior art and can determine their suitability using routine methods.

[0088] In another embodiment, one or more scaffolds of the present invention or fusion derivatives thereof can be administered in various forms or modes that make the compound available in an effective amount to treat or diagnose a disease in a subject suspected of having the disease. Numerous routes of administration are well known to those of skill in the art, including (but not limited to) oral, subcutaneous, intramuscular, intravenous, intraperitoneal, intradermal, spinal, topical, intranasal, intraocular, etc., and the most suitable route can be readily selected based on, for example, pharmacokinetic data obtained from in vivo experiments, common medical practice, and other sources of information widely available to those of skill in the art. In some embodiments, NMR, PET, CT, fluorescence imaging, and various other well-known in vivo imaging techniques can be used to diagnose a disease using one or more scaffolds of the present invention or derivatives thereof.

[0089] Another embodiment of the present invention relates to the co-administration or treatment with additional therapeutic agents, such as cytokines, steroids, chemotherapeutic agents, antibiotics, radiation, or other therapeutic agents and treatments known in the art. This is a well-known means of improving the therapeutic effect of drugs. The appropriate dosage, combination, and timing of additional therapies can be selected based on a variety of relevant factors known to those skilled in the art.

[0090] The present invention also provides methods for detecting compounds by utilizing the scaffolds of the present invention. Based on the binding specificity of the scaffolds obtained by library screening, such scaffolds can be used in assays to detect specific targets in a sample, for example, for diagnostic methods. In one embodiment, a method for detecting a compound comprises contacting the compound in a sample with a scaffold of the present invention under conditions that allow the formation of a compound:scaffold complex, and detecting the scaffold, thereby detecting the compound in the sample. In a further embodiment, the scaffold is labeled (e.g., radiolabeled, fluorescent, enzyme-linked, or colorimetric) to facilitate detection of the compound. In a further embodiment, the use of an in vivo implantable device utilizing the scaffold of the present invention or a derivative thereof can be used to detect a compound of interest.

[0091] The present invention also provides methods for capturing compounds using the scaffolds of the invention. Based on the binding specificity of scaffolds obtained by library screening, such scaffolds can be used in assays to capture specific targets in a sample for purification methods, etc. In one embodiment, a method for capturing a compound in a sample comprises contacting the compound in the sample with a scaffold of the invention under conditions that allow for the formation of a compound:scaffold complex, and removing the complex from the sample, thereby capturing the compound in the sample. In a further embodiment, the scaffold is immobilized to facilitate the removal of the compound:scaffold complex.

[0092] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein which equivalents are intended to be encompassed by the invention described herein.

[0093] Exemplary Embodiments 1. (i) Residue: (a) 16–18 (inclusive); (b) 29–37 (inclusive); (c) 43–47 (inclusive); (d) 60-61 (inclusive); (e) 66–75 (inclusive); (f) 80–84 (inclusive); (g) 92-93 (inclusive); (h) 103-107 (inclusive); (i) 124–125 (inclusive); (j) 135-137 (inclusive); (k) 149-150 (inclusive); (l) 160-162 (inclusive); (m) 173-176 (inclusive); (n) 180 to 181 (inclusive) 14 loop regions corresponding to the cognate loop regions of SEQ ID NO: 1, consisting of (ii) 14 loop regions linked to secondary structure elements corresponding to non-loop regions of SEQ ID NO: 1 Recombinant CheB containing c 1. A recombinant polypeptide scaffold comprising a domain, At least one of the loop regions is a non-naturally occurring variant of the cognate loop region of SEQ ID NO: 1; and Recombinant CheB c The recombinant CheB domain has at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to SEQ ID NO: 1 outside the non-naturally occurring variant loop region. c A recombinant polypeptide scaffold comprising a domain.

[0094] 2. The scaffold of embodiment 1, wherein the scaffold comprises a linear sequence of four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42) and FR4 (SEQ ID NO:43) connected by loop regions, wherein the framework regions have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to their cognate regions of SEQ ID NO:1; and wherein at least one loop region is a non-naturally occurring variant of its cognate loop region of SEQ ID NO:1.

[0095] 3. four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43), which have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to the cognate region of SEQ ID NO:1; and three loop regions L1 (SEQ ID NO:44), L2 (SEQ ID NO:44), and L3 (SEQ ID NO:47), where Xaa represents any amino acid; 3. The scaffold of embodiment 2, wherein L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4 to form a continuous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4.

[0096] 4. The scaffold of embodiment 2, comprising four framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43), which have at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity to their cognate regions of SEQ ID NO:1; and three loop regions L1 (SEQ ID NO:45), L2 (SEQ ID NO:46), and L3 (SEQ ID NO:47), wherein Xaa represents any amino acid, wherein L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4, to form a continuous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4.

[0097] 5. The scaffold of embodiment 3, wherein for the L1, L2 and L3 loop regions, Xaa represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine and tyrosine.

[0098] 6. The scaffold of embodiment 4, wherein for the L1 and L2 loop regions, Xaa represents any amino acid; and for the L3 loop region, Xaa represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, and tyrosine.

[0099] 7. A polynucleotide encoding the scaffold of embodiment 1.

[0100] 8. A cell genetically engineered to express the polynucleotide of embodiment 7.

[0101] 9. The scaffold of embodiment 1, further comprising a fluorophore, a radioisotope, a drug conjugate, an enzyme, a serum half-life extending polypeptide, or a target-binding polypeptide.

[0102] 10. The scaffold of embodiment 9, further comprising a linker having one or more glycine residues linking the scaffold to a fluorophore, radioisotope, drug conjugate, enzyme, serum half-life extending polypeptide or target-binding polypeptide.

[0103] 11. The scaffold of embodiment 1, which is capable of binding to a target other than the target bound by the additional target-binding polypeptide.

[0104] 12. The scaffold of embodiment 11, wherein the target-binding polypeptide is a polyhistidine tag.

[0105] 13. The scaffold of embodiment 11, wherein the target-binding polypeptide is a FLAG tag.

[0106] 14. A target detection device comprising the scaffold of embodiment 1.

[0107] 15. A composition comprising the scaffold of embodiment 1 and a pharmaceutically acceptable carrier.

[0108] 16. The scaffold of any of embodiments 1-6, which has been determined to bind to a target.

[0109] 17. Affinity (K) of at least 100 micromolar D 7. The scaffold of any of embodiments 1-6, wherein binding to the target is determined by

[0110] 18. The scaffold of embodiment 17, wherein said target is a cell surface antigen, a soluble antigen, an immobilized antigen, an immunosilent antigen, an intracellular antigen, an intranuclear antigen, an autoantigen, a non-autoantigen, a cancer antigen, a bacterial antigen, or a viral antigen.

[0111] 19. The scaffold of embodiment 17, which exhibits a thermal melting temperature (Tm) of at least 40°C.

[0112] 20. The scaffold of embodiment 17, conjugated to a heterologous agent, said agent being selected from the group consisting of polyethylene glycol (PEG), human serum albumin (HSA), an Fc region of an antibody, an IgG molecule, a cytotoxic drug, an imaging agent, a toxin, biotin, a nucleic acid, or a cytokine.

[0113] 21. A multidomain construct comprising the scaffold of embodiment 17, further comprising an epitope-binding domain, said epitope-binding domain being selected from the group consisting of a scaffold of an additional embodiment 17, a scaffold unrelated to embodiment 17, an antibody, an antibody fragment, a diabody, an scFv, a Fab, an Fv, or a binding peptide.

[0114] 22. The multi-domain construct of embodiment 21, which recognizes one epitope.

[0115] 23. The multi-domain construct of embodiment 21, which recognizes two epitopes.

[0116] 24. The multi-domain construct of embodiment 21, which recognizes three or more epitopes.

[0117] 25. The multidomain construct of any of embodiments 21-24, wherein said scaffold is linked to said epitope-binding domains by an IgG molecule or fragment thereof, an Fc region, a dimerization domain, a disulfide bond or an amino acid linker.

[0118] 26. The multidomain construct of any of embodiments 21-24, wherein said scaffold is covalently attached to said epitope-binding domains by an enzymatic or chemical reaction.

[0119] 27. The multidomain construct of any of embodiments 25-26, further comprising a fluorophore, a radioisotope, a drug conjugate, an enzyme or a serum half-life extending polypeptide.

[0120] 28. An isolated nucleic acid molecule encoding the multidomain construct of any of embodiments 21 to 25.

[0121] 29. The nucleic acid of embodiment 28, operably linked to an expression vector.

[0122] 30. A host cell comprising the construct of embodiment 29.

[0123] 31. A polypeptide display library comprising a plurality of variant scaffolds of any of embodiments 1-6.

[0124] 32. A collection of isolated nucleic acid molecules encoding the library of embodiment 31.

[0125] 33. The nucleic acid molecule of embodiment 32, operably linked to an expression vector.

[0126] 34. A method for obtaining a polypeptide scaffold that binds to a target, comprising: (a) contacting a target ligand with the library of any of embodiments 1-6 under conditions that allow the formation of a scaffold:target ligand complex; and (b) obtaining from the complex a scaffold that binds to the target ligand.

[0127] 35. The method of embodiment 34, further comprising randomizing at least one loop region of said scaffold of step (b) to generate a further randomized scaffold, and repeating steps (a) and (b) using said further randomized scaffold.

[0128] 36. The method of embodiment 34, further comprising randomizing at least one non-loop region of said scaffold of step (b) to generate a further randomized scaffold, and repeating steps (a) and (b) using said further randomized scaffold.

[0129] 37. A method for detecting a compound in a sample, comprising contacting the sample with a scaffold of any of embodiments 16-20 under conditions that allow the formation of a compound:scaffold complex, and detecting the complex, thereby detecting the compound in the sample.

[0130] 38. A method for capturing a compound in a sample, comprising contacting the sample with an immobilized scaffold of any of embodiments 16-20 under conditions that allow the formation of a compound:scaffold complex, and removing the immobilized scaffold, thereby capturing the compound in the sample.

[0131] 39. A method for detecting a compound in a sample, comprising contacting the sample with a multidomain construct of any of embodiments 21-27 under conditions that allow the formation of a compound:multidomain construct complex, and detecting the complex, thereby detecting the compound in the sample.

[0132] 40. A method for capturing a compound in a sample, comprising contacting the sample with an immobilized multidomain construct of any of embodiments 21-27 under conditions that allow the formation of a compound:multidomain construct complex, and removing the immobilized multidomain construct, thereby capturing the compound in the sample.

[0133] 41. A sterile, pyrogen-free composition comprising the scaffold of any of embodiments 16-20 or the multidomain construct of any of embodiments 21-27.

[0134] 42. A pharmaceutical composition comprising embodiment 41.

[0135] 43. A method of preventing, treating, managing or ameliorating a disease in a patient using the composition of embodiment 41 or 42.

[0136] 44. A method of diagnosing or imaging a disease in a patient using the composition of embodiment 41 or 42.

[0137] 45. The method of embodiment 43, further comprising an additional therapy, wherein said therapy is immunotherapy, biotherapy, chemotherapy, radiation therapy, or small molecule drug therapy.

[0138] 46. ​​The method of any of embodiments 43-45, wherein said disease is an autoimmune disease, an inflammatory disease, a proliferative disease, an infectious disease, a respiratory disease, a cardiovascular disease, a degenerative disease or a metabolic disease.

[0139] 47. A recombinant non-naturally occurring polypeptide scaffold, comprising a recombinant CheB having at least 50%, at least 60%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, or at least 99% amino acid sequence identity outside the loop region to SEQ ID NO:1. c domain; and at least one of said loop regions varies by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0140] 48. The scaffold of embodiment 47, comprising two loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0141] 49. The scaffold of embodiment 47, comprising three loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0142] 50. The scaffold of embodiment 47, comprising four loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0143] 51. The scaffold of embodiment 47, comprising five loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0144] 52. The scaffold of embodiment 47, comprising six loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0145] 53. The scaffold of embodiment 47, comprising seven loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0146] 54. The scaffold of embodiment 47, comprising eight or more loop region sequences that vary by deletion, substitution or addition of at least one amino acid from the corresponding loop region of SEQ ID NO:1.

[0147] 55. A polypeptide display library comprising a plurality of variant scaffolds of any of embodiments 47-54.

[0148] 56. A collection of isolated nucleic acid molecules encoding the library of embodiment 55.

[0149] 57. The nucleic acid molecule of embodiment 56, operably linked to an expression vector.

[0150] 58. The scaffold of any of embodiments 47-54, which has been determined to bind to a target.

[0151] 59. Affinity (K) of at least 100 micromolar D 55. The scaffold of any of embodiments 47-54, wherein binding to the target has been determined by

[0152] 60. The scaffold of embodiment 59, wherein said target is a cell surface antigen, a soluble antigen, an immobilized antigen, an immunosilent antigen, an intracellular antigen, an intranuclear antigen, an autoantigen, a non-autoantigen, a cancer antigen, a bacterial antigen, or a viral antigen.

[0153] 61. The scaffold of embodiment 59, which exhibits a thermal melting temperature (Tm) of at least 40°C.

[0154] 62. The scaffold of embodiment 59, conjugated to a heterologous agent, said agent being selected from the group consisting of polyethylene glycol (PEG), human serum albumin (HSA), the Fc region of an antibody, an IgG molecule, a cytotoxic drug, an imaging agent, a toxin, biotin, a nucleic acid, or a cytokine.

[0155] 63. A multidomain construct comprising the scaffold of embodiment 59, further comprising an epitope-binding domain, wherein said epitope-binding domain is selected from the group consisting of a scaffold of an additional embodiment 59, a scaffold unrelated to embodiment 59, an antibody, an antibody fragment, a diabody, an scFv, a Fab, an Fv, or a binding peptide.

[0156] 64. The multi-domain construct of embodiment 63, which recognizes one epitope.

[0157] 65. The multidomain construct of embodiment 63, which recognizes two epitopes.

[0158] 66. The multidomain construct of embodiment 63, which recognizes three or more epitopes.

[0159] 67. The multidomain construct of any of embodiments 63-66, wherein the scaffold is linked to the epitope-binding domains by an IgG molecule or fragment thereof, an Fc region, a dimerization domain, a disulfide bond, or an amino acid linker.

[0160] 68. The multidomain construct of any of embodiments 60-63, wherein the scaffold is covalently attached to the epitope-binding domains by an enzymatic or chemical reaction.

[0161] 69. The multidomain construct of any of embodiments 67-68, further comprising a fluorophore, a radioisotope, a drug conjugate, an enzyme or a serum half-life extending polypeptide.

[0162] 70. An isolated nucleic acid molecule encoding the multidomain construct of any of embodiments 63 to 67.

[0163] 71. The nucleic acid of embodiment 70, operably linked to an expression vector.

[0164] 72. A host cell comprising the construct of embodiment 71.

[0165] 73. A method for detecting a compound in a sample, comprising contacting the sample with a scaffold of any of embodiments 58-62 under conditions that allow the formation of a compound:scaffold complex, and detecting the complex, thereby detecting the compound in the sample.

[0166] 74. A method for capturing a compound in a sample, comprising contacting the sample with an immobilized scaffold of any of embodiments 58-62 under conditions that allow the formation of a compound:scaffold complex, and removing the immobilized scaffold, thereby capturing the compound in the sample.

[0167] 75. A method for detecting a compound in a sample, comprising contacting the sample with a multidomain construct of any of embodiments 63-69 under conditions that allow the formation of a compound:multidomain construct complex, and detecting the complex, thereby detecting the compound in the sample.

[0168] 76. A method for capturing a compound in a sample, comprising contacting the sample with an immobilized multidomain construct of any of embodiments 63-69 under conditions that allow the formation of a compound:multidomain construct complex, and removing the immobilized multidomain construct, thereby capturing the compound in the sample.

[0169] 77. A sterile, pyrogen-free composition comprising the scaffold of any of embodiments 58-62 or the multi-domain construct of any of embodiments 63-69.

[0170] 78. A pharmaceutical composition comprising embodiment 77.

[0171] 79. A method of preventing, treating, managing or ameliorating a disease in a patient using the composition of embodiment 77 or 78.

[0172] 80. A method of diagnosing or imaging a disease in a patient using the composition of embodiment 77 or 78.

[0173] 81. The method of embodiment 79, further comprising an additional therapy, wherein said therapy is immunotherapy, biotherapy, chemotherapy, radiation therapy, or small molecule drug therapy.

[0174] 82. The method of any of embodiments 79-81, wherein said disease is an autoimmune disease, an inflammatory disease, a proliferative disease, an infectious disease, a respiratory disease, a cardiovascular disease, a degenerative disease or a metabolic disease.

[0175] The invention is further illustrated by the following examples and the accompanying figures and sequence information. [Example]

[0176] Example The invention will now be described with reference to the following examples, which are provided for illustrative purposes only and the present invention should in no way be construed as being limited to those examples, but rather as encompassing any and all variations that arise as a result of the teachings provided herein.

[0177] Example 1 Randomized design of the scaffold of the present invention Optimization of the polypeptide sequence of the scaffolds of the present invention The scaffolds of the present invention desirably do not elicit an immunogenic response in mammals, e.g., humans. cThe polypeptide sequence of the protein structure containing the domain (PDB ID: 3SFT) (SEQ ID NO: 80) was screened against a database of peptides predicted to bind to the MHC-II molecules DRB1_0101, DRB1_0301, DRB1_0401, DRB1_0701, DRB1_0802, DRB1_1101, DRB1_1302, and DRB1_1501 (Jensen K. et. al, 2018). After identification of peptides within the top 5% ranking binding threshold, CheB c Repeated database interrogations of domain sequence variants were performed to identify variants with reduced predicted immunogenic potential. It was found that the amino acid residue substitutions Met53Gln and Ser125Glu relative to SEQ ID NO: 80 reduced the predicted binding of the scaffold of the present invention to MHC-II molecules. An additional Cys161Ser substitution was made to the above sequence to allow for the generation of a cysteine-free scaffold. The results of these three amino acid residue substitutions were incorporated into the polypeptide SEQ ID NO: 1.

[0178] Generation of recombinant test loop graft constructs of the scaffolds of the present invention Wild-type CheB from Thermotoga maritima c Analysis of published structural data for the protein containing the CheB domain (PDB ID: 3SFT) (SEQ ID NO: 80) (FIG. 1) allowed the identification of various loop regions. Three of these loop regions were selected for test loop grafting to compare the CheB loop regions against randomization. c The tolerance of the domain was assessed and the locations selected for test loop grafting for SEQ ID NO: 1 are shown in FIG.

[0179] A polynucleotide containing the coding region of SEQ ID NO: 1 was designed with flanking EcoRI and AscI restriction enzyme sites to generate a synthetic DNA (SEQ ID NO: 7) encoding the corresponding polypeptide SEQ ID NO: 4. This synthetic DNA (SEQ ID NO: 7) was obtained from FASMAC (Japan) and used as a PCR template.

[0180] DNA fragments encoding the framework regions FR1 (SEQ ID NO: 81), FR2 (SEQ ID NO: 82), FR3 (SEQ ID NO: 83), and FR4 (SEQ ID NO: 84) of the scaffold of the present invention were amplified from DNA SEQ ID NO: 7 by PCR using the appropriate flanking primers shown on the schematic in Figure 10 and listed in Table 1. For the DNA to be amplified encoding each framework fragment, PCR amplification was performed using 100 fmol of DNA template per 50 microliter reaction. PCR reactions were performed for 20 cycles at an annealing temperature of 55°C using Pfu Ultra II Fusion HS DNA polymerase (Agilent) according to the manufacturer's instructions.

[0181] Synthetic oligonucleotides containing the coding regions for test loop graft 1 (SEQ ID NO: 37), test loop graft 2 (SEQ ID NO: 38), and test loop graft 3 (SEQ ID NO: 39) were obtained from FASMAC (Japan). These were assembled with gel-purified DNA fragments encoding the framework regions to generate DNA fragments encoding test loop graft constructs of the scaffold of the present invention with test loop grafts at positions 2 and 3 (SEQ ID NO: 12), test loop grafts at positions 1 and 2 (SEQ ID NO: 13), test loop grafts at positions 1 and 3 (SEQ ID NO: 14), and test loop grafts at positions 1, 2, and 3 (SEQ ID NO: 15) (respective polypeptide sequence alignments are shown in Figure 6). Assembly of adjacent DNA fragments by successive rounds of overlap-extension PCR reactions was performed using 100 fmol of each DNA species per 50-microliter reaction. The PCR products were gel-purified between each amplification step and used as templates for subsequent rounds of overlap-extension PCR assembly until full-length products were obtained. The final step of overlap-extension PCR appended EcoRI and AscI restriction sites to the assembled product using primers EcoRIF and AscIR (Table 1). The PCR product was gel-purified and digested with a 20-fold excess of EcoRI-HF and AscI (New England Biolabs) at 37°C for 3 hours, after which the digested DNA was column-purified using the Wizard SV gel and PCR Clean-Up System (Promega). The resulting DNA insert was cloned into a modified pQE-80L vector (QIAGEN) containing the corresponding EcoRI and AscI cloning sites and transformed into E. coli XL1-Blue (Agilent). The isolated plasmid clones were sequenced, and clones encoding polypeptides comprising recombinant test loop graft constructs of the scaffold of the present invention having test loop grafts at positions 2 and 3 (SEQ ID NO: 8), test loop grafts at positions 1 and 2 (SEQ ID NO: 9), test loop grafts at positions 1 and 3 (SEQ ID NO: 10), and test loop grafts at positions 1, 2, and 3 (SEQ ID NO: 11) were identified.

[0182] Expression and purification of recombinant test loop graft constructs of the scaffolds of the invention Glycerol stocks of sequence-verified clones were used to inoculate 50 mL cultures of 2xYT medium containing 50 micrograms / mL kanamycin and 0.1% glucose, and the OD was measured at 37°C with vigorous shaking. 600 The culture was grown until its pH reached 0.5. The culture was then chilled on ice, IPTG was added to 0.5 mM, and the culture was grown overnight at 27°C with vigorous shaking. The culture was centrifuged at 3000 x g for 10 minutes at 4°C, and the cell pellet was resuspended in 27 mL of ice-cold PBS (pH 7.4) containing 300 mM NaCl. 3 mL of 10x bugbuster reagent (EMD Millipore) was then added, and the cells were lysed on ice for 30 minutes. The cell lysate was then centrifuged at 12,000 x g for 30 minutes at 4°C, and the supernatant containing the cell lysate was collected. These were then bound to a 1 mL bed volume of pre-equilibrated Talon Cell-thru resin (Clontech), and purification was continued according to the manufacturer's instructions, followed by elution in a volume of 5 mL. Purified proteins were visualized by running 10 microliter aliquots on NuPAGE 4-12% SDS-PAGE gels (Invitrogen) and staining with Coomassie blue stain (Figure 7). The eluted proteins were buffer exchanged with PBS (pH 7.4) by repeated centrifugation through an Amicon Ultra-4 10,000 MWCO column (Millipore) according to the manufacturer's instructions, and the proteins were collected in approximately 1 mL volumes. Protein concentrations were calculated based on the measured absorbance at 280 nm compared to the extinction coefficient predicted from the amino acid sequence deduced from DNA sequence data. Post-purification yields for the test proteins are reported in Table 2.

[0183] Thermal stability of recombinant test loop graft constructs of the scaffolds of the present invention The thermal stability of purified proteins comprising recombinant test loop graft constructs of the scaffolds of the present invention was determined by DSF (differential scanning fluorometry) measurements using SYPRO Orange dye (Merck) at a scan rate of 0.5°C / min using 500 micrograms / mL of protein in PBS buffer (pH 7.4) (Figure 9). The protein melting temperatures were determined from the temperature of the maximum of the first derivative of the fluorescence intensity curve. The protein melting temperatures are reported in Table 2. A melting temperature of approximately 90°C was observed for all recombinant test loop graft constructs of the scaffolds of the present invention evaluated, with only a few degrees of difference in thermal stability observed between the different test constructs (Tm range: 89.4-92.0°C). This suggests that the scaffolds of the present invention may support a wider variety of modifications than the test loop graft constructs evaluated herein, and that those skilled in the art, aware of these findings, can have a reasonable expectation of success in generating and utilizing such variants through routine experimentation.

[0184] Determination of monomer fraction The monomer fraction of purified protein comprising a test loop graft construct (SEQ ID NO: 11) of a scaffold of the present invention having test loop grafts at positions 1, 2, and 3 was determined by size exclusion chromatography after two weeks of storage at 4°C in PBS buffer (pH 7.4) at 1 mg / mL, followed by two weeks of storage at room temperature. SEC was performed on a Superdex 75 10 / 300 column (GE Lifesciences) using 500 micrograms of protein in PBS buffer (pH 7.4) (Figure 8). The purified protein was found to be 96.4% monomeric. The experimentally determined monomer mass was 22.6 kDa, which closely matches the predicted molecular weight of approximately 24 kDa.

[0185] Example 2 Construction of randomized libraries of scaffolds of the invention DNA fragments SEQ ID NO:81, SEQ ID NO:82, SEQ ID NO:83, and SEQ ID NO:84, encoding framework regions FR1 (SEQ ID NO:40), FR2 (SEQ ID NO:41), FR3 (SEQ ID NO:42), and FR4 (SEQ ID NO:43), respectively, were amplified from DNA SEQ ID NO:7 by PCR using the appropriate flanking primers shown on the schematic in Figure 10 and listed in Table 1. For the DNA to be amplified encoding each framework region, eight individual PCR amplifications were performed using 100 fmol of DNA template per 50-microliter reaction. PCR reactions were performed for 20 cycles at an annealing temperature of 72°C using Phusion Hot Start Flex DNA Polymerase (New England Biolabs) according to the manufacturer's instructions. PCR products were gel-purified using the Wizard SV Gel and PCR Clean-Up System (Promega).

[0186] Trinucleotide-coupled oligonucleotides SEQ ID NO:16, SEQ ID NO:17, and SEQ ID NO:18, containing DNA sequences encoding trinucleotide-randomized loop regions L1 (SEQ ID NO:44), L2 (SEQ ID NO:44), and L3 (SEQ ID NO:47) (Figure 10), respectively, were obtained from ELLA Biotech GmbH (Germany) and dissolved to 50 micromolar in TE buffer (10 mM Tris, 5 mM EDTA pH 8.0).

[0187] The first stage of assembly of the randomized loop region-encoding DNA fragments into framework DNA-encoding fragments (first PCR assembly) was performed by overlap-extension PCR using appropriate primers listed in Table 1. Three separate PCR assembly schemes were performed to assemble DNA fragments containing the coding regions of FR1+L1 (using primers FR1F and L1R) to generate the FR1-L1 coding fragment, FR2+L2 (using primers FR2F and L2R) to generate the FR2-L2 coding fragment, and FR3+L3+FR4 (using primers FR3F and FR4R) to generate the FR3-L3-FR4 coding fragment, as shown schematically in Figure 11. A total of five replicate 100-microliter PCR reactions were performed for each scheme, each containing 500 fmol of randomized loop fragment DNA template. PCR was performed for 18 cycles using Phusion Hot Start Flex DNA polymerase (New England Biolabs) at an annealing temperature of 72° C. Individual fragment assemblies were gel purified as described above.

[0188] Final assembly of the above fragments into full-length randomized library-encoding DNA fragments (second PCR assembly), including the coding region of FR1-L1-FR2-L2-FR3-L3-FR4 containing three trinucleotide-coupled randomized loop regions, was performed by overlap-extension PCR using 125 fmol of each of the above fragment assemblies per 50-microliter PCR reaction tube with external primers EcoRIF and AscIR (Table 1). A total of 176 PCR reaction tubes were used to amplify the full-length fragment assemblies for 20 cycles at an annealing temperature of 72°C.

[0189] Generation of full-length randomized library-encoding DNA fragments containing NNK-randomized coding loop regions was carried out as follows: Oligonucleotides SEQ ID NO:19 and SEQ ID NO:20, containing DNA sequences encoding NNK-randomized loop regions L1 (SEQ ID NO:45) and L2 (SEQ ID NO:46), respectively, were obtained from FASMAC (Japan). Trinucleotide-coupled oligonucleotide SEQ ID NO:18, containing DNA sequence encoding trinucleotide-randomized loop region L3 (SEQ ID NO:47), was obtained from ELLA Biotech GmbH (Germany).

[0190] Assembly of randomized loop region-encoding DNA fragments into framework DNA-encoding fragments was performed as described above, except that the final assembly of fragments into full-length randomized library-encoding DNA fragments (second PCR assembly) was performed by overlap-extension PCR with 125 fmol of each assembled fragment per 50-microliter PCR reaction tube using external primers EcoRIF and AscIR (Table 1). A total of 112 PCR reaction tubes were used to amplify the full-length fragment assemblies for 20 cycles at an annealing temperature of 72°C.

[0191] PCR products corresponding to the full-length randomized library-encoding DNA fragments containing three trinucleotide-coupled randomized loop regions and the full-length randomized library-encoding DNA fragments containing the NNK randomized loop region were individually gel-purified as described above, and these two libraries were then cloned and displayed separately on phage.

[0192] In total, 72 micrograms of gel-purified full-length randomized library-encoding DNA fragments containing three trinucleotide-coupled randomized loop-encoding regions were digested with 1400 U each of EcoRI-HF and AscI (New England Biolabs) in a 2.4 mL volume at 37°C for 7 hours to generate library inserts for ligation. Additionally, 48 micrograms of gel-purified full-length randomized library-encoding DNA fragments containing NNK randomized loop-encoding regions were digested with 960 U each of EcoRI-HF and AscI (New England Biolabs) in a 1.6 mL volume at 37°C for 7 hours to generate library inserts for ligation. The resulting digested insert DNAs were then separately column-purified using the Wizard SV gel and PCR Clean-Up System (Promega).

[0193] A modified pADL-10b phagemid vector (Antibody Design Labs) containing EcoRI and AscI restriction enzyme sites was used for library construction and generation of fusions of the scaffolds of the present invention with the phage pIII protein for display on phage particles. A 1 mg aliquot of this vector was digested in a 4 mL volume with 3000 U each of EcoRI-HF and AscI (New England Biolabs) at 37°C for 3 hours to generate digested vector DNA for ligation. DNA fragments corresponding to the digested vector DNA were gel-purified as described above.

[0194] Individual ligations were set up in 2.5 mL volumes with 15.5 micrograms of the digested vector and 5 micrograms of the digested insert (insert:vector at an approximate 2:1 molar ratio) using 10,000 U of T4 DNA ligase (New England Biolabs) overnight at 16°C. The ligations were heated to 65°C for 15 minutes, and the ligation buffer was exchanged for milliQ ultrapure water by repeated rotation and rehydration using Amicon Ultra 30K MWCO columns (Millipore).

[0195] Electrocompetent E. coli strain XL1-Blue (Agilent) was cultured at OD 600 The culture was prepared from a 1-liter culture vigorously grown in TB medium until a pH of 0.8 was reached. The culture was rapidly chilled on ice and centrifuged at 3000 × g at 4°C to recover the cell pellet. The cell pellet was washed three times by repeated resuspension in ice-cold milliQ ultrapure water, recovered by centrifugation as above, and finally resuspended in a final volume of 9 mL of ice-cold 10% glycerol. This was aliquoted to 1.5 mL volumes on ice and used for electroporation of the ligated DNA described above using a total of six flat-packed chamber 1.5 mL electroporation cuvettes (Harvard Apparatus) shocked with an exponentially decaying pulse of 1960 volts. The resulting transformed E. coli was grown in 250 mL of SOC medium at 37°C for 1 hour, and the harvested cell pellet was plated onto a total of eight 500 cm plates containing TB agar, 2% glucose, and 100 micrograms / mL carbenicillin. 2The resulting clones were spread onto selective media plates and incubated at 37°C for 16 hours. The resulting clones were harvested by scraping them from the plates onto 2xYT medium containing 2% glucose and 100 micrograms / mL carbenicillin, and glycerol was added to a final volume of 15%. The resuspended cells were divided into 1 mL aliquots and stored at -80°C as library E. coli glycerol stocks until further use. This ligation and transformation process was repeated 13 times, and approximately 1.5 x 10 clones were obtained for the DNA fragments encoding the three trinucleotide-coupled randomized loop regions, estimated from the number of colonies arising from diluted E. coli of the culture aliquots after electroporation. 10 and 2.5 × 10 for the DNA fragment encoding the NNK randomized loop region. 9 We generated libraries of different complexities, respectively.

[0196] Example 3 Phage display of randomized libraries of scaffolds of the invention Aliquots of E. coli glycerol stocks of the scaffold randomized libraries of the invention described above were thawed and diluted in a total of 5 liters of 2xYT medium (for libraries containing three trinucleotide-coupled randomized loop regions) and 1 liter of 2xYT medium (for libraries containing NNK randomized library loop regions) to an OD of 0.2. 600 Carbenicillin and glucose were added to a final concentration of 100 micrograms / mL carbenicillin and 0.1% (w / v) glucose. The cultures were then incubated at 37°C with vigorous shaking until OD 600 Grow until the β-actin concentration reaches 0.6, then add 2 × 10 per liter culture. 12E. coli was infected by adding VCSM13 helper phage (Agilent). Infection proceeded for 1 hour at 37°C, after which the culture was chilled on ice and kanamycin was added to 30 micrograms / mL. IPTG was also added to 0.5 mM to induce expression of the scaffold-pIII fusion gene. The culture was then grown overnight at 26°C.

[0197] The cultures were centrifuged at 8000 × g for 20 minutes at 4°C, and the supernatant containing the phage particles was carefully collected. They were chilled on ice, and 0.25x the volume of 20% (w / v) PEG, 2.5 M NaCl was added and left on ice for 1 hour to precipitate the phage. The precipitate was centrifuged at 8000 × g for 30 minutes at 4°C, and the phage pellet was washed by resuspending it in PBS. They were then precipitated with 20% (w / v) PEG, 2.5 M NaCl as described above, and washed again by resuspending the phage pellet in PBS. They were again precipitated and resuspended in PBS, and glycerol was added to a final volume of 20%. The phage were then aliquoted into 0.6 mL volumes (for libraries containing three trinucleotide-coupled randomized loop regions) and 0.09 mL volumes (for libraries containing NNK randomized loop regions) and stored at −80°C as phage library stocks.

[0198] Example 4 Selection of binders from scaffold phage display libraries First round phage display selection of a library of scaffolds of the invention against targets PD-L1 and HER2 Biotinylated human PD-L1 antigen containing the human IgG1 Fc domain and biotinylated human HER2 antigen containing part of the HER2 extracellular domain (Acro Biosystems) were used individually as panning targets as follows.

[0199] For each target, aliquots of two phage library stocks (one containing three trinucleotide-coupled randomized loop regions and one containing NNK randomized loop regions) were thawed and combined. Blocking reagent (BSA added to 3% (w / v) and Tween-20 added to 0.05% (v / v) in PBS) was added to a final volume of 1 mL. For PD-L1 target selection, non-biotinylated human IgG1 Fc protein (Acro Biosystems) was also added as a blocking reagent to a final concentration of 1000 pM. A 200 microliter aliquot of Dynal M-280 dynabeads suspension (Invitrogen) was then washed twice in PBS containing 3% BSA and 0.05% Tween-20. The blocked phage was added to the washed dynabeads and rotated at 4°C for 1 hour to remove phage bound to the beads. The beads were then collected using a magnet, and the phage supernatant was transferred to a new tube. Biotinylated antigen was then added to the phage supernatant to a final concentration of 50 nM, and the mixture was rotated overnight at 4°C to allow the phage to bind to the antigen. After this, 100 microliters of the dynabeads suspension was washed twice in PBS, 3% BSA, and 0.05% Tween-20, and the supernatant was discarded. The phage and antigen mixture was then added to the tube containing the washed dynabeads, and the mixture was rotated at 4°C for 30 minutes to capture the biotinylated antigen onto the dynabeads. The dynabeads were then collected using a magnet to pull down the phage bound to the biotinylated antigen captured on the surface of the dynabeads, and the beads were washed three times with 1 mL of a solution of PBS, 3% BSA, and 0.05% Tween-20. The beads were then washed three times with PBS, 0.05% Tween-20 as above, followed by three washes with PBS. The beads were then collected by magnet, the supernatant discarded, and bound phage eluted by incubating the beads with 300 microliters of 100 mM glycine, 500 mM NaCl, pH 2.2 for 10 minutes.The beads were then captured again by magnet and the supernatant containing the eluted phage was diluted to OD 2 in 2xYT medium in a volume of 15 mL. 600 The phage was added to E. coli XL1-Blue at 0.7 μg / mL. This was incubated at 37°C for 45 minutes to allow the phage to infect the E. coli, and the culture was then centrifuged at 3000 × g for 10 minutes at 4°C. The cell pellet was then resuspended in 2xYT medium and plated on a large 500cm plate containing TB agar, 2% glucose, and 100 micrograms / mL ampicillin. 2 The cells were spread on selective media plates for 16 hours at 37°C. Diluted aliquots of the infection output were also plated out as above to obtain colony counts, which were used to estimate the number of clones obtained from the selection. Approximately 4.8 x 10 5 clones and 3.1 x 10 5 Clones were obtained from the first round panning outputs of PD-L1 and HER2, respectively. The next day, colonies from each of the plates were harvested by scraping the plates with LB medium containing 1% glucose, 100 micrograms / mL ampicillin, and 15% glycerol, and the resuspended cells were divided into 0.5 mL aliquots and stored at -80°C as E. coli glycerol stocks of the first round panning selection outputs until further use.

[0200] Randomization of L1 and L3 loops of first-round library selection output Randomization of the L1 and L3 loops for each selection output was performed as follows. A pool of recombinant phagemid DNA was isolated from an aliquot of the E. coli glycerol stock of the first-round panning selection output described above using the FastGene Plasmid Mini Kit (Nippon Genetics, Japan). For L1 loop randomization, the pool of library DNA fragments from the first-round selection output, which did not contain the L1 loop-coding region, was amplified using primers FR2F and AscIR (Table 1). For each pool of DNA fragments to be amplified, PCR amplification was performed using 8 fmol of DNA template in each of two 50-microliter reactions. PCR amplification was performed for 18 cycles at an annealing temperature of 72°C using Phusion Hot Start Flex DNA polymerase (New England Biolabs) according to the manufacturer's instructions. The PCR products were gel-purified using the Wizard SV Gel and PCR Clean-Up System (Promega). These fragments were joined by PCR to the DNA fragment encoding FR1-L1 (Figure 11), which contained the randomized loop 1 coding region generated by trinucleotide coupling during library construction (Example 2). Primers EcoRIF and AscIR (Table 1) were used to join the fragments in seven 50-microliter PCR reactions for each target, each containing 125 fmol of each template, at an annealing temperature of 72°C for 12 cycles. PCR products were gel-purified as described above. For L3 loop randomization, primers EcoRIF and L2R (Table 1) were used to PCR-amplify the library scaffold DNA of the first-round selection output, which lacked the L3 loop coding region. For each pool of DNA fragments to be amplified, two individual PCR amplifications were performed using 8 fmol of DNA template in each of two 50-microliter reactions.PCR amplification was performed for 18 cycles at an annealing temperature of 72°C using Phusion Hot Start Flex DNA polymerase (New England Biolabs) according to the manufacturer's instructions. PCR products were gel-purified using the Wizard SV Gel and PCR Clean-Up System (Promega). These fragments were PCR-ligated to a DNA fragment encoding FR3-L3-FR4 (Figure 11), which contained a randomized loop 3 coding region generated by trinucleotide coupling and was previously generated during library construction (Example 2). Primers EcoRIF and AscIR (Table 1) were used to ligate the fragments for 12 cycles at an annealing temperature of 72°C in seven 50-microliter PCR reactions for each target, each containing 125 fmol of each template. PCR products were gel-purified as described above. The resulting PCR products containing the DNA pool encoding the first-round panning selection output in which loops 1 and 3 were randomized were combined, and 5 micrograms of this DNA was digested with 100 U each of EcoRI-HF and AscI (New England Biolabs) at 37°C for 4 hours. The digested DNA was then column-purified using the Wizard SV gel and PCR Clean-Up System (Promega) to generate insert DNA of the first-round panning selection output in which loops 1 and 3 were randomized. Ligation was performed overnight at 16°C in a volume of 1.25 mL using 5,000 U of T4 DNA ligase (New England Biolabs) with 7.75 micrograms of the digested phagemid vector (used in the library construction in Example 2) and 2.5 micrograms of the insert DNA of the first-round panning selection output in which loops 1 and 3 were randomized (insert:vector molar ratio of approximately 2:1). The ligation was heated at 65°C for 10 min, and the ligation buffer was exchanged for milliQ ultrapure water by repeated rotation and rehydration using an Amicon Ultra 30K MWCO column (Millipore).These ligations were then transformed into electrocompetent E. coli strain XL1-Blue (Agilent), and each transformed culture was harvested and plated on a large 500 cm plate containing TB agar, 2% glucose, and 100 micrograms / mL ampicillin using the procedure described for the library construction protocol (Example 2). 2 The DNA encoding the loop 1 and loop 3 randomized first-round panning selection output from PD-L1 selection was estimated from the colony counts of aliquots of diluted E. coli cultures after electroporation, and the DNA encoding the loop 1 and loop 3 randomized first-round panning selection output from PD-L1 selection was approximately 4 × 10 8 and 9 x 10 for DNA encoding the first round panning selection output in which loop 1 and loop 3 derived from HER2 were randomized. 7 A library with a complexity of 1000 was obtained. The resulting clones were harvested by scraping them from the plates in 2xYT medium containing 2% glucose, 100 micrograms / mL carbenicillin, and glycerol was added to a final volume of 15%. The resuspended cells were divided into 1 mL aliquots and stored at -80°C as E. coli glycerol stocks of randomized first-round panning selection output until further use.

[0201] Second and third rounds of phage display selection against targets PD-L1 and HER2 Thaw an aliquot of each E. coli glycerol stock of the randomized first-round panning selection output and dilute in 500 mL of 2xYT medium to an OD of 0.2. 600 Additionally, an aliquot of the E. coli glycerol stock from the first round of panning selection output was thawed and diluted in 100 mL of 2xYT medium to obtain an OD of 0.2. 600To each culture, carbenicillin and glucose were added to a final concentration of 100 micrograms / mL carbenicillin and 0.1% glucose. The cultures were then incubated at 37°C with vigorous shaking until OD 600 Grow until the β-actin concentration reaches 0.6, then add 2 x 10 per 100 mL of culture. 11 E. coli was infected by adding VCSM13 helper phage (Agilent). Infection proceeded for 1 hour at 37°C, after which the culture was chilled on ice and kanamycin was added to 30 micrograms / mL. IPTG was also added to 0.5 mM to induce expression of the scaffold-pIII fusion gene. The culture was then grown overnight at 26°C.

[0202] A 50 mL aliquot of each overnight culture was centrifuged at 8000 × g for 20 minutes at 4 ° C., and the supernatant containing the phage particles was carefully collected and filtered through a 0.45-micrometer filter (Sartorius). These filtrates were chilled on ice, and 0.25 × volume of 20% (w / v) PEG, 2.5 M NaCl was added and left on ice for 1 hour to precipitate the phages. The precipitates were centrifuged at 8000 × g for 30 minutes at 4 ° C., and the phage pellets were individually resuspended in 500 microliters of PBS. For each target, 62 microliters of phage from the randomized first-round panning selection output and 438 microliters of phage from the first-round panning selection output were combined to obtain a combined phage mixture in a volume of 500 microliters.

[0203] Each of these combined phage mixtures was then used to perform a second round of selection against their respective biotinylated antigen as described above, except that the biotinylated antigen was added to the phage supernatant to a final concentration of 10 nM. 5 clones and 1.1 x 10 5clones were obtained from the second-round panning outputs of PD-L1 and HER2, respectively. The next day, colonies from each of the plates were harvested by scraping the plates with LB medium containing 1% glucose, 100 micrograms / mL ampicillin, and 15% glycerol, and the resuspended cells were divided into 0.5 mL aliquots and stored at -80°C as E. coli glycerol stocks of the second-round panning selection outputs until further use.

[0204] Thaw an aliquot of the E. coli glycerol stock from the second round of panning selection output above and dilute in 100 mL of 2xYT medium to an OD of 0.2. 600 To each culture, carbenicillin and glucose were added to a final concentration of 100 micrograms / mL carbenicillin and 0.1% glucose. The cultures were then incubated at 37°C with vigorous shaking until OD 600 Grow until the β-actin concentration reaches 0.6, then add 2 x 10 per 100 mL of culture. 11E. coli was infected by adding 100 VCSM13 helper phage (Agilent). Infection proceeded for 1 hour at 37°C, after which the cultures were chilled on ice and kanamycin was added to 30 micrograms / mL. IPTG was also added to 0.5 mM to induce expression of the scaffold-pIII fusion gene. The cultures were then grown overnight at 26°C. A 50 mL aliquot of each overnight culture was centrifuged at 8000 × g for 20 minutes at 4°C, and the supernatant containing the phage particles was carefully collected and filtered through a 0.45 micrometer filter (Sartorius). These filtrates were chilled on ice, and 0.25x the volume of 20% (w / v) PEG, 2.5 M NaCl was added and allowed to stand on ice for 1 hour to precipitate the phage. The precipitates were centrifuged at 8000 x g for 30 minutes at 4°C, and the phage pellets from each panning output were individually resuspended in 1 mL of PBS. Each of these phage preparations was divided into two separate tubes (containing 500 microliters of phage per tube) and used to perform a third round of panning with their respective biotinylated antigens at final concentrations of 5 nM and 500 pM. Panning was performed as described above for each sample. For the third round of selection of PD-L1, approximately 3.3 x 10 6 clones and 8.2 x 10 5 Approximately 1.5 x 10 clones were obtained from the 5 nM and 500 pM panning outputs, respectively. For the third round of selection for HER2, approximately 1.5 x 10 clones were obtained from the 5 nM and 500 pM panning outputs, respectively. 6 clones and 3.2 x 10 5 Clones were obtained from the 5 nM and 500 pM panning outputs, respectively. Colonies from each plate were harvested by scraping the plate with LB medium containing 1% glucose, 100 micrograms / mL ampicillin, and 10% glycerol, and the resuspended cells were divided into 0.5 mL aliquots and stored at -80°C as E. coli glycerol stocks of the third round panning selection output until further use.

[0205] Example 5 Identification of target-binding clones of the scaffold of the present invention by ELISA A pool of recombinant phagemid DNA was isolated from each aliquot of the glycerol stock from the third round of selection output described above using the FastGene Plasmid Mini Kit (Nippon Genetics, Japan). Phagemid DNA (5 micrograms) was digested with 50 units each of EcoRI-HF and AscI (New England Biolabs) at 37°C for 2 hours, and the insert DNA was gel-purified as described above. A 100 ng aliquot of the resulting DNA insert was ligated with 100 ng of modified pQE-80L (QIAGEN) vector containing the corresponding EcoRI and AscI cloning sites using 400 U of T4 DNA ligase (New England Biolabs) in a volume of 20 microliters at 16°C for 2 hours. The ligation mixture was then heated to 65°C for 10 minutes and used to transform chemically competent E. coli XL1-Blue (Agilent) according to the manufacturer's instructions and plated out overnight at 37°C on 2xYT agar plates containing 2% glucose and 50 micrograms / mL kanamycin. The following day, 95 individual colonies were picked from each transformation output and grown for 4 hours at 37°C with gentle shaking in 96-well plates ("expression plates") containing 110 microliters per well of 2xYT medium containing 0.1% glucose and 50 micrograms / mL kanamycin (plate well 12H of each "expression plate" was not inoculated with bacteria). After this, 10 microliters from each well was transferred to a replicate 96-well plate ("storage plate") containing 100 microliters per well of TB medium containing 1% glucose and 50 micrograms / mL kanamycin. The storage plate was grown overnight at room temperature with shaking. IPTG was added to each well of the expression plate to a final concentration of 0.5 mM IPTG to induce expression of individual clones, while expression was allowed to proceed overnight at room temperature with gentle shaking.The next day, 100 microliter aliquots of TB medium containing 1% glucose, 50 micrograms / mL kanamycin, and 30% glycerol were added to each well of the storage plate. The storage plate was then sealed with an adhesive aluminum foil sheet and frozen at -80°C to serve as glycerol stocks of individual clones. After this, 40 microliters of lysis buffer (24.7 g / L boric acid, 18.7 g / L NaCl, 1.49 g / L EDTA, pH 8.0) containing 2.5 mg / mL human lysozyme (Merck) and 20 U / mL benzonase (Merck) was added to each well of the expression plate and shaken for 1 hour at room temperature. Next, 40 microliters of 12.5% ​​(w / v) skim milk powder in PBS was added to each well of the expression plate (final concentration: 2.5% (w / v) skim milk), and the plate was shaken for 30 minutes at room temperature. The resulting blocked cell lysates containing the scaffolds of the present invention from the individually expressed clones were screened for binding to the target antigen by ELISA as follows: The antigen was dissolved in PBS to 1 microgram / mL, and 100 microliters was coated onto the surface of each well of a 96-well MaxiSorp plate (Nunc) overnight at 4°C. The next day, the wells of the MaxiSorp plate were washed with PBST buffer (PBS containing 0.05% Tween-20), and the wells were blocked for 2 hours with 400 microliters per well of 5% (w / v) skim milk powder in PBST. The blocking buffer was then discarded, and the wells of the plate were washed with PBST. The blocked cell lysates were then transferred to the MaxiSorp plate and allowed to bind to the blocked, immobilized antigen for 2 hours at room temperature with gentle shaking. After this, the lysates were discarded, and the wells of the MaxiSorp plate were washed four times with PBST. One hundred microliters of a 1 / 4000 dilution of anti-FLAG M2 HRP-conjugated antibody (Sigma) in PBST containing 2.5% (w / v) skim milk was then added to each well and allowed to bind for 1 hour, after which it was discarded and the plate was washed four times with PBST.Next, 100 microliters of ELISA POD Substrate TMB Kit (HYPER) detection reagent (Nacalai Tesque, Japan) was added per well, and the color reaction was stopped by adding 100 microliters of 1 M phosphoric acid. The absorbance of each well was read at a wavelength of 450 nm (Figures 12A and 12B). Clones that produced positive binding signals were identified and expanded from inocula taken from individual wells of the glycerol stock storage plate described above. Cultures were grown overnight with shaking at 37°C in 2 mL of TB medium containing 1% glucose and 50 micrograms / mL kanamycin. Plasmids were isolated from these cultures using the FastGene Plasmid Mini Kit (Nippon Genetics, Japan), and sequencing of the DNA region encoding the scaffold of the present invention was performed by Eurofins Genomics (Japan).

[0206] Example 6 Small-scale protein purification and characterization of target-binding scaffolds E. coli glycerol stocks of sequence-confirmed target-binding clones from storage plates (described in Example 5) were used to grow 50 mL cultures of 2xYT medium containing 50 micrograms / mL kanamycin and 0.1% glucose at 37°C with vigorous shaking until the OD 600The culture was inoculated until a pH of 0.5 was reached. The culture was then chilled on ice, IPTG was added to 0.5 mM, and the culture was grown overnight at 27°C with vigorous shaking. The culture was centrifuged at 3,000 x g for 10 minutes at 4°C, and the cell pellet was resuspended in 27 mL of ice-cold PBS (pH 7.4) containing 300 mM NaCl. 3 mL of 10x bugbuster reagent (EMD Millipore) was then added, and the cells were lysed on ice for 30 minutes. The cell lysate was then centrifuged at 12,000 x g for 30 minutes at 4°C, and the supernatant containing the cell lysate was collected. These were then bound to a 1 mL bed volume of pre-equilibrated Talon Cell-thru resin (Clontech), and purification was continued according to the manufacturer's instructions, followed by elution in a volume of 5 mL. Ten microliter aliquots were run on NuPAGE 4-12% SDS-PAGE gels (Invitrogen) and the purified protein was visualized by staining with Coomassie blue stain (Figures 13-15). The eluted protein was buffer exchanged into PBS (pH 7.4) by repeated centrifugation through an Amicon Ultra-4 10,000 MWCO column (Millipore) according to the manufacturer's instructions, and the protein was collected in a volume of approximately 1 mL. Protein concentrations were calculated based on the measured absorbance at 280 nm compared to the extinction coefficient predicted from the amino acid sequence deduced from DNA sequence data.

[0207] The thermal stability of the scaffold protein of the present invention was determined by DSF measurement using SYPRO Orange dye (Merck) at a scan rate of 0.5°C / min with the protein at 500 micrograms / mL in PBS buffer (pH 7.4) (Figures 16A-16E). The melting temperature of the protein was determined from the temperature at the maximum of the first derivative curve of the fluorescence intensity.

[0208] The binding affinity of individual scaffolds of the present invention was estimated by ELISA. Antigens were dissolved in PBS to 1 microgram / mL, and 100 microliters were coated onto the surface of each well of a 96-well MaxiSorp plate (Nunc) overnight at 4°C. The following day, the wells of the MaxiSorp plate were washed with PBST buffer (PBS containing 0.05% Tween-20), and the wells were blocked with 400 microliters per well of 5% (w / v) skim milk powder in PBST for 2 hours. The blocking buffer was then discarded, and the wells of the plate were washed with PBST. Purified target-binding scaffold proteins of the present invention were diluted in a 96-well plate using a three-fold serial dilution method at various concentrations ranging from 10.8 micromolar to 20 pM in PBS containing 2.5% (w / v) skim milk. The diluted target-binding scaffold proteins were then transferred to the antigen-coated MaxiSorp plate and allowed to bind to the blocked, immobilized antigen for 2 hours with gentle shaking at room temperature. After this, the diluted target-binding scaffold protein solution was discarded, and the wells of the MaxiSorp plate were washed four times with PBST. Next, 100 microliters of a 1 / 4000 dilution of anti-FLAG M2 HRP conjugate antibody (Sigma) in PBST containing 2.5% (w / v) skim milk was added to each well and allowed to bind for 1 hour. This was then discarded, and the plate was washed four times with PBST. Next, 100 microliters of ELISA POD Substrate TMB Kit (HYPER) detection reagent (Nacalai Tesque, Japan) was added per well, and the color reaction was stopped by adding 100 microliters of 1 M phosphoric acid. The absorbance of each well was read at a wavelength of 450 nm. EC of binding was calculated. 50 was calculated from a four-parameter logistic plot of the measured absorbance values ​​(FIGS. 17A-17B).

[0209] Example 7 Assessing the randomization potential of proteins with sequence homology to the scaffolds of the present invention To determine whether the randomization scheme of the present invention is broadly applicable to proteins that share sequence identity with the scaffolds of the present invention, a polypeptide sequence homology search was performed to identify homology candidates. c The polypeptide sequence of the domain (SEQ ID NO: 1) was used to search the NCBI (National Center for Biotechnology Information) protein sequence database for homologous polypeptides using the blastp algorithm. c We identified several proteins that show homology to the CheB domain. One of them, the domain of the chemotaxis protein CheY from Fervidobacterium pennivorans (GenBank ID: ANE42371.1, amino acid residues 147-337) (SEQ ID NO: 48), is similar to CheB. c The Fervidobacterium sp. domain protein showed 78% homology with the nucleotide sequence of SEQ ID NO: 1 (FIG. 18), and individual loop-grafted variants of the Fervidobacterium sp. domain protein were predicted to show lower sequence homology with SEQ ID NO: 1. Therefore, the Fervidobacterium sp. protein domain was selected as a candidate for test loop grafting.

[0210] Using information obtained from the sequence alignment, a test loop graft similar to the one used to test the randomized design of the scaffold of the present invention was designed and incorporated into the Fervidobacterium sp. protein along with the Cys159Ser substitution in SEQ ID NO:48 to generate a cysteine-free test loop graft construct (SEQ ID NO:49) (Figures 19 and 20). The polypeptide sequence of the test loop-grafted Fervidobacterium sp. protein construct (SEQ ID NO:50), flanked by EcoRI and AscI sites, was encoded by synthetic DNA (SEQ ID NO:51) obtained from Eurofins Genomics (Japan), cloned into a modified pQE-80L vector (QIAGEN) containing the corresponding EcoRI and AscI cloning sites, and transformed into E. coli XL1-Blue (Agilent). Glycerol stocks of sequence-verified clones were used to inoculate 50 mL cultures of 2xYT medium containing 50 micrograms / mL kanamycin and 0.1% glucose, and grown at OD with vigorous shaking at 37°C. 600The culture was grown until its pH reached 0.5. The culture was chilled on ice, IPTG was added to 0.5 mM, and the culture was grown overnight at 27°C with vigorous shaking. The culture was centrifuged at 3000 x g for 10 minutes at 4°C, and the cell pellet was resuspended in 27 mL of ice-cold PBS (pH 7.4) containing 300 mM NaCl. 3 mL of 10x bugbuster reagent (EMD Millipore) was then added, and the cells were lysed on ice for 30 minutes. The cell lysate was then centrifuged at 12,000 x g for 30 minutes at 4°C, and the supernatant containing the cell lysate was collected. This was then bound to a 1 mL bed volume of pre-equilibrated Talon Cell-thru resin (Clontech), and purification was continued according to the manufacturer's instructions, followed by elution in a 5 mL volume. Ten microliter aliquots were run on NuPAGE 4-12% SDS-PAGE gels (Invitrogen), and the purified protein was visualized by staining with Coomassie blue stain (Figure 15). The eluted protein was buffer exchanged with PBS (pH 7.4) by repeated centrifugation through an Amicon Ultra-4 10,000 MWCO column (Millipore) according to the manufacturer's instructions, and the protein was collected in an approximately 1 mL volume. Protein concentration was calculated based on the measured absorbance at 280 nm compared with the extinction coefficient predicted from the amino acid sequence deduced from DNA sequence data. The thermostability of the test loop-grafted Fervidobacterium sp. protein was determined by DSF measurements using SYPRO Orange dye (Merck) at a scan rate of 0.5°C / min, using the protein at 500 micrograms / mL in PBS buffer (pH 7.4) (Figure 21). The melting temperature of the protein (78°C) was determined from the temperature at the maximum of the first derivative curve of the fluorescence intensity.

[0211] While the foregoing invention has been described in some detail for purposes of clarity and understanding, it will be apparent to those skilled in the art from a perusal of this disclosure that various changes in form and detail may be made therein without departing from the true scope of the invention.

[0212] References Binz H., Stumpp M., Forrer P., Amstutz P., Pluckthun A. (2003). Designing repeat proteins: Well-expressed, soluble and stable proteins from combinatorial libraries of consensus ankyrin repeat proteins. Journal of Molecular Biology 332, 489-503. Cho K., Crane B., Park S. (2011). An insight into the interaction mode between CheB and chemoreceptor from two crystal structures of CheB methylesterase catalytic domain. Biochemical and Biophysical Research Communications 411, 69-75. Du B., Han H., Wang Z., Kuang L., Wang L., Yu L., Wu M., Zhou Z., Qian M. (2010). Targeted drug delivery to hepatocarcinoma in vivo by phage-displayed specific binding peptide. Molecular Cancer Research 8, 135-144. Dudgeon K., Rouet R., Christ D. (2013). Rapid prediction of expression and refolding yields using phage display. Protein Engineering, Design and Selection 26, 671-674. Gilbreth R., Koide S. (2012). Structural insights for engineering binding proteins based on non-antibody scaffolds. Current Opinion in Structural Biology 22, 413-420. Honegger A., Malebranche A., Rothlisberger D., Pluckthun A. (2009). The influence of the framework core residues on the biophysical properties of immunoglobulin heavy chain variable domains. Protein Engineering, Design & Selection 22, 121-134. Jensen K., Andreatta M., Marcatili P., Buus S., Greenbaum J., Yan Z., Sette A., Peters B., Nielsen M. (2018). Improved methods for predicting peptide binding affinity to MHC class II molecules. Immunology 154, 394-406. Miller B., Demarest S., Lugovskoy A., Huang F., Wu X., Snyder W., Croner L., Wang N., Amatucci A., Michaelson J., Glaser S. (2010). Stability engineering of scFvs for the development of bispecific and multivalent antibodies. Protein Engineering Design and Selection 23, 549-557. Nagi A., Regan L. (1997). An inverse correlation between loop length and stability in a four-helix bundle protein. Folding and Design 2: 67-75 Regan L. (1999). Protein redesign. Current Opinion in Structural Biology 9:494-499. Schilling J., Schoppe J., Pluckthun A. (2014). From DARPins to LoopDARPins: novel LoopDARPin design allows the selection of low picomolar binders in a single round of ribosome display. Journal of Molecular Biology 426, 691-721. Schmidt A., Kochanowski K., Vedelaar S., Ahrne E., Volkmer B., Callipo L., Knoops K., Bauer M., Aebersold R., Heinemann M. (2015). The quantitative and condition-dependent Escherichia coli proteome. Nature Biotechnology 34, 104-110. Vogt M., Skerra A. (2004). Construction of an artificial receptor protein (“anticalin”) based on the human apolipoprotein D. Chembiochem 5:191-199. Willuda J., Honegger A., Waibel R., Schubiger A., Stahel R., Uwe Z., Pluckthun A. (1999). High thermal stability is essential for tumor targeting of antibody fragments: Engineering of a humanized anti-epithelial glycoprotein-2 (epithelial cell adhesion molecule) single-chain Fv fragment. Cancer Research 59, 5758-5767. Xu L., Kohli, N., Rennard R., Yang J., Razlog M., Zhang K., Baum J., Johnson B., Tang J., Schoeberl B., Fitzgerald J., Nielsen U., Lugovskoy A. (2013). Rapid optimization and prototyping for therapeutic antibody-like molecules. mAbs 5, 237-254. Zhao N., Schmitt M., Fisk J. (2016). Phage display selection of tight specific binding variants from a hyperthermostable Sso7d scaffold protein library. The FEBS Journal 283, 1351-1367.

Claims

1. (i) Residue: (a) 16-18 (inclusive); (b) 29 to 37 (inclusive); (c) 43 to 47 (inclusive); (d) 60-61 (inclusive); (e) 66-75 (inclusive); (f) 80 to 84 (inclusive); (g) 92-93 (inclusive); (h) 103 to 107 (inclusive); (i) 124 to 125 (inclusive); (j) 135 to 137 (inclusive); (k) 149-150 (inclusive); (l) 160 to 162 (inclusive); (m) 173 to 176 (inclusive); (n) 180 to 181 (inclusive) corresponding to the cognate loop region of SEQ ID NO: 1, consisting of: (ii) 14 loop regions linked to secondary structure elements corresponding to non-loop regions of SEQ ID NO: 1 Recombinant CheB containing c 1. A randomized polypeptide display library comprising a recombinant polypeptide scaffold comprising a domain, At least one of the loop regions is a non-naturally occurring variant of the cognate loop region of SEQ ID NO: 1; and The recombinant CheB c The recombinant CheB domain has at least 90% amino acid sequence identity to SEQ ID NO: 1 outside the non-naturally occurring variant loop region. c A randomized polypeptide display library comprising a recombinant polypeptide scaffold comprising a domain.

2. The scaffolding comprises: (i) four framework regions designated FR1, FR2, FR3, and FR4, having at least 90% amino acid sequence identity to the amino acids of SEQ ID NO: 40 for FR1, SEQ ID NO: 41 for FR2, SEQ ID NO: 42 for FR3, and SEQ ID NO: 43 for FR4; (ii) four framework regions connected by three loop regions Including; 2. The randomized polypeptide display library of claim 1, wherein at least one loop region is a non-naturally occurring variant of the cognate loop region of SEQ ID NO:

1.

3. 3. The randomized polypeptide display library of claim 1, wherein at least two loop regions are non-naturally occurring variants of the cognate loop regions of SEQ ID NO: 1, and the non-naturally occurring variant loop regions comprise between 2 and 20 variant amino acid positions.

4. The scaffolds each independently have the general formula: G 1 X a1 G 2 X a2 G 3 (In the formula, G 1 , G 2 and G 3 independently represent 1 to 3 glycines; X a1 and X a2 independently represent any amino acid of 1 to 10 residues) or The scaffold has three loop regions designated L1, L2 and L3; L1 is represented by SEQ ID NO:44 or SEQ ID NO:45, L2 is represented by SEQ ID NO:44 or SEQ ID NO:46, and L3 is represented by SEQ ID NO:47; for the L1, L2, and L3 loop regions, each Xaa independently represents any amino acid; 3. The randomized polypeptide display library of claim 2, wherein L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4 to form a contiguous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4.

5. L1 is represented by SEQ ID NO:44, L2 is represented by SEQ ID NO:44, and L3 is represented by SEQ ID NO:47; for the L1, L2, and L3 loop regions, each Xaa independently represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, or tyrosine; or L1 is represented by SEQ ID NO:45, L2 is represented by SEQ ID NO:46, and L3 is represented by SEQ ID NO:47; For the L1 and L2 loop regions, each Xaa independently represents any amino acid; 5. The randomized polypeptide display library of claim 4, wherein for the L3 loop region, each Xaa independently represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, or tyrosine.

6. The randomized polypeptide display library of claim 1 , wherein the scaffold is displayed on the surface of a ribosome, a bacteriophage, a virus, a bacterium, a yeast, or a mammalian cell.

7. Recombinant CheB c 1. A randomized polypeptide display library comprising a recombinant polypeptide scaffold comprising a domain, CheB c The domain has 14 loop regions and 15 non-loop regions, the loop regions corresponding to segments including amino acids 16-18, 29-37, 43-47, 60-61, 66-75, 80-84, 92-93, 103-107, 124-125, 135-137, 149-150, 160-162, 173-176, and 180-181 of SEQ ID NO:1; the non-loop region corresponds to a segment comprising amino acids 1-15, 19-28, 38-42, 48-59, 62-65, 76-79, 85-91, 94-102, 108-123, 126-134, 138-148, 151-159, 163-172, 177-179, and 182-193 of SEQ ID NO:1; all 15 non-loop regions are connected, in naturally occurring order, by all 14 loop regions, thereby forming a single linear polypeptide comprising an engineered two-turn α / β sandwich fold supersecondary structure region; at least one of the loop regions is a randomized loop region that differs from a cognate loop region sequence by at least one amino acid, thereby defining a target ligand binding region; The recombinant CheB c A randomized polypeptide display library comprising a recombinant polypeptide scaffold, wherein the domains have at least 90% identity to SEQ ID NO:1 outside the randomized loop regions.

8. The scaffolding comprises: (i) four framework regions designated FR1, FR2, FR3, and FR4, having at least 90% amino acid sequence identity to the amino acids of SEQ ID NO: 40 for FR1, SEQ ID NO: 41 for FR2, SEQ ID NO: 42 for FR3, and SEQ ID NO: 43 for FR4; (ii) four framework regions connected by three loop regions Including; 8. The randomized polypeptide display library of claim 7, wherein at least one loop region is a non-naturally occurring variant of the cognate loop region of SEQ ID NO:

1.

9. 9. The randomized polypeptide display library of claim 7 or 8, wherein at least two loop regions are non-naturally occurring variants of the cognate loop regions of SEQ ID NO: 1, and the non-naturally occurring variant loop regions comprise from 2 to 20 variant amino acid positions.

10. The scaffolds each independently have the general formula: G 1 X a1 G 2 X a2 G 3 (In the formula, G 1 , G 2 and G 3 independently represent 1 to 3 glycines; X a1 and X a2 independently represent 1 to 10 residues of any amino acid) or The scaffold has three loop regions designated L1, L2 and L3; L1 is represented by SEQ ID NO:44 or SEQ ID NO:45, L2 is represented by SEQ ID NO:44 or SEQ ID NO:46, and L3 is represented by SEQ ID NO:47; for the L1, L2, and L3 loop regions, each Xaa independently represents any amino acid; 9. The randomized polypeptide display library of claim 8, wherein L1 is linked between FR1 and FR2, L2 is linked between FR2 and FR3, and L3 is linked between FR3 and FR4 to form a contiguous polypeptide comprising the arrangement FR1-L1-FR2-L2-FR3-L3-FR4.

11. L1 is represented by SEQ ID NO:44, L2 is represented by SEQ ID NO:44, and L3 is represented by SEQ ID NO:47; for the L1, L2, and L3 loop regions, each Xaa independently represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, or tyrosine; or L1 is represented by SEQ ID NO:45, L2 is represented by SEQ ID NO:46, and L3 is represented by SEQ ID NO:47; For the L1 and L2 loop regions, each Xaa independently represents any amino acid; 11. The randomized polypeptide display library of claim 10, wherein for the L3 loop region, each Xaa independently represents serine, aspartic acid, arginine, alanine, leucine, threonine, asparagine, tryptophan, glycine, glutamic acid, valine, or tyrosine.

12. 12. The randomized polypeptide display library of claim 7, wherein the scaffold is displayed on the surface of a ribosome, a bacteriophage, a virus, a bacterium, a yeast, or a mammalian cell.

13. At least 10 6 The randomized polypeptide display library of claim 1 , having a sequence diversity of

14. A collection of isolated nucleic acid molecules encoding the randomized polypeptide display library of any one of claims 1 to 13.

15. A collection of isolated nucleic acid molecules encoding the randomized polypeptide display library of any one of claims 1 to 13 operably linked to an expression vector.

16. 16. A cell genetically engineered to express the collection of isolated nucleic acid molecules of claim 14 or 15.

17. 14. A method for obtaining a polypeptide scaffold that binds to a target, the method comprising: (a) contacting a target ligand with the randomized polypeptide display library of any one of claims 1 to 13 under conditions that allow the formation of a scaffold:target ligand complex; and (b) obtaining from the complex the scaffold that binds to the target ligand.

18. 18. The method of Claim 17, further comprising: (a) isolating a nucleic acid molecule encoding the scaffold that binds to the target ligand; (b) operably linking the nucleic acid to an expression vector; and (c) expressing the nucleic acid operably linked to the expression vector in a cell.

Images