JP2007520423A - Protein design automation to design modified immunogenic protein libraries - Google Patents

Abstract

  The present invention relates to the use of various computer processing methods to modulate the immunogenicity of a protein by identifying and modifying amino acid sequences that are likely to elicit an immune response in a host organism. In particular, proteins are screened for MHC binding sequences, T cell epitopes and B cell epitopes.

Description

  This application claims the priority date benefit of USS 09 / 903,378, filed July 10, 2001.

Field of Invention   The present invention relates to the use of various computer processing methods to modulate the immunogenicity of a protein by identifying and modifying amino acid sequences that are likely to elicit an immune response in a host organism. In particular, proteins are screened for MHC binding motifs, T cell receptors and B cell receptor binding sequences.

Background of the Invention   The distinction between foreign and “self” is a major point in immune surveillance. The identification of foreign pathogen-derived proteins such as viruses and bacteria is a critical step in acquired immunity. A similar recognition process occurs in transplant organ rejection and autoimmune diseases, and can also occur in humans when exogenous proteins or other macromolecules are used repeatedly or continuously systemically.

  There are two main weapons of acquired immunity: humoral immunity and cellular immunity. Immunoglobulins are the most important humoral immune response. As a cell surface receptor on B lymphocytes, immunoglobulins are responsible for eliciting a wide variety of cellular responses such as activation, differentiation and programmed cell death. When secreted as antibodies, immunoglobulins bind to foreign antigens and neutralize them directly or arm and collect effector systems such as antibody-dependent cell lysis by complement or monocytic phagocytes The necessary steps are started (Fundamental Immunology, 4th edition, WE Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 3, pp 37-74).

  T cells are responsible for cellular immunity. T cells directly kill target cells, provide assistance to such killer, activate cells of other immune systems (ie macrophages), help B cells elicit an antibody response It is known to down-regulate the activity of various immune system cells and to secrete cytokines, chemokines, and other regulatory factors. These activities are often mediated by different types of T cytokines, chemokines, and other modulators. These activities are often mediated by different types of T cells such as α: β T cells, type 1 and type 2 helper cells. T cell activation occurs when a T cell recognizes a particular antigen via a receptor on its surface (ie, a T cell receptor or TCR). α: β T cells (ie CD8 + and CD4 + T cells) are only in association with one of the molecules encoded within the major histocompatibility complex (MHC) and when it is a suitable allelic variant Only recognize the antigen. This phenomenon is called MHC restriction (Fundamental Immunology, 4th edition, W. E. Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 11, pp 367-409).

  Major histocompatibility complex (MHC) molecules bind to foreign protein-derived polypeptide fragments (antigens) and present these peptides to receptors on the surface of T cells leading to an immune response, leading to an immune response. Play a central role. MHC molecules achieve their primary role in the immune response by fulfilling two distinct molecular functions: peptide binding and interaction with T cells, usually via the α: β T cell receptor (TCR). It is. Peptide binding by MHC I or MHC II molecules is determined by the cell expressing the MHC molecule (antigen-presenting cell, APC), its protein (MCH I) or the protein (MCH II) taken from the nearby extracellular environment. ) Is a selective event that makes it possible to become any sample (Fundamental Immunology, 4th edition, WE Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 8, pp 263-285).

  The interaction between the TCR on one cell and the complementary peptide-MHC complex on another cell causes a cascade of intracellular signals that depend on the identity of the T cell and the antigen presenting cell. Finally, TCR-peptide-MHC recognition regulates immune responses, including graft and tumor rejection, antiviral cytolysis, and collection and control of other immune cells such as antibody-producing B cells (Madden, DR , (1995) Annu. Rev. Immunol., 13: 587-622).

  MHC molecules are highly polymorphic and show allelic variation in various human populations (Buus, supra). Hundreds of MHC class I and II alleles are known, each displaying a different binding affinity for a particular antigenic peptide sequence. This structural bias for allele-dependent peptide selectivity is centered on differences in amino acid residues within the MHC peptide binding pocket (Buus, supra). X-ray crystal structures of MHC class I and II molecules bound to specific antigenic peptides show that the N and C-terminal (ie anchor position) peptide residues are in close physical contact with the MHC class I binding pocket, while It has been revealed that peptides that bind to class II are further extended by additional peptide residues that make contact with the MHC class II pocket (Buus, supra).

  Detailed sequence analysis of peptides extracted from MHC molecules revealed several allele-specific amino acid selectivities (Buus, supra). A database of thousands of peptide sequences known to bind to MHC molecules has been compiled (Rammensee, H., et al. (1999) Immunogenetics, 50: 213-219) and possible to analyze full-length protein sequences Several techniques have been developed to predict the presence of sex antigenic sequences (Hiemstra, HS et al. (2000) Curr. Op. Immunol., 12: 80-84; Malios, RR, (1999) Bioinformatics, 15: 432-439; Sturniolo, T., et al. (1999) Nature Biotechnology, 17: 555-561; Brusic, V., et al., (1998) Bioinformatics, 14: 121-130; Mallios, RR, (1998) J. Comp. Biol., 5: 703-711; Savoie, CJ et al. (1999) Pac Symp Biocomput, 182-9; Altuvia, Y., et al. (1997) Human Immunology, 58 : 1-11; Shastri, N. (1996) Curr. Op. Immunol., 8: 271-277; Hammer, J. (1995) Curr. Op. Immunol., 7: 263-269; Meister, GE, et al. (1995) Vaccine, 13: 581-591; Udaka, K., et al. (1995) J. Exp. Med., 181: 20972108; Hammer, J. et al. (1994) Behring. Inst. Mitt 94: 124-132; Hammer, J., et al. (1994) J. Exp. Med., 180: 2353-2358; and Rudenshky, A. Y., et al. (1991) Nature, 353: 622-627). The overall peptide binding affinity is sequence and MHC allele specific, but the contribution of each peptide residue is independent of the identity of the neighboring residues and can be summed individually (Altuvia, et al., The above). The presence of anchor residues and the length of MHC class I binding peptides lead to a better predictive model for MHC class I molecules than MHC class II molecules (Abrams and Schlom, (2000) Curr. Op. Immunol., 12: 85-91).

  It is not clear which residues of the antigenic peptide are bound by TCR, but side chain substitution experiments have mapped the outline of TCR binding sites on a number of peptide-MHC complexes. Typically, different TCRs have been found to contact different but overlapping subsets of MHC and peptide side chains. The TCR “footprint” is located in the center of the bound peptide and includes the MHC side chain at the top of both α helices that form the peptide binding groove. Despite the majority of the peptide surface being buried, the bound peptides clearly contribute significantly to TCR recognition. More recent results suggest that each amino acid in the peptide sequence independently contributes to the affinity of the MHC-peptide-TCR complex (Hemmer, B., et al., (1998), J Immunol., 160: 3631-3636).

  An important component of humoral immunity is the broad repertoire of antibodies (ie, immunoglobulins) produced by B lymphocytes. Antigen contact with specific B cells causes the transmembrane signaling function of the B cell antigen receptor (BCR). This in turn induces early events in B cell activation, including increased MHC class II molecule expression and formation of antibody secreting cells.

  Reduced polypeptide immunogenicity is due to rational site-directed mutagenesis (Meyer, et al., (2001) Protein Science 10: 491-503), exhaustive site-directed mutagenesis. Introduction (Laroche, et al., (2000) Blood, 96: 1425-1432; WO00 / 34317; WO98 / 52976) and direct chemical coupling of polyethylene glycol derivatives (Tsutsumi, et al., (2000) Proc. Natl Acad. Sci. USA, 97: 8548-8553). However, these methods are extremely time consuming, especially when multiple mutations are considered simultaneously. Although rational selection of surface residues can lead to reduced immunogenicity, substitution of some residues can destabilize and cause poor folding. In addition, removal of charged residues exposed to the solvent can be energetically inconvenient.

  One way to overcome these problems is to use computerized methods that are either more or less immunogenic compared to the target protein, but have the structural properties to ensure proper folding and activity. It is to design the array that holds.

  Accordingly, it is an object of the present invention to use a computerized method to screen for potential MHC, TCR, or BCR binding peptides. A wide variety of methods are known for the generation and evaluation of sequences. These include sequence profiling (Bowie and Eisenberg, Science 253 (5016): 164-70, (1991)), rotamer library selection (Dahiyat and Mayo, Protein Sci 5 (5): 895-903 (1996) ); Dahiyat and Mayo, Science 278 (5335): 82-7 (1997); Desjarlais and Handel, Protein Science 4: 2006-2018 (1995); Harbury et al, PNAS USA 92 (18): 8408-8412 (1995) ); Kono et al., Proteins: Structure, Function and Genetics 19: 244-255 (1994); Hellinga and Richards, PNAS USA 91: 5803-5807 (1994)); and residue pair potential (Jones, Protein Science 3 : 567-574, (1994)), but is not limited thereto.

  In particular, U.S.S.N. 60 / 061,097, 60 / 043,464, 60 / 054,678, 09 / 127,926, and PCT US98 / 07254 include a number of methods for assessing sequence stability. A method called “Protein Design Automation” or PDA® that utilizes a scoring function is described.

  Furthermore, it is an object of the present invention to provide a computerized method for screening a sequence library in order to select a small library of protein sequences that can be created and evaluated for immunogenic modification.

Summary of the Invention   In accordance with the objects outlined above, the present invention provides methods for producing polypeptides that exhibit enhanced immunogenicity. The method includes inputting a backbone structure of a target protein along with variable residue positions into a computer, applying a set of primary mutated amino acid sequences by applying at least one protein design algorithm, and computing the computer Analyzing the set of primary variant amino acid sequences by computer processing by applying an immunogenic filter of treatment. A candidate protein is then created and tested to determine whether the candidate protein has been enhanced in immunogenicity compared to the target protein. This same method can be used to produce polypeptides that exhibit reduced immunogenicity.

  In a further aspect, the present invention provides a method for producing a polypeptide that exhibits enhanced immunogenicity. The method includes inputting a backbone structure of a target protein with variable residue positions into a computer, applying at least one computational immunogenic filter to generate a set of primary mutated amino acid sequences, at least one Computational analysis of the set of primary mutant amino acid sequences using a protein design algorithm, and identifying at least one mutant protein with enhanced immunogenicity. This same method can be used to produce polypeptides that exhibit reduced immunogenicity.

  In a further aspect, the present invention provides a method for producing a polypeptide that exhibits enhanced immunogenicity. The method applies at least one protein design algorithm comprising inputting a backbone structure of a target protein with variable residue positions into a computer, at least one scoring function including at least one computational immunogenic filter. Thereby generating a set of primary mutated amino acid sequences by computer processing and identifying at least one variant protein having enhanced immunogenicity. This same method can be used to produce polypeptides that exhibit reduced immunogenicity.

  In a further aspect, the present invention provides a method for producing a polypeptide that exhibits enhanced immunogenicity. The method includes inputting a backbone structure of a target protein with variable residue positions into a computer, applying at least one computerized protein design algorithm and at least one computerized immunogenic filter in any order. And identifying at least one mutant protein having enhanced immunogenicity. This same method can be used to produce polypeptides that exhibit reduced immunogenicity.

  In a further aspect, the present invention provides a method of eliciting an enhanced immune response in a patient. The method includes inputting a backbone structure of a target protein with variable residue positions into a computer, applying at least one computerized protein design algorithm and at least one computerized immunogenic filter in any order. Identifying at least one mutant protein having enhanced immunogenicity, and administering the mutant protein to a patient.

  The computerized design algorithm may be applied prior to or after the application of the computerized immunogenic filter. Alternatively, the computerized protein design algorithm includes a computerized filter as a scoring function.

  The step of generating by computer processing may comprise applying a computerized immunogenic filter comprising a scoring function for the MHC class I motif, MHC class II motif, B cell epitope or T cell epitope. Other computational steps include the use of Dead-End-Elimination (DEE) computer calculations, Monte Carlo searches, or genetic algorithms. Additional scoring functions include van der Waals potential scoring functions, hydrogen bond potential scoring functions, atomic solvation scoring functions, secondary structure trend scoring functions, and electrostatic scoring functions.

  In a further aspect, the polypeptide may comprise one or more immunogenic sequences. The immunogenic sequences can be identical or different. The immunogenic sequence can be selected from the group consisting of an MHC class I motif, an MHC class II motif, a B cell epitope and a T cell epitope.

  In a further aspect, the target protein is selected from the group comprising Zn-alpha 2-glycoprotein, human serum albumin, immunoglobulin G and other non-immunogenic proteins.

Brief Description of Drawings   FIG. 1 depicts full-length gene synthesis and possible total mutation introduction by PCR. Overlapping oligonucleotides corresponding to the full-length gene (black bar, step 1) are synthesized, heated and annealed. Addition of Pfu DNA polymerase to the annealed oligonucleotide leads to 5 ′ → 3 ′ synthesis of DNA (step 2), producing longer DNA fragments (step 3). Repeated heating and annealing cycles (step 4) result in the production of longer DNA containing several full length molecules. These can be selected by a second round of PCR using primers (arrows) corresponding to the ends of the full-length gene (step 5).

  FIG. 2 shows a preferred scheme for synthesizing the libraries of the present invention. Any starting gene can be used, such as a wild type gene, or a gene for a global minimum gene. Oligonucleotides containing different amino acids at different mutation sites can be used in PCR using standard primers. This generally requires less oligonucleotide and results in fewer errors.

  FIG. 3 shows the overlap extension method. The top row of FIG. 3 is a template DNA showing the location of the region to be mutated (black box) and the binding site of the associated primer (arrow). Primers R1 and R2 represent a pool of primers, each containing a different mutation; as described herein, this can be done using various ratios of primers as desired. The mutation site is adjacent to a region with sufficient homology to perform hybridization. In this example, three separate PCR reactions are performed in stage 1. The first reaction includes oligos F1 and R1 in addition to the template. The second reaction includes F2 and R2 in addition to the template, and the third reaction includes F3 and R3 in addition to the template. The reaction product is shown. Stage 2 uses the product of stage 1 tube 1 and stage 1 tube 2. After purification to remove the primers, they are added to a new PCR reaction along with F1 and R4. During the denaturation step of PCR, the overlapping region anneals and the second strand is synthesized. The product is then amplified with the outer primer. In stage 3, the purified stage 2 product is used in the third PCR reaction with the product of stage 1 tube 3 and primers F1 and R3. The final product corresponds to the full-length gene and contains the desired mutation.

  FIG. 4 shows the ligation of PCR reaction products to synthesize the library of the present invention. In this technique, the primer also contains an endonuclease restriction site (RE), either blunt ended, 5 ′ overhanged, or 3 ′ overhanged. We set up three separate PCR reactions in stage 1. The first reaction includes oligos F1 and R1 in addition to the template. The second reaction includes F2 and R2 in addition to the template, and the third reaction includes F3 and R3 in addition to the template. The reaction product is shown. In step 2, the product of step 1 is purified and then cleaved with an appropriate restriction endonuclease. The cleavage products from step 2 tube 1 and step 2 tube 2 are both ligated with DNA ligase (step 3). Step 4 then amplifies the product using primers F1 and R4. The entire process is repeated by cleaving the amplified products, ligating them to the cleavage products of stage 2 tube 3, and amplifying the final product with primers F1 and R3. If the two restriction enzyme sites (RET and RE2) are different, it is possible to ligate all three PCR products from step 1 in one reaction.

  FIG. 5 depicts blunt end ligation of PCR products. In this technique, primers such as F1 and R1 do not overlap but are adjacent. Again three separate PCR reactions are performed. The products of tube 1 and tube 2 are ligated and amplified with outer primers F1 and R4. This product is then ligated with the product of tube 1 of stage 1. The final product is then amplified with primers F1 and R3.

DETAILED DESCRIPTION OF THE INVENTION The present invention, for selecting a small library of protein sequences with altered immunogenicity (10 ¹³ can include the following members), the protein sequence library (10 ⁸⁰ or less or The method of using computerized screening (which may include the above members) is targeted. For example, if a protein with reduced immunogenicity is desired, to identify residues known to elicit an immune response and replace them with compensating residues that maintain the protein's natural folding and stability Computerized filters can be used, resulting in proteins that are non-immunogenic or less immunogenic than the starting protein.

  Alternatively, it may be desirable to design a protein with increased immunogenicity. In this case, computational filters can be applied to change residues and introduce antigen motifs to ensure proper folding and stability of the resulting protein.

  In general, this can be done in one of two general ways. In a first embodiment, computational processing is used to generate a list of mutant proteins with altered properties such as stability. A computerized filter is then applied to select variants that are more likely to have altered immunogenicity.

  Alternatively, a computational filter is first applied to generate a list of variants that are prone to altered immunogenicity, and then variants that are considered to be folded or stable are selected. Therefore, processing by computer processing is performed.

  In particular, computational filters are used to screen for peptide fragments or amino acid residues that may bind to MHC class I and class II molecules, T cells and B cells. For example, it can be used to search a database of MHC ligands and peptide motifs and identify potential MHC class I or class II binding sequences (Rammensee, H., et al. (1999) Immunogenetics, 50: 213 -219). Computer processing methods are then used to structurally and chemically compensate for amino acid residues associated with binding to MHC molecules. For example, if a mutant protein that is less immunogenic than the target protein is desired, peptide sequences or amino acid residues that are predicted to elicit an immune response are identified and the residues that are predicted to be non-immunogenic are identified. Computer processing methods can be used to screen for sequences that are properly folded and stable from the resulting sequence and then substituted with a group.

  Rules have been clarified to determine appropriate substitution of antibody-binding surface residues (Meyer, DL, et al. (2001) Protein Science, 10: 491-503; Laroche, Y., (2000) Blood 96: 1425-1432; and Schwartz, HL, (1999) J. Mol. Biol., 287: 983-999). For example, aromatic surface residues are involved in antigen-antibody binding. Aromatic surface residues such as tyrosine can be replaced with small residues such as serine, alanine or glycine. Similarly, large patches of charged side chains can be replaced with small hydrophilic residues such as serine or alanine. Computer processing methods can then be applied to determine sequence changes that compensate for natural folding and maintenance of stability.

  There are situations where it is desirable to increase the immunogenicity of the target protein. For example, activation of T cell populations against specific epitopes is closely related to controlling or eliminating viral pathogens or tumorigenesis. In this case, computational methods can be used to introduce T cell epitopes anywhere within the target protein. In addition, T cell epitopes can be introduced into regions of the target protein that are less fixed and less structurally restricted, such as loop regions, using the computer processing methods described herein. In doing so, computer processing methods are used to change the residues adjacent to the epitope insertion to ensure energetic compatibility between the native protein and the grafted epitope.

  Accordingly, the present invention provides a method for modulating the immunogenicity of a target protein. “Modulation” as used herein means that the immune response to the target protein is altered. That is, if the target protein elicits an immune response in a given species, the amino acid sequence of the target protein is changed so that the immune response is reduced or enhanced. As used herein, “decrease” means that at least one immunological response is reduced compared to a wild-type protein. As used herein, “enhanced” means that at least one immunological response is increased compared to a wild-type protein. As will be appreciated by those skilled in the art, not all identified sequences that are capable of eliciting a response need be modified. For example, immune responses generally do not occur against autologous circulating proteins such as immunoglobulins and other serum proteins. Thus, at least 5% of the sequences capable of eliciting a response are modified. Preferably, at least 10% of the sequence is modified, more preferably at least 15% of the sequence is modified, more preferably at least 20% of the sequence is modified, even more preferably at least 30% of the sequence is modified. Preferably, at least 40% of the sequence is even more preferred, at least 50% of the sequence is more preferred, and 100% of the sequence is most preferred.

  It should be noted that altered immunogenicity is defined within a particular host organism. That is, in a preferred embodiment, the target protein (as defined below) is modified to show altered immunogenicity in humans. Alternative host organisms include rodents (rats, mice, hamsters, guinea pigs, etc.), primates, livestock animals (sheep, goats, pigs, cows, horses, etc.), and domestic animals (cats, dogs, rabbits). Such as, but not limited to.

  As used herein, “immunogenic” refers to the ability of a protein to elicit an immune response. The ability of a protein to elicit an immune response depends on the amino acid sequence or sequences within the protein. As outlined below, immunogenicity includes both humoral and cellular components of the immune response. Amino acid sequences capable of eliciting an immune response are referred to herein as “immunogenic sequences”. Preferably, the immunogenic sequence comprises an “MHC binding site (ie, MHC binding motif)”, “T cell epitope” and “B cell epitope” as outlined below.

  As defined herein, the definition of immunogenicity is broad enough to encompass the term “antigenic”. “Antigen” refers to the ability of a protein to elicit an antibody response by itself when recognized as a non-self molecule.

  The reaction elicited by a protein having an immunogenic sequence includes both components of the immune system: humoral immunity and cellular immunity. Thus, an “immune response” in the context of the present invention includes any component of a humoral or cellular immune response. In general, when a protein having an immunogenic sequence is administered to a human, the protein is under surveillance of both humoral and cellular weapons of the immune system. If a protein is recognized as foreign and the immune system is not yet tolerant to the immunogenic sequences within the protein, the immune system will react to the protein. In a humoral immune response, immature B cells displaying immunoglobulin (Ig) on the surface have an affinity that matches the individual immunoglobulin and allow the B cell epitope to be accessible to the Ig. Can be bound to one or more sequences (B cell epitopes) within the protein. The process of Ig binding to the protein provides a soluble form of native Ig that can stimulate B cells to divide and form a complex with the protein in the presence of a suitable cytokine. Removal can be promoted.

  An effective B cell response also includes a T cell response in parallel to provide the cytokines and other signals necessary to generate soluble antibodies. An effective T cell response requires the uptake of protein fragments by antigen presenting cells (APCs); APCs include B cells or other cells such as macrophages, dendritic cells and other monocytes It is. The APC then presents the protein complexed with MHC class II molecules on the cell surface. Such peptide-MHC II complexes can be recognized by helper T cells via the T cell receptor (TCR) and this assists B cells in differentiation into antibody producing cells. Leading to T cell stimulation and cytokine secretion. As can be seen from the above discussion, an effective primary immune response to an immunogenic protein generally requires a combination of B and T cell responses to B and T cell specific sequences or epitopes.

  Alternatively, if the immunogenic sequence is specific for MHC class I molecules, the MHC I antigen processing / presentation pathway is involved. MHC class I molecules collect fragments of infectious agent-derived proteins or “self” molecules and then display these fragments on the surface of the APC. The bound peptide is recognized by the TCR of cytotoxic T lymphocytes and is the primary antigenic determinant of the cellular immune response. Thus, the regulation of immunogenicity involves identifying peptides that stimulate T cell responses, ie T cell epitopes, and changing the sequence of those peptides so that the cellular response to the protein is reduced or enhanced. That is included. In addition, the modulation of immunogenicity includes the identification of peptides that stimulate B cell responses, ie “B cell epitopes” or “BCRs”, such that the humoral response to the protein is altered. Changing the sequence of the peptide is also included. As will be appreciated by those skilled in the art, since a single protein can contain both T and B cell epitopes, both modulations can modify both humoral and cellular weapons of the immune system.

  In a preferred embodiment, the target protein is modified to modify the MHC I response. MHC class I molecules collect protein fragments derived from infected viruses, intracellular parasites, or self-proteins that are normally expressed or deviated from control by tumorigenesis; To display. On the cell surface, the MHC I-peptide complex bound to the cells exposed on the APC is displayed to the T cells. The second feature of the MHC I molecule is the ability to interact with the TCR that allows APCs with specific MHC-peptide complexes to engage the appropriate TCR. This is the first step in the activation of the cellular program leading to the lysis of APC as a target and / or the secretion of lymphokines by T cells. Interaction with TCR depends on both peptides and MHC molecules. MHC class I molecules exhibit selective restriction on CD8 + cells. A further function of MHC class I molecules is to act as an element of signal transduction to natural killer cells (Fundamental Immunology, 4th edition, WE Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 8, pp 263- 285).

  In a preferred embodiment, the target protein is modified to modify the MHC II response. Utilizing the same molecular mechanism as MHC class I molecules, MHC class II molecules bind to peptides derived from the degradation of proteins ingested by APCs expressing MHC II and are recognized by specific T cells. Display them on the cell surface. The MHC II antigen presentation pathway is based on the initial association of the MHC II αβ heterodimer with a dual function molecule, ie, the invariant chain (Ii). Ii acts as a chaperone that directs the αβ heterodimer to the acidic protein processing site of the endosome where the αβ heterodimer meets the antigenic peptide. The process of loading the antigenic peptide onto the MHC II molecule leads to cell surface presentation of the MHC II peptide complex. T cells that recognize MHC II can then be induced to secrete lymphokines and proliferate. MHC class II molecules exhibit selective restriction on CD4 + cells (Fundamental Immunology, 4th edition, W. E. Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 8, pp 263-285).

  In a preferred embodiment, the target protein is modified to modify the TCR response. TCR occurs as two separate heterodimers, αβ or γδ, both of which are expressed with non-polymorphic CD3 polypeptides γ, δ, ε, ζ. CD3 polypeptides, particularly ζ and its variants, are very important for intracellular signaling. αβTCR heterodimer expressing cells are dominant in most lymphocyte compartments and are responsible for classical helper or cytotoxic T cell responses. In most cases, the αβTCR ligand is a peptide antigen bound to a class I or class II MHC molecule (Fundamental Immunology, 4th edition, WE Paul, ed., Lippincott-Raven Publishers, 1999, Chapter 10, pp 341-367). .

  In a preferred embodiment, the target protein is modified to modify the BCR response. Contact of the antigen with specific B cells causes the transmembrane signaling function of the B cell antigen receptor (BCR). BCR molecules are internalized rapidly after antigen binding, leading to antigen uptake and degradation in endosomes or lysosomes. In the case of protein antigens, antigen-derived peptides bind in the groove of class II MHC molecules. Upon binding, this complex is delivered to the cell surface where it acts as a stimulator for specific helper T cells. Antigen recognition by helper T cells induces helper T cells to form tight and long-lasting interactions with B cells and synthesize B cell proliferation and differentiation factors. B cells activated in this way proliferate and finally differentiate into antibody-secreting cells (also called plasma cells) (Fundamental Immunology, 4th edition, WE Paul, ed., Lippincott-Raven Publishers, 1999, Chapters). 6-7, pp 183-261).

  Accordingly, the present invention is directed to a method for modulating the immunogenicity of a target protein. As used herein, “target protein” means at least two amino acids covalently bound, and includes proteins, polypeptides, oligopeptides and peptides. The protein comprises naturally occurring amino acids and peptide bonds or synthetic peptidomimetic structures, ie “analogs” such as peptoids (Simon et al., Proc. Natl. Acad. Sci. USA 89 (20): 9367. -71 (1992)) and generally depends on the synthesis method. Therefore, in the present specification, “amino acid” or “peptide residue” means both naturally-occurring amino acids and synthetic amino acids. For example, homophenylalanine, citrulline and norleucine are considered amino acids for the present invention. “Amino acid” also includes imino acid residues such as proline and hydroxyproline. In addition, any amino acid representing a component of the mutant protein of the invention can be replaced with the same amino acid but with the opposite chirality. Thus, any amino acids that are naturally produced in the L configuration (these are also referred to as R or S, depending on the structure of the chemical), can be substituted with amino acids of the same chemical structure but of the opposite chirality. This is commonly referred to as a D amino acid, but is further called R or S depending on its composition and chemical configuration. In general, such derivatives have the property of greatly increasing stability, and therefore when administered by oral, intravenous, intramuscular, intraperitoneal, topical, rectal, intraocular or other routes. It is advantageous for the formation of compounds with a longer half-life in vivo.

  In a preferred embodiment, the amino acid is in the (S) or L configuration. If non-naturally occurring side chains are used, non-amino acid substituents may be used, for example to prevent or retard in vivo degradation. Proteins containing non-naturally occurring amino acids may be synthesized or may be made recombinantly; van Hest et al., FEBS Lett 428: (1-2) 68-70 May 22 1998 and Tang et al ., Abstr. Pap Am. Chem. S218: U138-U138 Part 2 August 22, 1999. Both are hereby incorporated by reference.

  Aromatic amino acids are D- or L-naphyl alanine, D- or L-phenylglycine, D- or L-2-thienylalanine, D- or L-1-, 2-, 3- or 4- Pyrenylalanine, D- or L-3-thienylalanine, D- or L- (2-pyridinyl) -alanine, D- or L- (3-pyridinyl) -alanine, D- or L- (2-pyrazinyl) -alanine , D- or L- (4-isopropyl) -phenylglycine, D- (trifluoromethyl) -phenylglycine, D- (trifluoromethyl) -phenylalanine, Dp-fluorophenylalanine, D- or Lp- Biphenylphenylalanine, D- or Lp-methoxybiphenylphenylalanine, D- or L-2-indole (Al ) Alanine, and D- or L-alkyl ainine, where alkyl is substituted or unsubstituted methyl, ethyl, propyl, hexyl, butyl, pentyl, isopropyl, iso-butyl, sec-isotyl, iso -Pentyl and C1-C20 non-acidic amino acids).

Acidic amino acids maintain a negative charge, such as (phosphono) alanine, glycine, leucine, isoleucine, threonine, or serine; or sulfated (eg, —SO ₃ H) threonine, serine, or tyrosine as non-limiting examples Non-carboxylic acid amino acids and derivatives or analogs thereof.

  Other substitutions may include unnatural hydroxylated amino acids that may be made by combining “alkyl” with any natural amino acid. As used herein, the term “alkyl” refers to 1 to 24 such as methyl, ethyl, n-propyl, isopropyl, n-butyl, isobutyl, t-butyl, octyl, decyl, tetradecyl, hexadecyl, eicosyl, tetracycyl and the like. Means a branched or unbranched saturated hydrocarbon group of carbon atoms. Alkyl includes heteroalkyl having nitrogen, oxygen and sulfur atoms. Preferred alkyl groups in the present invention contain 1 to 12 carbon atoms. The basic amino acid is any of the naturally occurring amino acids lysine, arginine, ornithine, citrulline, or (guanidino) -acetic acid, or other (guanidino) alkyl-acetic acid (where “alkyl” is as defined above). It can be substituted with an alkyl group at the position. Nitrile derivatives (eg containing a CN moiety instead of COOH) can also replace asparagine or glutamine, and methionine sulfoxide can replace methionine. Methods for preparing such peptide derivatives are well known to those skilled in the art.

  In addition, any amide bond in any mutant polypeptide can be replaced with a ketomethylene moiety. Such derivatives have the property of increasing stability against enzymatic degradation, and thus when administered by oral, intravenous, intramuscular, intraperitoneal, topical, rectal, intraocular or other routes. It is expected to be advantageous for the formation of compounds with longer half lives in vivo.

  Further amino acid modifications of the amino acids of the mutant polypeptides of the invention may include the following: reacting a cysteinyl residue with an alpha-haloacetate salt (and corresponding amine) such as 2-chloroacetic acid or chloroacetamide. To obtain a carboxymethyl or carboxyamidomethyl derivative. Cysteinyl residues include bromotrifluoroacetone, alpha-bromo-beta- (5-imidozoyl) propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, It can also be derivatized by reaction with compounds such as p-chloromercury benzoate, 2-chloromercury-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

  A histidyl residue can be derivatized by reaction with a compound such as diethyl procarbonate (eg, at pH 5.5-7.0, since this material is relatively specific for the histidyl side chain), and para-bromophenacyl (Bromophenacyl) bromide may also be used; for example, in that case, the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0.

  Ricinyl and amino terminal residues can be reacted with succinic acid or other carboxylic anhydrides. Induction with these reagents is expected to have the effect of reversing the charge of the ricinyl residue.

  Other reagents suitable for deriving alpha-amino containing residues include imide esters such as methylpicolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzene sulfonic acid; O-methylisourea; Transaminase-catalyzed reaction with compounds such as pentanedione; and glyoxylate. Arginyl residues may be modified according to known method steps by reaction with one or more conventional reagents, notably phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione and ninhydrin. Inducing arginine residues, the pKa of the guanidine functional group is high, so it is necessary to carry out the reaction under alkaline conditions. Furthermore, these reagents can be reacted with lysine groups as well as the epsilon-amino group of arginine. Specific modifications of tyrosyl residues themselves, such as to introduce spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane are well known.

  N-acetylimidizole and tetranitromethane can be used to form O-acetyl tyrosyl species and 3-nitro derivatives, respectively. Carboxyl side chain groups (aspartyl or glutamyl) include 1-cyclohexyl-3- (2-morpholinyl- (4-ethyl) carbodiimide or 1-ethyl-3- (4-azonia-4,4-dimethylpentyl) carbodiimide Can be selectively modified by reaction with carbodiimide (R′—N—C—N—R ′) In addition, aspartyl and glutamyl residues can be converted to asparaginyl and glutaminyl residues by reaction with ammonium ions. .

  Glutaminyl and asparaginyl residues are frequently deamidated to the corresponding glutamyl and aspartyl residues. Alternatively, these residues can be deamidated under mildly acidic conditions. Either form of these residues is within the scope of the present invention.

  The target protein can be any protein whose three-dimensional structure is known or can be generated, ie there are three-dimensional coordinates for each atom of the protein. In general, this can be measured using X-ray crystal techniques, NMR techniques, novel modeling, homology modeling, and the like. In general, when an X-ray structure is used, a structure with 2 構造 resolution or higher resolution is preferable, but it is not always necessary.

  The target proteins of the invention can be derived from prokaryotes and eukaryotes such as bacteria (including extreme bacteria such as archaea), fungi, insects, fish and mammals. Suitable mammals include, but are not limited to, rodents (rats, mice, hamsters, guinea pigs, etc.), primates, livestock animals (including sheep, goats, pigs, cows, horses, etc.). In the most preferred embodiment, it is human.

  That is, as used herein, “target protein” preferably refers to a protein for which a library of variants with altered immunogenicity is desired. As will be appreciated by those skilled in the art, any number of target proteins are useful in the present invention. Specifically, fragments and domains of known proteins, including functional domains such as enzyme domains, binding domains, and small fragments such as turns, loops, are included within the definition of “protein”. That is, a portion of the protein can be used as well. Furthermore, “protein” as used herein encompasses proteins, oligopeptides and peptides. In addition, protein variants, i.e. non-naturally occurring protein analog structures, can also be used.

  Suitable proteins include industrial, pharmaceutical and agricultural proteins, including ligands, cell surface receptors, antigens, antibodies, cytokines, hormones, transcription factors, signal modules, cytoskeletal proteins and enzymes. It is not limited. Suitable types of enzymes include, but are not limited to, hydrolases such as proteases, carbohydrases, lipases, isomerases such as racemases, epimerases, tautomers, or mutases, transferases, kinases, oxidoreductases and phosphatases. . Suitable enzymes are listed in the Swiss-plot enzyme database. Suitable protein backbones include, but are not limited to, those found in protein databases compiled and provided by Research Collaboratory for Structural Bioinformatics (RCSB, formerly the Brookhaven National Lab).

  Particularly preferred pharmaceutical target proteins include cytokines (IL-1ra (+ receptor complex), IL-1 (receptor alone), IL-1a, IL-1b (mutant and / or receptor complex) IL-2, IL-3, IL-4, IL-5, IL-6, IL-8, IL-10, IFN-β, INF-γ, IFN-α-2a; IFN-α- 2B, TNF-α; CD40 ligand (chk), human obesity protein leptin, granulocyte colony stimulating factor, bone morphogenic protein-7, ciliary neurotrophic factor, granulocyte macrophage colony stimulating factor, monocyte chemistry-inducing protein 1 , Macrophage migration inhibitory factor, human glycosylation inhibitor, human lantes, human macrophage inflammatory protein 1 beta, human growth hormone, leukemia inhibitory factor, human melanoma growth Stimulating activity, neutrophil activating peptide-2, Cc-chemokine Mcp-3, platelet factor M2, neutrophil activating peptide 2, eotaxin, stromal cell-derived factor-1, insulin, insulin-like growth factor I, insulin-like growth factor II, transforming growth factor B1, transforming growth factor B2, transforming growth factor B3, transforming growth factor A, vascular endothelial growth factor (VEGF), acidic fibroblast growth factor, basic Fibroblast growth factor, endothelial cell growth factor, nerve growth factor, brain-derived neurotrophic factor, ciliary neurotrophic factor, platelet-derived growth factor, human hepatocyte growth factor, glial cell-derived neurotrophic factor (and PDB1 / 55 cytokines in 12/99)); urokinase; erythropoietin Other extracellular signaling moieties including but not limited to hedgehog sonic, hedgehog desert, hedgehog indian, hCG; including but not limited to TPA and factor VIIa Coagulation factors; p53, p53 tetramerization domain, Zn finger (of which more than 12 have structure), homeodomain (of which 8 have structure), leucine zipper (of which 4 are A transcription factor; including but not limited to cFv; an antibody; a hemagglutinin tetramerization domain and a hiv Gp41 ectodomain (fusion domain) ) Viral proteins; SH2 domains (including but not limited to) 8 of which are known), an SH3 domain (of which 11 have structure), and an intracellular signal module including, but not limited to, a plexin homology domain; a human tissue factor cytokine of Gp130 Extracellular region of binding region, G-CSF receptor, erythropoietin receptor, fibroblast growth factor receptor, TNF receptor, IL-1 receptor, IL-1 receptor / IL1ra complex, IL -4 receptor, INF-γ receptor alpha chain, MHC class I, MHC class II, T cell receptor, insulin receptor, insulin receptor, insulin receptor tyrosine kinase and human growth hormone receptor. Including but not limited to receptors with known structures (including variants), including receptors, However, it is not limited to these.

  The definition of pharmaceutical protein also includes soluble proteins that can act as vehicles for immunogenic sequence delivery. Examples of soluble proteins include, but are not limited to, albumin, globulins, other proteins present in blood and other body fluids, and any other protein that is substantially non-immunogenic. . As used herein, “substantially non-immunogenic protein” means any protein that does not elicit an immune response in a subject. Substantially non-immunogenic proteins can be naturally occurring, synthetic, or modified using recombinant techniques known to those skilled in the art. Preferably, soluble proteins used as delivery vehicles include Zn-alpha 2-glycoprotein (Sanchez, LM, (1997) Proc. Natl. Acad. Sci., 94: 4626-4630; Sanchez, LM, et al., (1999) Science, 283: 1914-1919; both of which are hereby incorporated by reference), human serum albumin (HSA), IgG and other substantially non-immunogenic proteins. Including, but not limited to.

  In particular, preferred industrial target proteins include protease (including but not limited to papain, subtilisin) cellulases (endoglucanases I, II and III, exoglucanases, xylanases, ligninases, cellobiohydrolase I, Having a known structure, including but not limited to II and III, carbohydrase (including but not limited to glucoamylase, α-amylase, glucose isomerase) and lipase (Including mutants), but are not limited to these.

  In particular, preferred agricultural target proteins include xylose isomerase, pectinase, cellulase, peroxidase, rubisco, ADP glucose furophosphorylase, and enzymes involved in oil biosynthesis, sterol biosynthesis, carbohydrate biosynthesis and secondary metabolite synthesis Including, but not limited to, those having a known structure (including mutants).

  In a preferred embodiment, the methods of the invention include starting with a target protein and using computerized analysis to generate a set of primary sequences. There are a variety of computerized methods that can be used, including sequence alignments, structural alignments, structural prediction models, databases, or (preferably) protein design automation computerized analysis of related proteins. Collectively, these computer processing methods are referred to herein as “computerized protein design algorithms”. Similarly, created by perturbing the starting structure (using any number of techniques such as molecular mechanics calculations, Monte Carlo analysis, etc.) and altering the protein (including changes in torsion angles of the main and side chains) A primary mutant sequence library can be generated through sequence screening using a set of scaffold structures. The optimal sequence is selected for each starting structure (or several sets of superordinate sequences) to create a primary mutant sequence library.

  Some of these techniques result in a “scoring”, “ranking”, or “filtering” based on a certain criterion of a list of sequences in the primary library. In some embodiments, the sequence listing generated without ranking can then be ranked or filtered using the techniques outlined below.

  In general, there are various computer processing methods that can be used to generate a primary mutant sequence library. Again, all of these can be considered as computerized protein design algorithms. In a preferred embodiment, sequence-based methods are used. Alternatively, a structure-based method is used, such as the PDA® detailed below. Other models for assessing the relative energy of sequences with high accuracy include Warshel, Computer Modeling of Chemical Reactions in Enzymes and Solutions, Wiley & Sons, New York, (1991, which is incorporated herein by reference. ) Is included.

  Similarly, molecular mechanics calculations can be used to compute sequences by computing mutant sequence scores individually and editing the ranked list.

  In a preferred embodiment, residue pair potentials can be used to score sequences during computational screening (Miyazawa et al. Macromolecules 18 (3): 5 34-552 (1985), hereby incorporated by reference).

  In a preferred embodiment, a sequence profile evaluation (Bowie et al. Science 253 (5016): 164-70 (1991) is incorporated herein by reference) and / or average power to score sequences. (Hendlich et al. J. Mol. Biol. 216 (1): 167-180 (1990), which is also incorporated herein by reference). These methods can function to screen for fidelity to protein structure to assess the harmony between sequence and 3D protein structure. By using various scoring functions to evaluate the sequence, various regions of the sequence space can be sampled by computerized screening.

  In addition, scoring functions can be used to screen for sequences that create metal or cofactor binding sites in proteins (Hellinga et al. Fold Des. 3 (1): R1-8 (1998), published by reference). Part of the description). Similarly, scoring functions can be used to screen for sequences that create disulfide bonds in proteins. Try these possibilities to specifically alter the protein structure to introduce new structural motifs.

  In a preferred embodiment, sequence and / or structural alignment programs can be used to generate a primary library. As known in the art, there are numerous sequence-based alignment programs; for example, Smith-Waterman search, Needleman-Wunsch, Double Affine Smith-Waterman, frame search, Griskov / GCG profile search, Griskov / GCG Includes profile scan, profile frame search, Bucher generalized profile, Hidden Markov model, Hframe, Double Frame, Blast, Psi-blast, Clustal, and GeneWise.

  The source of the sequence can vary widely, and SCOP (Hubbard et al. Nucleic Acids Res 27 (1): 254-256. (1999)); PFAM (Bateman et al. Nucleic Acids Res 27 (1): 260- 262. (1999)); VAST (Gibrat et al. Curr Opin Struct Biol 6 (3): 377-385. (1996)); CATH (Orengo et al. Structure 5 (8): 1093-1108. (1997) All of which are hereby incorporated by reference); PhD Predictor (http://www.embl-heidelberg.de/predictprotein/predictprotein.html); Prosite (Hofmann et al. Nucleic Acids Res 27 (1): 215-219. (1999) part of this specification by reference); PIR (http://www.mips.biochem.mpg.de/proj/protseqdb/); GenBank (http: //www.ncbi.nlm.nih.gov/); PDB (www.rcsb.org) and BIND (Bader et al. Nucleic Acids Res 29 (1): 242-245. (2001) Including, but not limited to, obtaining a sequence from one or more known databases.

  In addition, sequences from these databases can be subjected to continuous analysis or gene prediction; Wheeler et al. Nucleic Acids Res 28 (1): 10-14. (2000) and Burge and Karlin, J Mol Biol 268 (1) : 78-94. (1997) (both are hereby incorporated by reference).

  There are a number of sequence alignment schemes that can be used, as is known in the art. For example, sequence homology based on alignment methods can be used to create protein sequence alignments with respect to the target structure (Altschul et al. J. Mol. Biol. 215 (3): 403 (1990), by reference) Part of this specification). These sequence alignments are examined to determine the observed sequence variation. These sequence variations are tabulated to define the primary library. In addition, these methods can also be used to generate secondary libraries, as further outlined below.

  Sequence-based alignments can be used in a variety of ways. For example, some of the related proteins can be aligned as is known in the art and define “variable” and “conserved” residues; that is, altered residues among members of the family. Groups or residues that remain the same can be defined. These results can be used to generate the probability table outlined below. Similarly, these sequence variations can be tabulated, and from these, a secondary library is defined as defined below. Alternatively, permissible sequence variations can be used to define the possible amino acids at each position in computerized screening. Another variation biases the scores of amino acids that occur in sequence alignments, thereby increasing the likelihood that those amino acids will be found in computerized screening, but still allow for other amino acids to be considered. This bias results in a focused primary library, but does not exclude amino acids that are not found in the alignment from consideration. In addition, numerous other types of bias may be introduced. For example, diversification may be forced; ie, “conserved” residues are selected and modified to force the protein to diversify, thus sampling a larger portion of the sequence space. Alternatively, positions that are highly variable (i.e., less conserved) among family members can be randomized using either all or a subset of amino acids. Similarly, abnormal residues may eliminate either positional abnormalities or side chain abnormalities.

  Similarly, structural alignments of structurally related proteins can be made to generate a sequence alignment. There are such a wide variety of known structural alignment programs. For example, NCBI's VAST (http://www.ncbi.nlm.nih.gov:80/Structure/VAST/vast.shtml); SSAP (Orengo and Taylor, Methods Enzymol 266 (617-635 (1996)) SARF2 ( (Alexandrov, Protein Eng 9 (9): 727-732. (1996)) CE (Shindyalov and Bourne, Protein Eng 11 (9): 739-747. (1998)); (Orengo et al. Structure 5 (8): 1093-108 (1997); see Dali (Holm et al. Nucleic Acid Res. 26 (1): 316-9 (1998), all of which are incorporated herein by reference). These structurally generated sequence alignments can be examined to determine the variation of the generated sequence.

  A primary mutant sequence library can be generated by predicting secondary structure from a sequence and then selecting sequences that are consistent with the predicted secondary structure. Threading (Bryant and Altschul, Curr Opin Struct Biol 5 (2): 236-244. (1995)); Profile 3D (Bowie et al. Methods Enzymol 266 (598-616 (1996)); MONSSTER (Skolnick et al J Mol Biol 265 (2): 217-241. (1997); Rosetta (Simons et al. Proteins 37 (S3): 171-176 (1999); PSI-BLAST (Altschul and Koonin, Trends Biochem Sci 23 (11) ): 444-447. (1998)); Impala (Schaffer et al. Bioinformatics 15 (12): 1000-1011. (1999)); HMMER (McClure et al. Proc Int Conf Intell Syst Mol Biol 4 (155-164) (1996)); Clustal W (http://www.ebi.ac.uk/clustalw/); BLAST (Altschul et al. J Mol Biol 215 (3): 403-410. (1990)); Helix-coil Including but not limited to transition theory (Munoz and Serrano, Biopolymers 41: 495, 1997), neural networks, local structure alignments, and others (see, eg, Selbig et al. Bioinformatics 15: 1039, 1999) There are not many secondary structure prediction methods.

  Similarly, as outlined above, other computer processing methods are known, such as sequence profiling (Bowie and Eisenberg, Science 253 (5016): 164-70 (1991)), rotamer library selection (Dahiyat and Mayo, Protein Sci5 (5): 895-903 (1996), Dahiyat and Mayo, Science278 (5335): 82-7 (1997); Desjarlais and Handel, Protein Science 4: 2006-2018 (1995), Harbury et al. PNAS USA92 (18): 8408-8412 (1995); Konoet al. Proteins: Structure, Function and Genetics 19: 244-255 (1994); Hellinga and Richards, PNAS USA 91: 5803-5807 (1994)) and residue pairs Potential (Jones, Protein Science 3: 567-574 (1994)), PROSA (Heindlich et al. J. Mol. Biol. 216: 167-180 (1990); THREADER (Jones et al. Nature 358: 86-89 ( 1992), and Simons et al. (Proteins, 34: 535-543, 1999), Levitt and Gerstein (PNAS USA, 95: 59135920, 1998), Godzik et al. PNAS, V89, PP 12098-102; Godzik and Skolnick (PNAS USA, 89: 12098102, 1992), Godzik et al. (J. Mol. B iol. 227: 227-38, 1992) and other two-fold methods (Gribskov et al. PNAS 84: 4355-4358 (1987) and Fischer and Eisenberg, Protein Sci. 5: 947-955 (1996), Rice and Eisenberg J. Mol. Biol. 267: 1026-1038 (1997)), but are not limited to these. Part of the description. Furthermore, described in Koehl and Levitt (J. Mol. Biol. 293: 1161-1181 (1999); J. Mol. Biol. 293: 1183-1193 (1999); incorporated herein by reference). Other computer processing methods, such as those described, can be used to create protein sequence libraries, and in some cases, smaller secondary live for use in experimental screening for improved properties and functions Can be used to generate a rally.

  In addition, there are computer processing methods based on force field calculations such as SCMF that can be used in the same way as SCMF. Delarue et al. Pac. Symp. Biocomput. 109-21 (1997), Koehl et al. J. Mol. Biol. 239: 249 (1994); Koehl et al. Nat. Struc. Biol. 2: 163 (1995) Koehl et al. Curr. Opin. Struct. Biol. 6: 222 (1996); Koehl et al. J. Mol. Bio. 293: 1183 (1999); Koehl et al. J. Mol. Biol. 293: 1161 (1999); Lee, J. Mol. Biol. 236: 918 (1994); Vasquez Biopolymers 36: 53-70 (1995); all of which are hereby incorporated by reference. . Other force field calculations that can be used to optimize the conformation of sequences within computer processing methods or to generate new optimized sequences as outlined here , OPLS-AA (Jorgensen et al. J. Am. Chem. Soc. (1996), v 118, pp 11225-11236; Jorgensen, WL; BOSS, Version 4.1; Yale University: New Haven, CT (1999) OPLS (Jorgensen et al. J. Am. Chem. Soc. (1988), v 110, pp 1657ff; Jorgensen et al. J Am. Chem. Soc. (1990), v 112, pp 4768ff); UNRES ( United Residue Forcefield; Liwo et al. Protein Science (1993), v 2, pp1697-1714; Liwo et al. Protein Science (1993), v 2, pp1715-1731; Liwo et al. J. Comp. Chem. (1997 ), v 18, pp849-873; Liwo et al. J. Comp. Chem. (1997), v 18, pp874-884; Liwo et al. J. Comp. Chem. (1998), v 19, pp259-276 ; Forcefield for Protein Structure Prediction (Liwo et al. Proc. Natl. Acad. Sci. USA (1999), v 96, pp5482-5485); ECEPP / 3 (Liwo et al. JP rotein Chem 1994 May; 13 (4): 375-80); AMBER 1.1 force field (Weiner et al. J. Am. Chem. Soc. v 106, pp765-784); AMBER 3.0 force field (UC Singh et al. Proc. Natl. Acad. Sci. USA. 82: 755-759); CHARMM and CHARMM22 (Brooks et al. J. Comp. Chem. V 4, pp 187-217); cvff3.0 (Dauber-Osguthorpe et al. (1988) Proteins: Structure, Function and Genetics, v4, pp31-47); including, but not limited to, cff91 (Maple et al. J. Comp. Chem. V15, 162-182); and DISCOVER (cvff and cff91) and AMBER force fields are used in the INSIGHT molecular modeling package (Biosym / MSI, San Diego California) and HARMM is used in the QUANTA molecular modeling package (Biosym / MSI, San Diego California) (these Are all incorporated herein by reference). In fact, as outlined below, these force field methods can be used to generate a secondary library directly; that is, they do not generate a primary library; rather, these methods create a probability table. Based on that, secondary libraries can be generated directly, for example by using these force fields during SCMF calculations.

  In a preferred embodiment, the computer processing method used to generate the primary library is U.S.S.N.60 / 061,097, 60 / 043,464, 60 / 054,678, 09 / 127,926, 09. / 782,004 and PCT US98 / 07254, all of which are hereby incorporated by reference, the Protein Design Automation® (PDA®) technique. Again, as outlined herein, each of the above methods can be referred to as a “protein design algorithm”, “computerized protein design algorithm”, or “computerized protein design method”.

  Briefly, the PDA® protein design technique can be described as follows. A known protein structure is used as a starting point. The residue to be optimized is then identified, which can be the entire sequence or a subset (s) thereof. Next, the side chain at any position to be changed is removed. The resulting structure consisting of the protein main chain and the remaining side chains is called a template. Each variable residue position is then preferably classified as a core residue, surface residue or boundary residue. Each classification defines a subset of possible amino acid residues at that position (eg, core residues are generally selected from a set of hydrophobic residues, surface residues are generally selected from hydrophilic residues, and boundaries The residue can be either). Each amino acid can be represented by an independent set of all conformers of each allowed side chain, called a rotamer. That is, to reach the optimal sequence for the main chain, all possible rotamer sequences must be screened, where each main chain position is in each possible rotamer state. It can be occupied by an amino acid, or a subset of amino acids, and thus a subset of rotamers.

  Two sets of interactions are then calculated at all positions for each rotamer. That is, the interaction of rotamer side chains with all or part of the main chain (also referred to as “singles” energy, rotamer / template or rotamer / main chain energy), and rotamer side chains and other Interaction with other possible all rotamers at all positions or a subset of other positions ("doubles" energy, also called rotamer / rotomer energy). The energy of each of these interactions is calculated through the use of various scoring functions, including van der Waals force energy, hydrogen bond energy, secondary structure tendency energy, surface region solvation energy and electrostatics. Is included. Therefore, the total energy of the interaction of each rotamer with both the backbone and other rotamers is calculated and stored in matrix form.

The unambiguous nature of the set of rotamers allows a simple calculation of the number of rotamer sequences to be tested. For a length-n backbone with m possible rotamers at each position, there are m ⁿ possible rotamer sequences, the number increasing exponentially with sequence length, in real time It is impractical or impossible to calculate. Therefore, in order to solve this combination search problem, a “dead end elimination” (DEE) calculation is performed. The DEE calculation shows that if the worst total interaction of the first rotamer is still better than the best total interaction of the second rotamer, the second rotamer is a global optimum solution. Based on the fact that it cannot be part. Since the energies of all rotamers have already been calculated, the DEE method requires a sum that spans the entire sequence length to test and eliminate rotamers, and the calculation is much faster. DEE can be rerun while comparing rotamer pairs or rotamer combinations, and ultimately a single sequence representing the global optimum energy is determined.

  Once a global answer is found, a Monte Carlo search can be performed to generate a ranked list of sequences in the vicinity of the DEE answer. Starting from the DEE answer, change the random position to another rotamer and calculate the new sequence energy. If the new sequence meets the acceptance criteria, it is used as a starting point for another jump. After a predetermined number of jumps, a rank ordered list of sequences is generated.

  Monte Carlo search is a sampling technique for exploring the array space around a global minimum or finding a new local minimum distance in the array space. There are other sampling techniques that can be used, including Boltzmann sampling, genetic algorithm techniques, and simulated annealing, as further outlined below. In addition, for all sampling techniques, the type of jump allowed is changeable (eg random jump to random residues, biased jump (eg to or from wild type), biased Jump to a residue (eg, to or from a similar residue). Similarly, for all sampling techniques, the acceptance criteria for whether sampling jumps are allowed can be changed.

  The protein backbone (nitrogen, carbonyl carbon, α-carbon and carbonyl oxygen along the α-carbon to β-carbon vector direction, as outlined in USS N.09 / 127926. (In the case of natural proteins) can be modified prior to computational analysis by changing a set of parameters called super-secondary structure parameters.

  Once the protein structure backbone is generated (as outlined above, with modifications) and entered into the computer, it is added if no apparent hydrogen is included in the structure (eg, the structure is X-ray crystal If generated by science, hydrogen must be added). After hydrogen addition, structural energy minimization is performed to relax hydrogen and other atoms, bond angles and bond lengths. In a preferred embodiment, this is done by performing a conjugate gradient minimization (Mayoet al. J. Phys. Chem. 94: 8897 (1990)) of atomic coordinate positions consisting of several steps, Dreiding's force field is minimized without any involvement. Generally, about 10 to about 250 steps are preferred, and about 50 is most preferred.

  The protein backbone structure includes at least one variable residue position. As is known in the art, protein residues or amino acids are generally sequentially numbered starting from the N-terminus of the protein. Therefore, a protein having a methionine at its N-terminus has a methionine at the 1st position of the residue or amino acid, and the next residues are at positions 2, 3, 4 and so on. At each position, the wild-type (ie, naturally produced) protein can have one of at least 20 amino acids in any number of rotamers. As used herein, “variable residue position” refers to the amino acid position of a protein that is not fixed as a specific residue or rotamer, generally a wild-type residue or rotamer in the design method. Means.

  In a preferred embodiment, all residue positions of the protein are variable. That is, any amino acid side chain can be modified in the method of the present invention. This is particularly desirable for small proteins, but the present invention allows for the design of large proteins as well. There is no theoretical limit to the length of protein that can be designed in this way, and there are practical computational limitations.

  In another preferred embodiment, only some of the residue positions of the protein are variable and the rest are “fixed”, ie they are identified in the three-dimensional structure as being in a set conformation. Is done. In some embodiments, the fixed position remains in its original conformation (which may or may not be correlated with the specific rotamer of the rotamer library being used). Alternatively, the residue can be fixed as a non-wild type residue. For example, when a known site-directed mutagenesis technique indicates that a particular residue is desirable (eg, to eliminate a proteolytic site or to alter the substrate specificity of an enzyme), the residue is fixed as a particular amino acid. Can be done.

  Alternatively, as discussed below, the methods of the invention can be used to newly assess mutations. In another preferred embodiment, the fixed position may be “floating”; the amino acid at that position is fixed, but another rotamer of that amino acid is tested. In this embodiment, the variable residue can be at least one, or generally from 0.1% to 99.9% of the total number of residues. Thus, for example, it is possible to change only a small number (or one) of a residue or a majority of residues (any possibility between them).

  In a preferred embodiment, the residues that can be fixed include, but are not limited to, structural or biological functional residues; alternatively, the biological functional residues are specifically It does not have to be fixed. For example, residues known to be important for biological activity, such as enzyme active sites, enzyme substrate binding sites, binding sites for binding partners (ligand / receptor, antigen / antibody, etc.), biological Residues that form a phosphorylation or glycosylation site critical for function, or a structurally important residue such as a disulfide bridge, a metal binding site, a critical hydrogen bonding residue, a backbone such as proline or glycine Residues critical for conformation, residues critical for interaction packing, etc. may all be fixed in one conformation, or as a single rotamer, or may be “floating”.

  Similarly, residues that can be selected as variable residues include undesirable biological properties such as susceptibility to proteolysis, dimerization or aggregation sites, glycosylation sites that can induce an immune response, unwanted binding activity. Undesirable allostery, binding may be conserved but impart undesirable enzyme activity, and the like.

  In a preferred embodiment, each variable position is classified as a core, surface or boundary residue position, but in some cases, as described below, the variable position can be set to glycine to minimize the backbone. Further, as outlined herein, the residues need not be classified, can be selected as variable, and any set of amino acids can be used. Any combination of core, surface and boundary location can be used. That is, core, surface and boundary residues; core and surface residues; core and boundary residues, and surface and boundary residues, and core residues alone, surface residues alone or boundary residues alone.

  Classification of residue positions as cores, surfaces or boundaries can be done in several ways, as will be apparent to those skilled in the art. In a preferred embodiment, the classification is done by scanning images of the original protein backbone structure including side chains and assigning the classification based on the subjective assessment of those skilled in protein modeling. Alternatively, a preferred embodiment is a template as outlined in U.S.S.N.60 / 061,097, 60 / 043,464, 60 / 054,678, 09 / 127,926 and PCT US98 / 07254. A Cα-Cβ vector orientation assessment is used for a solvent accessible surface calculated using only Cα atoms. Alternatively, the surface area can be calculated.

  Once each variable position is classified as a core, surface or boundary, a set of amino acid side chains, ie a set of rotamers, is assigned to each position. That is, the program selects a set of possible amino acid side chains that are recognized as considerations at a particular position. Subsequently, once the possible amino acid side chains are selected, the set of rotamers evaluated at a particular position can be determined. Accordingly, the core residue is generally selected from the group of hydrophobic residues consisting of alanine, valine, isoleucine, leucine, phenylalanine, tyrosine, tryptophan and methionine (in some embodiments, the Van der Waals scoring function When α-scaling factor is low, methionine is removed from the set), a set of rotamers for each core position (total rotamers when using a backbone independent library, and rotamer dependent) The subset), when using a functional backbone, potentially includes these 8-amino acid side-chain rotamers. Similarly, the surface location is generally selected from the group of hydrophilic residues consisting of alanine, serine, threonine, aspartic acid, asparagine, glutamine, glutamic acid, arginine, lysine and histidine. Thus, the set of rotamers for each surface position includes these 10 residue rotamers. Finally, the boundary position is generally selected from alanine, serine, threonine, aspartic acid, asparagine, glutamine, glutamic acid, arginine, lysine, histidine, valine, isoleucine, leucine, phenylalanine, tyrosine, tryptophan and methionine. Thus, the set of rotamers for each boundary position includes all rotamers of these 17 residues (assuming that cysteine, glycine and proline are not used, but they can be used). . In addition, in some preferred embodiments, 18 naturally occurring amino acids are used, particularly all except cysteine and proline, which are known to be destructive.

  Therefore, as will be appreciated by those skilled in the art, there is a computational advantage in classifying residue positions since the number of calculations can be reduced. It should also be noted that there may be situations where the set of cores, boundaries and surface residues is altered from the above. For example, in some situations, one or more amino acids are added or subtracted from the set of allowed amino acids. For example, some proteins that dimerize or multimerize or have a ligand binding site may contain hydrophobic surface residues and the like. Furthermore, residues that do not allow favorable interaction with the helix “capping” or α-helix dipole can be subtracted from the set of allowed residues. This amino acid group change can be made on a residue by residue basis.

  In a preferred embodiment, proline, cysteine and glycine are not included in the list of possible amino acid side chains and thus rotamers of these side chains are not used. However, in a preferred embodiment, the φ angle where the variable residue position is greater than 0 ° (ie, 1) the carbonyl carbon of the preceding amino acid, 2) the nitrogen atom of the current residue, 3) the α carbon of the current residue, and 4) The dihedral angle defined by the carbonyl carbon of the current residue), the position is set to glycine to minimize main chain distortion.

Once a group of possible rotamers has been assigned to each variable residue position, processing proceeds as outlined in U.S.S.N.09 / 127,926 and PCT US98 / 07254. This processing step requires analysis of the interaction between rotamers and the interaction between rotamers and protein backbones in order to produce an optimal protein sequence. In extreme simplification, processing involves first calculating the interaction energy of a rotamer relative to the main chain itself or other rotamers by using some scoring function. Preferred PDA® technique scoring functions include van der Waals potential scoring functions, hydrogen bond potential scoring functions, atomic solvation scoring functions, secondary structure propensity scoring functions, and electrostatic scoring functions. However, it is not limited to these. In addition, as reported below, at least one scoring function is used to score each position, such as position classification or advantageous interaction with the α-helix dipole. It can vary depending on other considerations. As outlined below, the total energy used in the calculation is the sum of the energies of each scoring function used at a particular position, as generally shown in Equation 1: Equation 1 E _total = nE _vdw + nE _as + nE _h-bond + nE _as + nE _elec

In Equation 1, the total energy is van der Waals potential energy (E _vdw ), atomic solvation energy (E _as ), hydrogen bond energy (E _h-bond ), secondary structure energy (E _ss ), and electrostatic interaction. It is the sum of energy (E _elec ). The word n is 0 or 1 depending on whether this word should be considered with respect to a particular residue position.

  U.S.S.N. 60 / 061,097, 60 / 043,464, 60 / 054,678, 09 / 127,926 and PCT US98 / 07254, alone or in combination Any combination of these scoring functions can be used. Once the scoring function to be used has been identified for each variable position, the preferred first step in the computational analysis involves measuring the interaction of each possible rotamer with all or part of the protein. That is, the energy of interaction of each possible rotamer with each backbone or other rotamer at each variable residue position, as measured by one or more of the scoring functions, is calculated. In a preferred embodiment, the interaction of each rotamer with the rest of the protein, both the entire template and all other rotamers, takes place. However, as outlined above, it is possible to model only a portion of a protein, for example a domain of a large protein, and thus in some cases it is not necessary to consider the entire protein. The term “portion” as used herein refers to a fragment of a protein with respect to a protein. This fragment can range in size from 10 amino acid residues to the entire amino acid sequence minus one amino acid. Thus, as used herein, the term “portion” refers to a fragment of a nucleic acid with respect to that nucleic acid. This fragment can range in size from 10 nucleotides to the entire nucleic acid sequence minus 1 nucleotide.

  In a preferred embodiment, the first stage of computer processing is performed by calculating two sets of interactions for each rotamer at all positions. That is, the interaction of the rotamer side chain with the template or main chain (“single-s” energy) and all other possibilities at all other positions with the rotamer side chain, whether to change or float its position Interaction ("doubles" energy) with various rotamers. It should be understood that the backbone in this case includes both the protein structure backbone atom and the fixed residue, if any, where the fixed residue is defined as a specific conformation of an amino acid. .

That is, "single" (rotamer / template) energy is used for all possible rotamer and backbone interactions at all variable residue positions using some or all of the scoring functions. Calculated. Thus, for the hydrogen bond scoring function, all rotamer and main chain hydrogen bond atoms are evaluated and E _HB is calculated for each possible rotamer at all variable positions. Similarly, for van der Waals scoring functions, all rotamer atoms are compared to template atoms (typically excluding the main chain atom of its own residue) and all variable residue positions. Calculate E _vdW for each possible rotamer in. Furthermore, van der Waals energy is generally not calculated when atoms are joined by three or fewer bonds. For atomic solvation scoring functions, the surface of the rotamer is measured against the template surface and E _as is calculated for each possible rotamer at all variable residue positions. Secondary structure trend scoring functions can also be considered as singles energy, so the total singles energy can include an _Ess term. As will be apparent to those skilled in the art, depending on the physical distance between the rotamer and the template position, many of these energy terms approach zero; that is, the farther apart the two parts, the lower the energy .

For the calculation of “doubles” energies (rotamers / rotamers), the interaction energy of each possible rotamer is compared to all possible rotamers at all other variable residue positions. Thus, the “doubles” energy uses all or all of the scoring functions to calculate all possible rotamers at all variable residue positions and all possible rotations at all other variable residue positions. Calculated for interactions with isomers. Thus, for the hydrogen bond scoring function, all hydrogen bond atoms of the first rotamer and all possible second rotamer atoms are evaluated, and E _HB is possible at any two variable positions. Calculated for each rotamer pair. Similarly, for the van der Waals scoring function, all atoms of the first rotamer are compared with all atoms of all possible second rotamers, and each possible rotation at every two variable residue position. E _vdW is calculated for the isomer pair. For the atomic solvation scoring function, the surface of the first rotamer is measured against the surface of all possible second rotamers and each possible rotamer pair at all two variable residue positions. E _as is calculated for. Secondary structure propensity scoring functions need not be implemented as “doubles” energy because they are considered components of “single” energy. As will be appreciated by those skilled in the art, many of these doubles energy terms approach zero, depending on the physical distance between the first and second rotamers. That is, the further away the two parts, the lower the energy.

  In addition, as will be appreciated by those skilled in the art, various force fields can be used in PDA® technique calculations, such as Dreifing I and Dreiding II (Mayo et al, J. Phys. Chem. 948897 (1997)), AMBER (Weiner et al., J. Amer. Chem. Soc. 106: 765 (1984) and Weiner et al., J. Comp. Chem. 106: 230 (1986)), MM2 (Allinger J. Chem. Soc. 99: 8127 (1977), Liljefors et al., J. Com. Chem. 8: 1051 (1987)); MMP2 (Sprague et al., J. Comp. Chem. 8: 581 (1987)); CHARMM (Brooks et al., J. Comp. Chem. 106: 187 (1983)); GROMOS; and MM3 (Allinger et al., J. Amer. Chem. Soc. 111: 8551 (1989)), OPLS-AA (Jorgensen, et al., J. Am. Chem. Soc. (1996), v 118, pp 11225-11236; Jorgensen, WL; BOSS, Version 4.1; Yale University: New Haven, CT (1999)); OPLS (Jorgensen, et al., J. Am. Chem. Soc. (1988), v 110, pp 1657ff; Jorgensen, et al., J. Am. Chem. Soc. (1990), v112, pp 4768ff); UNRES (United Residue Forcefield ; Liwo, et al., Protein Science (1993), v2, pp 1697-1714; Liwo et al., Protein scienc e (1993), v2, pp1715-1731; Liwo, et al., J. Comp. Chem. (1997), v 18, pp849-873; Liwo, et al., J. Comp. Chem. (197), v18, pp874-884; Liwo, et al., J. Comp. Chem. (1998), v 19, pp259-276; Forcefield for Protein Structure Prediction (Liwo, et al., Proc. Natl. Acad. Sci. USA (1999), v 96, pp5482-5485); ECEPP / 3 (Liwo et al., J Protein Chem 1994 May; 13 (4): 375-80); AMBER 1.1 force field (Weiner, et al., J. Am. Chem. Soc. V106, pp765-784); AMBER 3.0 force field (UC Singh et al., Proc. Natl. Acad. Sci. USA. 82: 755-759); CHARMM and CHARMM22 (Brooks, et al. , J. Comp. Chem. V4, pp 187-217); cvff3.0 (Dauber-Osguthorpe, et al., (1988) Proteins: Structure, Function and Genetics, v4, pp31-47); cff91 (Maple, et al., J. Comp. Chem. v15, 162-182); and DISCOVER (cvff and cff91) and AMBER force fields are also included in the INSIGHT molecular modeling package (Biosym / MSI, San) Diego California), HARMM is a QUANTA molecular modeling Kkeji (Biosym / MSI, San Diego California) was used to, all of which are incorporated herein by reference.

  Once the singles and doubles energies are calculated and stored, a second stage of computer processing can be performed. As outlined in U.S.S.N.09 / 127926 and PCT US98 / 07254, preferred embodiments utilize a dead end exclusion (DEE) stage, and preferably a Monte Carlo stage.

  In overview, the PDA® technique has three elements that can be varied to modify the output (eg, primary library): a scoring function used for processing; a filtering technique, and a sampling technique. These functions can be used sequentially or substantially simultaneously. For example, a scoring function may be used in parallel with the filtering technique.

  In a preferred embodiment, the scoring function can be modified. In a preferred embodiment, the scoring function outlined above can be biased or weighted in various ways. For example, a bias towards or away from the reference sequence or family of reference sequences can be made; for example, a bias towards wild-type or homologous residues can be used. Similarly, the entire protein or fragment thereof can be biased; for example, the active site can be biased toward wild-type residues, or domain residues can be directed toward certain desirable physical properties. Furthermore, a bias towards or against increased energy can be created. Additional scoring function biases include, but are not limited to, applying electrostatic potential gradients or hydrophobic gradients, adding substrates or binding partners to the calculation, or biasing towards the desired charge or blocking properties. .

  In addition, in other embodiments, there are a variety of additional scoring functions that can be used. Additional scoring functions include, but are not limited to, torsional potential, or residue pair potential or residue entropy potential. Such additional scoring functions can be used alone or as a function for library processing after initial scoring of the library.

  In a preferred embodiment, various processing filtering techniques can be performed, including but not limited to DEE and its associated counterparts. Further filtering techniques include branch-and-bound techniques for finding optimal sequences (Cordon and Mayo, Structure Fold. Des. 7: 1089-98, 1999) and exhaustive enumeration of sequences. It is not limited. However, it should be noted that certain techniques can be performed without filtering techniques; for example, sampling techniques can be used to find good sequences without filtering.

  As will be appreciated by those skilled in the art, once an optimal sequence or set of sequences has been generated (or they need not be optimized or ordered), various sequence space sampling methods can be added to the preferred Monte Carlo method. Or instead of a Monte Carlo search. That is, once a sequence or set of sequences has been generated, preferred methods utilize sampling techniques that allow the generation of further related sequences for testing.

  These sampling methods include the use of amino acid substitutions, insertions or deletions, or recombination of one or more sequences. As outlined herein, the preferred embodiment utilizes a Monte Carlo search, which is a series of biased, systematic, or random jumps. However, there are other sampling techniques that can be used, including Boltzmann sampling, genetic algorithm techniques, and simulated annealing. In addition, for all sampling techniques, the types of jumps allowed can be modified (e.g. random jumps to random residues, biased jumps (e.g. towards wild type or away) ), Jump to a biased residue (toward or away from a similar residue, etc.), jump to the position of multiple residues together (two residues always change together, or Jumps in which the entire set of residues changes to other sequences (eg, recombination), as well as altering the acceptance criteria for whether sampling jumps are allowed for all sampling techniques It also allows for a wide search at high temperatures and a narrow search near local optimal values at low temperatures, which are incorporated herein by reference to Metropolis et al., J. Chem. Phys v21, pp1087, 1953 Teru.

  In addition, it should be noted that the preferred method of the present invention results in a ranked list of sequences; that is, the sequences are ranked or filtered based on certain objective criteria. However, as outlined herein, no ranking is done, for example, by directly generating the probability table to list (eg, using SCMF analysis or sequence alignment techniques) without ranking the sequences. It is possible to create a set of sequences. The sampling techniques outlined herein can be used in either situation.

  In a preferred embodiment, Boltzmann sampling is performed. As will be appreciated by those skilled in the art, a Boltzmann sampling temperature reference can be modified to perform a broad search at high temperatures or a narrow search near local optimal values at low temperatures (eg, Metropolis et al. al., J. Chem. Phys. 21: 1087, 1953).

  In a preferred embodiment, the sampling technique utilizes a genetic algorithm as described, for example, by Holland (Adaptation in Natural and Artificial Systems, 1975, Ann Arbor, U. Michigan Press). In general, genetic algorithms take generated sequences and recombine them computationally in a manner similar to “gene mixing” in the same manner as nucleic acid recombination events. Thus, a “jump” in genetic algorithm analysis is generally a multi-position jump. In addition, correlated multiple jumps can also be performed, as outlined below. Such jumps can be performed one or more recombination at a time, at various crossover positions, and can involve recombination of two or more sequences. In addition, deletions or insertions (random or biased) can be made. In addition, as outlined below, genetic algorithm analysis may be used after secondary library generation.

  In a preferred embodiment, the sampling technique uses simulated annealing as described, for example, in Kirkpatrick et al. (Science, 220: 671-680, 1983). Simulated annealing modifies the good or bad jump cutoff by modifying the temperature. That is, the severity of the cutoff is changed by changing the temperature. This makes it possible to perform a wide search at a high temperature to a new sequence space region and switch to a detailed search of the region by a narrow search at a low temperature.

  In addition, as outlined below, these sampling methods can be used in further processes to generate additional secondary libraries (sometimes referred to herein as tertiary libraries).

  Thus, primary libraries can be generated in a variety of computational methods, including structure-based methods such as PDA®, or sequence-based methods, or combinations as outlined herein.

  Computer processing produces a set of optimized candidate mutation sequences. The optimized candidate mutant protein sequence generally differs from the target protein sequence in a region critical for MHC, TCR or BCR binding. Preferably, each optimized candidate mutated sequence comprises at least about 1 mutated amino acid from the start or target sequence, with 3-5 being preferred. Preferably, the mutated residue is located in a non-contiguous region.

  Accordingly, in a preferred embodiment, the present invention is directed to a method of processing a target protein or fragment thereof by computer processing to produce a candidate mutant protein or a set of candidate mutant protein sequences.

  Thus, in a preferred embodiment, the candidate mutant protein of the invention comprises at least one M It has a different amino acid sequence from the target protein at the HC, TCR or BCR binding site. Preferably, if a less immunogenic protein is desired, the candidate mutant protein differs from the target protein by eliminating at least one MHC, TCR or BCR binding site. Alternatively, if a more immunogenic protein is desired, the candidate mutant protein differs from the target protein through the addition of at least one MHC, TCR or BCR binding site.

  Thus, computer processing results in a set of primary mutant sequences, which can be optimized protein sequences when using certain ranking or scoring functions. These optimized protein sequences are generally (but not always) significantly different from the target sequence employing the backbone. That is, each optimized protein sequence preferably contains at least about 5-10% mutated amino acids from the starting target or wild-type sequence, preferably at least about 15-20% change, and particularly at least about 30% change. preferable.

  In a preferred embodiment, a computerized immunogenic filter is applied to the set of primary library sequences. As used herein, “computerized immunogenic filter” refers to any of a number of scoring functions derived from data on MHC molecules, or binding of peptides to T or B cell epitopes. A computerized immunogenic filter is computed as part of the original computer calculation (eg, substantially simultaneously; eg, as one of the computer processing steps or as a scoring function during the original computer calculation). Can be applied prior to (eg, as a pre-filter) or after the original computer computation (eg, as a post-filter). For example, in a preferred embodiment, a computerized immunogenic filter is used as a post-filter: ie, to re-score a set of primary library sequences to eliminate potentially immunogenic sequences. Or a scoring function is used to introduce non-immunogenic sequences.

  In a preferred embodiment, a computerized immunogenic filter is applied at the same time, ie substantially simultaneously, when generating the primary library sequence.

  In another preferred embodiment, a computerized immunogenic filter is applied before the primary sequence set is computationally generated. Using this approach, depending on the desired result, a set of primary sequences is generated that potentially lack or contain immunogenic sequences. PDA® techniques are then performed on these sequences to identify those that retain the natural fold and are at least as stable as the starting target protein.

  In a preferred embodiment, PDA is used to structurally and chemically compensate for either the removal or addition of amino acid residues encoding linear epitopes represented by MHC class I and II molecules recognized by the TCR. (Registered trademark) technique is used.

  In a preferred embodiment, PDA® is used to structurally and chemically compensate for either the removal or addition of amino acid residues encoding conformational epitopes that are sensed by membrane-bound antibodies on naive B cells. ) Technique.

  Current knowledge about the rules of peptide selection by MHC molecules is the sequencing of natural peptide libraries extracted from peptides and MHC proteins, peptide binding to MHC molecules and mutagenesis into unknown CTL epitope sequences for T cell responses Derived from the analysis of the effects of MHC and the crystal structure analysis and molecular force field studies of the determined MHC peptide complex (Meister, GE, et al. (1995) Vaccine, 13: 581-591; Malios, RR , (1999) Bioinformatics Savoie, CJ et al. (1999) Pac Symp Biocomput., 182-9; Brusic, V., et al., (1998) Bioinformatics, Mallios, RR, (1998) J. Comp. Biol. , 5: 703-711; Altuvia, Y., et al. (1997) Human Immunology, 58: 1-11; Udaka, et al., (1995) J. Exp. Med., 181: 2097-2108; Hammer , J. et al. (1994) Behring. Inst. Mitt. 94: 124-132; Hemmer, B., et al., (2000) J. Immunol., 164: 861-871). In addition, a database of thousands of peptide sequences known to bind to MHC molecules was compiled (Buus, supra; Brusic, V., et al., (1998) Nucleic Acids Res., 26: 368-371 Rammensee, HG., Et al., (1999) Immunogenetics, 50: 213-219), and several techniques for analyzing the full-length protein sequence and predicting the presence of potential immunogenic sequences (Hiemstra, HS et al. (2000) Curr. Op. Immunol., 12: 80-84; Malios, RR, (1999) Bioinformatics, 15: 432-439; Sturniolo, T., et al. (1999) Nature Biotechnology, 17: 555-561; Brusic, V., et al., (1998) Bioinformatics, 14: 121-130; Mallios, RR, (1998) J. Comp. Biol., 5: 703 -711; Shastri, N. (1996) Curr. Op. Immunol., 8: 271-277; Hammer, J. (1995) Curr. Op. Immunol., 7: 263-269; Meister, GE, et al. (1995) Vaccine, 13: 581-591; Udaka, K., et al. (1995) J. Exp. Med., 181: 20972108; Hammer, J. et al. (1994) Behring. Inst. Mitt. 94 : 124-132; Hammer, J., et al. (19 94) J. Exp. Med., 180: 2353-2358; and, Rudenshky, AY, et al. (1991) Nature, 353: 622-627; Marshall, KW, et al., (1995) J. Immunology, 154: 5927-5933; Novak, EJ, (2001) J. Immunology, 166: 6665-6670; Cochlovius, B., et al., (2000) J. Immunology, 165: 4731-4741; Raddrizzani and Hammer, ( 2000) Brief Bioinform., 1 (2): 179-89; Hemmer, B., et al., (1998) J. Immunology, 160: 3631-3636; Gulukota, K., et al., (1997) J Mol. Biol., 1258-1267; Parker, et al., (1994) J. Immunology, 152: 163175; Berzofsky, JA, et al European Patent publication number 0279 994 A2); Fikes, J. et al., WO 01/41788, all of which are incorporated herein by reference).

In a preferred embodiment, primary mutant sequences are screened for peptide fragments that are potentially capable of binding to MHC class I molecules. MHC I ligands are mostly octa- or nonapeptides and represent MHC allele-specific sequence motifs determined by collectively sequencing natural segregants. Crystal structure analysis identified a peptide bond cleft, or groove, bordered by two α helices and one β pleated sheet. The cleavage is stabilized from below by non-covalently associated β2 microglobulin. A specific pocket in the binding groove accommodates the anchor residue of the peptide. Orientation of the peptides, NH ₂ - is determined by conserved side chain of MHC I protein to compensate for and COOH- terminus of charge.

  A given MHC class I peptide binding groove can bind to hundreds or thousands of different peptides that are identical or homologous only in a few side chain positions. A structural comparison of a number of class I peptide-MHC complexes revealed that this flexibility was achieved by the structurally equivalent binding of a small subset of each peptide residue. Among other things, the charge of the peptide backbone or the binding of polar atoms results in intrinsic side chain independent peptide-MHC interactions. This assembly of hydrogen bonds and van der Waals contacts helps to stabilize the binding of any peptide capable of adopting the required backbone conformation. Further interaction with a small number of peptide side chains supplements the main chain binding energy, making the peptides that bind to specific MHC molecules have some sequence selectivity (Madden, DR (1995) Annu. Rev. Immunol., 13: 587-622). Rules for identifying MHC I binding sites have been described in Altuvia, Y., et al (1997) Human Immunology, 58: 1-11; Meister, GE., Et al (1995) Vaccine: 6 : 581-591; Parker, KC, et al., (1994) J. Immunolgy, 152: 163; Gulukota, K., et al., (1997) J. Mol. Biol., 267: 1258-1267; Buus , S., (1999) Current Opinion Immunology, 11: 209-213; incorporated herein by reference). In addition, databases of MHC binding peptides such as TYPEITHI and MHCPEP are also available and can be used to identify potential MHC I binding sites (Rammensee, HG., Et al., (1999) Immunogenetics, 50 : 213-219; Brusic, V., et al., (1998) Nucleic Acids Research, 26: 368-371; incorporated herein by reference). Other methods for identifying MHC binding motifs include the allele-specific polynomial algorithm described by Fikes, J., et al., WO 01/41788.

  In a preferred embodiment, potential MHC class I binding sites are amino acid residues that structurally and chemically compensate for anchor residues removed to reduce or eliminate peptide binding to MHC class I molecules. Replaced. A possible MHC I binding motif is SYFPEITHI (Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219; http://134.2.96.221/scripts/MHCServer.dll/home.html) ); http://wehih.wehi.edu.au/mhcpep/, by adapting to published motif databases such as MHCEP (Brusic, B., et al., supra) or by neural networks (Gulukota , K, supra), polynomial (Gulukota, K., supra) ranking (Parker, KC, supra) and allele-specific polynomial algorithms (Fikes, J., et al., WO 01/41788) Identify by established methods such as In a further embodiment, non-anchor residues are replaced.

  In a preferred embodiment, specific cleavage motifs for antigen processing and presentation are removed. As used herein, “specific cleavage motif” means a motif that is specifically recognized as a proteolytic cleavage site by proteases involved in processing of antigenic determinants present in a protein (Schneider, SC, et al., (2000) J. Immunol., 165: 20-23; entirely incorporated herein by reference.). In other words, a specific cleavage motif, if present, is a motif that makes an antigenic determinant accessible for binding to MHC molecules and subsequent presentation on the APC surface. Preferably, the proteosome cleavage site is removed, reducing the availability of antigenic determinants for binding to MHC class I molecules. Possible proteosome cleavage sites are Kutter, C., et al., (2000) J. Mol. Biol., 298: 417-429 and Nussbaum, AK, et al., (2001) Immunogenetics, 53:87. Identified by using a prediction algorithm such as that described by -94, both of which are hereby incorporated by reference in their entirety.

  In a preferred embodiment, potential MHC class I binding sites are added to the target protein as a means of inducing cellular immunity. Preferably, at least one MHC class I binding site is added for each target protein. More preferably, at least two MHC class I binding sites are added for each target protein. More preferably, 3 to 5 MHC class I binding sites are added for each target protein. In other embodiments, up to 16 MHC class I binding sites may be added per target protein (Stienekemeier, M., et al., (2001) Proc Natl Acad Sci USA, 98: 13872-13877; Are incorporated herein by reference in their entirety). PDA® technology is used to ensure correct folding and stability of the modified target protein. Suitable target proteins are produced using soluble proteins such as Zn-alpha 2-glycoprotein (Sanchez, LM, et al., (1999) Science 283: 1914-9), or other target proteins of interest. Primary sequence libraries, but are not limited to these. A possible MHC I binding motif is SYFPEITHI (Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219; http://134.2.96.221/scripts/MHCServer.dll/home.html) ); http://wehih.wehi.edu.au/mhcpep/, by adapting to published motif databases such as MHCEP (Brusic, B., et al., supra) or by neural networks (Gulukota , K, supra), polynomials (Gulukota, K., supra), ranking (Parker, KC, supra) and allele-specific polynomial algorithms (Fikes, J., et al., WO 01/41788) ) By established methods such as

  In a preferred embodiment, specific cleavage motifs (defined above) are added for antigen processing and presentation. Preferably, a proteosome cleavage site is added to increase the availability of antigenic determinants for binding to MHC class I molecules. Possible proteosome cleavage sites are Kutter, C., et al., (2000) J. Mol. Biol., 298: 417-429 and Nussbaum, AK, et al., (2001) Immunogenetics, 53:87. Identified by using a prediction algorithm such as that described by -94, both of which are hereby incorporated by reference in their entirety.

  In a preferred embodiment, primary mutant sequences are screened for peptide fragments predicted to bind to MHC class II molecules. Class II ligands consist of 12 to 25 amino acids, 9 of which occupy the binding groove; 2 to 4 are anchored in the pocket. As in class I ligands, non-anchor amino acids play a secondary but still important role (Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219). The rules for identifying MHC II binding sites are described in Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219). Rules for identifying MHC II binding sites have been described in Hammer, J. et al., ( 1994) Behring. Inst. Mitt., 94: 124-132; Hammer, J. et al., (1994) J. Exp. Med., 180: 2353-2358; Mallios, RR (1998) J. Com. Biol ., 5: 703-711; Brusic, V., et al., (1998) Bioinformatics, 14: 121-130; Mallios, RR (1999) Bioinformatics, 15: 432-439; Marshall, KW, et al., (1995) J. Immunology, 154: 5927-5933; Novak, EJ, et al., (2001) J. Immunology, 166: 6665-6670; Cochlovius, B., et al., (2000) J. Immunology, 165: 4731-4741; and by Fikes, J., et al., WO 01/41788, all of which are hereby incorporated by reference.

  In a preferred embodiment, potential MHC class II binding sites are replaced with amino acid residues that structurally and chemically compensate for the removed anchor residues to eliminate MHC II binding sites. Preferably, the potential MHC II binding site is SYFPEITHI (Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219; http://134.2.96.221/scripts/MHCServer.dll/home .html) or MHCEP (http://wehih.wehi.edu.au/mhcpep/) to identify and match published motif databases. Alternatively, binding to class II molecules can be predicted by virtual matrix methods (Sturniolo, T, et al. (1999) Nature Biotechnology, 17: 555-561) and Raddrizzani, L. and Hammer, J., (2000) Brief. Bioinform., Both of which are hereby incorporated by reference) or the allele-specific polynomial algorithm described in Fikes, J., et al., WO 01/41788.   In a further embodiment, non-anchor residues are substituted.

  In a preferred embodiment, specific cleavage motifs as defined above for antigen processing and presentation are removed. Proteases involved in the processing of antigenic determinants for MHC class II molecules present in certain proteins include cathepsins B, D, E, L and asparaginyl endopeptidases (Schneider, SC, et al., (2000) J. Immunol., 165: 20-23; entirely incorporated herein by reference). Preferably, proteolytic cleavage sites are removed to reduce the availability of antigenic determinants for binding to MHC class II molecules. Possible proteolytic cleavage sites are Schneider, SC, et al., (2000) J. Immunol., 165: 20-23; and Medd and Chain, (2000) Cell & Developmental Biology, 11: 203-210. (Both of which are hereby incorporated by reference in their entirety).

  In a preferred embodiment, potential MHC class II binding sites are added to the target protein as a means of inducing cellular immunity. Preferably, at least one MHC class II binding site is added for each target protein. More preferably, at least two MHC class II binding sites are added per target protein. More preferably, 3 to 5 MHC class II binding sites are added per target protein. In other embodiments, up to 16 MHC class II binding sites may be added per target protein (Stienekemeier, M., et al., (2001) Proc Natl Acad Sci USA, 98: 13872-13877; As a part of this specification). PDA® technology is used to ensure correct folding and stability of the modified target protein. Suitable target proteins are produced using soluble proteins such as Zn-alpha 2-glycoprotein (Sanchez, LM, et al., (1999) Science 283: 1914-9), or other target proteins of interest. Primary sequence libraries, but are not limited to these. A possible MHC II binding motif is SYFPEITHI (Rammensee, H., et al., (1999) Immunogenetics, 50: 213-219; http://134.2.96.221/scripts/MHCServer.dll/home.html) ); http://wehih.wehi.edu.au/mhcpep/, by adapting to published motif databases such as MHCEP (Brusic, B., et al., supra), or virtual matrices ( Sturniolo, T, et al. (1999) Nature Biotechnology, 17: 555-561; Raddrizzani, L. and Hammer, J., (2000) Brief Bioinform., 1: 179-89) and allele-specific polynomial algorithms ( Identification by established methods such as Fikes, J., et al., WO 01/41788).

  In a preferred embodiment, a specific cleavage motif as defined above for antigen processing and presentation is added. Preferably, cathepsin B, D, E, L and asparaginyl endopeptidase proteolytic cleavage sites are added to increase the availability of antigenic determinants for binding to MHC class II molecules. Possible proteolytic cleavage sites are Schneider, SC, et al., (2000) J. Immunol., 165: 20-23; and Medd and Chain, (2000) Cell & Developmental Biology, 11: 203-210. (Both of which are hereby incorporated by reference in their entirety).

  In a preferred embodiment, potential MHC class I and class II binding sites are transferred to a primary sequence library generated using the target protein or other target protein of interest, with cellular immunity as defined above. Add as a means to guide.

  In preferred embodiments, only sequences modified by the computer processing methods described herein are considered.   In other embodiments, peptide sequences present in autologous proteins (ie, circulating human proteins such as immunoglobulins, albumin, etc.) are ignored.

  In a preferred embodiment, the primary mutant sequence is screened for peptide fragments that are predicted to function as T cell epitopes. In a preferred embodiment, potential T cell epitopes are replaced with amino acid residues that structurally and chemically compensate for the removed residues to eliminate T cell epitopes. Preferably, potential T cell epitopes are identified by fitting to published motif databases (Walden, P., (1996) Curr. Op. Immunol., 8: 68-74). Other T cell epitope identification methods useful in the present invention include Hemmer, B., et al. (1998) J. Immunol., 160: 3631-3636; Walden, P., et al. (1995) Biochemical Society. Transactions, 23; Anderton, SM, et al., (1999) Eur. J. Immunol., 29: 1850-1857; Correia-Neves, M., et al., (1999) J. Immunol., 163: 5471 Shastri, N., (1995) Curr. Op. Immunol., 7: 258-262; Hiemstra, HS, (2000) Curr. Op. Immunol., 12: 80-84; and Meister, GE, et al., (1995) Vaccine, 13: 581-591, all of which are incorporated herein by reference in their entirety.

  In a preferred embodiment, specific cleavage motifs as defined above for antigen processing and presentation are removed. Cleavage sites involved in processing of antigenic determinants present for MHC class I and / or class II molecules are removed as described above. In this way, proteolytic cleavage sites can be removed to reduce the availability of antigenic determinants for binding to MHC class II molecules. In addition, proteosome cleavage sites can be removed to reduce the availability of antigenic determinants for binding to MHC class II molecules.

  In preferred embodiments, non-peptide backbone elements are introduced into T cell epitopes to generate MHC class I or class II ligands with antagonistic properties. As used herein, “non-peptide backbone element” means a non-naturally occurring or synthetic amino acid as described above. As used herein, “antagonistic” means an epitope that is recognized by a T cell but interferes with its activation even in the presence of an activating epitope, ie, a homologous epitope. In general, antagonistic ones are derived from known epitopes by amino acid substitutions that introduce peptide side chain charges or large size modifications. Preferably, N-hydroxylated peptide derivatives or β-amino acids are introduced into T cell epitopes to generate antagonists (eg, Hin, S., et al., (1999) J. Immunology, 163: 2363-2367; Reinelt S., et al., (2001) J. Biol. Chemistry, 276: 24525-24530, which is hereby incorporated by reference in its entirety.).

  In other embodiments, T cell epitopes are introduced into the primary sequence library in regions that do not affect the natural folding and stability of the target protein. The T cell epitope is selected from a database of known MHC I binding peptides, MHC II binding peptides and T cell epitopes as described above. Preferably, at least one T cell epitope is added for each target protein. More preferably, at least two T cell epitopes are added per target protein. More preferably, 3 to 5 T cell epitopes are added for each target protein. In other embodiments, up to 16 T cell epitopes can be added per target protein (Stienekemeier, M., et al., (2001) Proc Natl Acad Sci USA, 98: 13872-13877; As a part of this specification). PDA® technology is used to ensure correct folding and stability of the modified target protein.

  In a preferred embodiment, a specific cleavage motif as defined above for antigen processing and presentation is added. Cleavage sites involved in the processing of antigenic determinants present for MHC class I and / or class II molecules are added as described above. Thus, proteolytic cleavage sites can be added to increase the availability of antigenic determinants for binding to MHC class II molecules. In addition, proteosome cleavage sites can be added to increase the availability of antigenic determinants for binding to MHC class I molecules.

  In a preferred embodiment, non-peptide backbone elements are introduced into T cell epitopes to produce MHC class I or class II ligands with agonist properties. As used herein, “agonist” means an epitope that is recognized by and activates T cells.

  In a preferred embodiment, the primary mutant sequence is screened for peptide fragments that are predicted to bind to the antibody. In a preferred embodiment, as described by Meyer et al. (Meyer, DL, et al. (2001), Protein Sci., 10: 491-503; Schwartz, HL., Et al. (1999) J. Mol Biol 287: 983-999; and Laroche, Y., et al., (2000) Blood, 96: 1425-1432), replacing potential B cell epitopes with smaller neutral residues. And reduce the immunogenicity of the sequence.

  In other embodiments, B cell epitopes are introduced into the primary sequence library or soluble target protein in a region that does not affect the natural folding and stability of the target protein. In particular, charged, aromatic or large hydrophobic residues are added on the surface of the target protein. Preferably, at least one B cell epitope is added for each target protein. More preferably, at least two B cell epitopes are added per target protein. More preferably, 3 to 5 B cell epitopes are added per target protein. In other embodiments, up to 16 B cell epitopes may be added per target protein (Stienekemeier, M., et al., (2001) Proc Natl Acad Sci USA, 98: 13872-13877; As part of the description). PDA® technology is used to ensure correct folding and stability of the modified target protein.

  In some embodiments, any combination of T cell epitopes, B cell epitopes, MHC class I and / or MHC class II binding motifs is converted into a primary target library or soluble target protein such as a Zn-alpha 2-glycoprotein. Introduce as above.

  In a preferred embodiment, at least one candidate mutant protein is identified in which at least one sequence capable of interacting with an MHC class I or class II molecule, TCR or BCR has been modified. Any method of identifying the actual or actual MHC, TCR or BCR sequence may be used in the present invention. Acceptable methods include computational or physical methods. Acceptable computer processing methods include OptiMer and EpiMer (Meister, GE., Et al. (1995) Vaccine, 6: 581-591); interactive stepwise discriminant analysis metal algorithm (Mallios, RR., (1999) Bioinformatics, 15: 432-439); and structure-based (Altuvia, Y., (1997) Human Immunology 58: 1-11) and combining evolutionary algorithms with artificial neural networks Prediction methods (Brusic, V., et al. (1998) Bioinformatics, 14: 121-130), virtual matrix (Sturniolo, T., et al. (1999) Nature Biotechnology, 17: 555-561) and BONSAI This includes the use of algorithms such as decision trees (Savoie, CJ., Et al (1999) Pac Symp Biocomput., 182-9). All references cited in this paragraph are hereby incorporated by reference in their entirety.

  Acceptable physical methods include high affinity binding assays (Hammer, J., et al. (1993) Proc. Natl. Acad. Sci. USA, 91: 4456-4460; Sarobe, P. et al. ( 1998) J. Clin. Invest., 102: 1239-1248), T cell proliferation and CTL assay (Hemmer, B., et al., (1998) J. Immunol., 160: 3631-3636), stabilization assay Competitive inhibition assays to purified MHC molecules or MHC with MHC, or sequencing followed by extraction (Brusic, V., et al., (1998) Nucleic Acids Res., 26: 368-371). All references cited in this paragraph are hereby incorporated by reference in their entirety.

  Once potential MHC, TCR or BCR sequences have been identified, these sequences are then altered by substitution of one or more amino acids as described below. Once the candidate mutant protein is changed, the protein is then tested to determine if its activity is similar to the target protein. A variant may retain sufficient activity or may usefully retain a sufficient percentage of activity.

  The mutant proteins and nucleic acids of the present invention can be distinguished from naturally occurring target proteins. As used herein, “naturally produced” or “wild type” or grammatical equivalent means an amino acid sequence or nucleotide sequence found in nature and includes allelic changes; that is, an amino acid sequence or nucleotide sequence is usually It has not been intentionally changed. Thus, “non-naturally occurring” or “synthetic” or “recombinant” or grammatical equivalents herein means an amino acid sequence or nucleotide sequence that is not found in nature; Are usually deliberately changed. Once a recombinant nucleic acid has been created and reintroduced into a host cell or organism, it is replicated non-recombinantly, ie, using the cellular machinery of the host cell in vivo rather than in vitro manipulation. It will be appreciated that once recombinantly produced nucleic acid is subsequently replicated non-recombinantly, it is still considered recombinant for the purposes of the present invention. Thus, the mutant proteins and nucleic acids of the invention are non-naturally occurring; that is, they do not exist in nature.

  Thus, in a preferred embodiment, the mutant protein has an amino acid sequence that differs from the target sequence by at least 1-5% of the residues. That is, the mutant protein of the present invention is about 97-99% or less identical to the target amino acid sequence. Accordingly, the protein preferably has a global homology of the protein sequence to the target sequence of about 99% or less, more preferably about 98% or less, even more preferably about 97% or less, and more preferably about 95% or less. Then, it is a “candidate mutant protein”. In some embodiments, the homology is as low as about 75-80%.

  Homology in this context means sequence similarity or identity, with identity being preferred. As is known in the art, a number of different programs can be used to identify whether a protein (or a nucleic acid as discussed below) has sequence identity or similarity with a known sequence. Sequence identity and / or similarity is measured using standard techniques well known in the art, including the local sequence identity algorithm of Smith & Waterman, Adv. Appl. Math., 2: 482 (1981), Needleman & Wunsch, J. Mol. Biol., 48: 443 (1970) sequence identity alignment, Pearson & Lipman, Proc. Natl. Acad. Sci. USA, 85: 2444 (1988) Computer implementation of the algorithm (Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Drive, Madison, WI, GAP, BESTFIT, FASTA and TFASTA), Devereux et al., Nucl. Acid Res., 12: 387- Including, but not limited to, the Best Fit sequence program described in 395 (1984), preferably using default settings or by assay. Preferably, FstDB calculates percent identity based on the following parameters: mismatch penalty 1; gap penalty 1; gap size penalty 0.33; joining penalty 30, "Current Methods in Sequence Comparison and Analysis," Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc. All references are incorporated herein by reference.

  An example of a useful algorithm is PILEUP. PILEUP creates multiple sequence alignments from related sequences using progressive pair alignment. It can also draw a tree showing the clustering relationships used to create the alignment. PILEUP uses a simplified version of the progressive alignment method of Feng & Doolittle, J. Mol. Evol. 35: 351-360 (1987); this method is Higgins & Sharp, CABIOS 5: 151-153 (1989). Similar to the method described. Useful PILEUP parameters include a default gap weight 3.00, a default gap length weight 0.10, and weighted end gaps.

  Another example of a useful algorithm is Altschul et al., J. Mol. Biol. 215, 403-410, (1990); Altschul et al., Nucleic Acids Res. 25: 3389-3402 (1997); Karlin et al., Proc. Natl. Acad. Sci. USA 90: 5873-5787 (1993). A particularly useful BLAST program is the WU-BLAST-2 program obtained from Altschul et al., Methods in Enzymology, 266: 460-480 (1996); http: //blast.wustl/edu/blast/README.html] It is. WU-BLAST-2 uses several search parameters, most of which are set to default values. The adjustable parameters are set with the following values: overlap span = 1, overlap fraction = 0.125, word threshold (T) = 11. HSP S and HSP S2 parameters are dynamic numbers, established by the program itself depending on the composition of the particular sequence and the composition of the particular database searching for the sequence of interest, but the values increase sensitivity Can be adjusted as follows.

An additional useful algorithm is gapped BLAST as reported in Altschul et al., Nucl. Acids Res., 25: 3389-3402. gapped BLAST uses BLOSUM-62 alternative score; wherein threshold T parameter set to 9; the X _u 16; 2 hits method by causing a free extension of the gap; imposes costs 10 + k to the gap length k Set; _Xg is set to 40 in the database search phase and 67 in the output phase of the algorithm. A gap door alignment starts with a score corresponding to about 22 bits.

  The percent amino acid sequence identity value is determined by dividing the number of matching identical residues by the total number of residues of the “longer” sequence in the aligned region. A “longer” sequence is one that has the most actual residues in the aligned region (ignoring gaps introduced by WU-Blast-2 to maximize the alignment score).

  Similarly, “percent (%) nucleic acid sequence identity” with respect to the coding sequence of a polypeptide identified in the present invention is the percentage of nucleotide residues in the candidate sequence that are identical to the nucleotide residues of the coding sequence of the target protein. Define as A preferred method is to use WU-BLAST-2's BLASTN module with default parameters and overlap span and overlap fraction set to 1 and 0.125 respectively.

  Alignment may include introducing gaps into the alignment sequence. In addition, for sequences containing more or fewer amino acids than the target protein, it is understood that in certain embodiments, the percent sequence identity is determined based on the number of identical amino acids relative to the total number of amino acids. In calculating percent identity, relative weights are not assigned to the expression of various sequence changes such as insertions, deletions, substitutions, and the like.

  In one embodiment, only identity is given a positive score (+1) and a value of “0” is assigned to any form of sequence variation, including gaps. This eliminates the need for weighted scales or parameters as described below for sequence similarity calculations. For example, percent amino acid sequence identity is calculated by dividing the number of matching identical residues by the total number of residues of the “shorter” sequence in the aligned region and multiplying by 100. A “longer” sequence is one that has the most actual residues in the aligned region.

  Therefore, the mutant protein of the present invention may be shorter or longer than the target protein. A part or fragment of the target sequence is included in the definition of the mutant protein. A fragment of a mutant protein is considered a mutant alpha protein if it a) shares at least one antigenic epitope; b) has at least the indicated homology; c) and preferably exhibits the biological activity of the target protein. .

  As outlined in more detail below, in a preferred embodiment, the candidate mutein contains further amino acid changes compared to the target protein than those outlined herein. In addition, as outlined herein, any of the changes depicted herein may be combined in any way to form additional novel mutant proteins.

  In addition, by adding other sequences such as purification tags, fusion sequences, as described, for example, in USS 09 / 798,789, which is hereby incorporated by reference. Can create candidate mutant proteins that are longer than the target protein. For example, the mutant proteins of the invention may be fused with other therapeutic proteins or other proteins such as Fc or serum albumin for pharmacokinetic purposes. See, eg, US Pat. Nos. 5,766,883 and 5,876,969, both of which are hereby incorporated by reference.

  Mutant proteins that contain variable residues at the core, surface, and boundary residues are also included within the invention.   In a preferred embodiment, the mutant protein of the invention is a human conformer. As used herein, “conformer” refers to a protein that has substantially the same main-chain 3D structure, but with significant differences in amino acid side chains. That is, the muteins of the present invention define a set of conformers where all the proteins of the set share the backbone structure and still have at least 1-3-5% different sequences. The three-dimensional backbone structure of the mutant protein therefore substantially corresponds to the three-dimensional backbone structure of the human target protein.

  “Main chain” in this context means non-side chain atoms, ie nitrogen, carbon and oxygen of the carbonyl group, and the α carbon, and hydrogen bonded to the nitrogen and α carbon. In view of the conformer, the protein must have a main chain atom of no more than 2 か ら from the human target protein structure, preferably no more than 1.5 好 ま し く, particularly preferably no more than 1 Å. In general, these distances are determined in two ways. In one embodiment, each potential conformer is crystallized and its three-dimensional structure is determined. Alternatively, since the former is technically difficult, each possible conformer sequence is executed in the PDA® program to determine whether it is a conformer.

  A candidate mutant protein may be identified as being encoded by a candidate mutant nucleic acid. In the case of nucleic acids, the overall homology of nucleic acid sequences is the same criteria as amino acid homology, but takes into account the degeneracy of the genetic code and the codon bias of various organisms. Thus, nucleic acid sequence homology may be lower or higher than that of protein sequences, with lower homology being preferred.

  In a preferred embodiment, the candidate mutant nucleic acid encodes a candidate mutant protein. To those skilled in the art As will be appreciated, all encode the mutant proteins of the present invention due to the degeneracy of the genetic code. A large number of nucleic acids can be generated. Therefore, if a specific amino acid sequence is identified, those skilled in the art Is a method that does not change the amino acid sequence of the mutated protein, simply one or more codes. Any number of different nucleic acids can be made by altering the sequence of the nucleic acids.

In certain embodiments, nucleic acid homology is determined by hybridization studies. Conditions with high stringency are known in the art; for example, Maniatis et al., Molecular Cloning: A Laboratory Manual, 2d Edition, 1989 and Short Protocols in Molecular Biology, which are incorporated herein by reference. See ed. Ausubel, et al. Stringency conditions are sequence-dependent and will vary for different environments. Longer sequences hybridize specifically at higher temperatures. Detailed guidance on nucleic acid hybridization can be found in Tijssen, Techniques in Biochemistry and Molecular Biology--Hybridization with Nucleic Acid Probes, "Overview of principles of hybridization and the strategy of nucleic acid assays" (1993). Generally, stringency conditions are selected to be about 5-10 ° C. lower than the thermal melting point (T _m ) for the specific sequence at a defined ionic strength and pH. T _m is the temperature at which 50% of the probe complementary to the target will hybridize to the target sequence in equilibrium (at a defined ionic strength, pH and nucleic acid concentration) (because the target sequence is present in excess) _{In m} , 50% of the probes are occupied in equilibrium). Stringency conditions are such that the salt concentration is below about 1.0 M sodium ion, typically about 0.01 to 1.0 M sodium ion (or other salt) concentration, pH 7.0 to 8.3, The temperature is as low as about 30 ° C. for short probes (eg, 10-50 nucleotides) and as low as about 60 ° C. for long probes (eg, 50 nucleotides or more). Stringency conditions may be achieved by the addition of destabilizing substances such as formamide.

  In other embodiments, lower stringency hybridization conditions are used; for example, mild or low stringency conditions may be used, as known in the art; Maniatis and Ausubel (supra) and See Tijssen (supra).

  The candidate mutant protein and nucleic acid of the present invention are recombinant. As used herein, “nucleic acid” may refer to DNA or RNA, or a molecule containing both deoxy- and ribonucleotides. Nucleic acids include genomic DNA, cDNA and oligonucleotides, including sense and antisense nucleic acids. Such nucleic acids may also contain changes in the ribose-phosphate backbone to increase the stability and half-life of such molecules in a physiological environment.

  Nucleic acids may be double stranded, single stranded, or contain portions of both double stranded or single stranded sequence. As will be appreciated by those skilled in the art, drawing one strand (“Watson”) defines the sequence of the other strand (“click”), so the sequence shown in FIG. Including. As used herein, the term “recombinant nucleic acid” generally refers to a nucleic acid originally formed in vitro in a form that is not normally found in nature by manipulating the nucleic acid with an endonuclease. Therefore, both candidate mutant nucleic acids isolated in linear form and expression vectors formed in vitro by ligation of DNA molecules that are not normally bound are both considered recombinant for the purposes of the present invention. . Once a recombinant nucleic acid has been created and reintroduced into a host cell or organism, it is understood to replicate non-recombinant, ie, using the in vivo cellular machinery of the host cell rather than in vitro manipulation; Once such nucleic acid is produced recombinantly, it can still be replicated non-recombinantly and still be considered recombinant for the purposes of the present invention.

  Similarly, a “recombinant protein” is a protein made using recombinant techniques, ie through expression of a recombinant nucleic acid as described above. A recombinant protein is distinguished from naturally occurring protein by at least one or more properties. For example, the protein can be isolated or purified from some or all of the protein or compound normally associated in the wild-type host, and thus can be substantially pure. For example, an isolated protein is preferably at least about 0.5%, more preferably at least 5% of the total protein weight in a given sample without at least some of the substances normally associated in the natural state. % Make up. A substantially pure protein comprises at least about 75%, preferably at least about 80%, and particularly preferably at least about 90% of the total protein weight. This definition includes producing a candidate mutein from one organism in a different organism or host cell. Alternatively, the protein can be made at a significantly higher concentration than would normally be seen by using an inducible promoter or a high expression promoter so that the protein is made at an increased concentration level. Furthermore, as discussed below, all of the mutant proteins outlined herein are in a form not normally found in nature because they contain amino acid substitutions, insertions and deletions, preferably substitutions.

  Amino acid sequence variants of the candidate variant sequences outlined herein are also included in the definition of candidate variant proteins of the invention. That is, the candidate mutant protein can contain additional variable positions compared to the target protein. These variants correspond to one or more of three classes: substitutional, insertional or deletional variants. These mutants typically encode the mutant by site-directed mutagenesis of nucleotides in the DNA encoding the candidate mutant protein using cassette or PCR mutagenesis or other techniques well known in the art. And is then prepared by expressing the DNA in recombinant cultured cells as outlined above. However, candidate mutant protein fragments having up to about 100-150 residues can be prepared by in vitro synthesis using established techniques. Amino acid sequence variants are characterized by the property that changes are predetermined, which distinguishes these variants from naturally occurring allelic or interspecies variants of the amino acid sequence of the candidate mutant protein. Is done. Variants typically exhibit the same qualitative biological activity as naturally occurring analogs, but variants with altered characteristics can also be selected as outlined in more detail below. .

  The site or region for introducing an amino acid sequence variation is determined in advance, but the mutation per se need not be determined in advance. For example, in order to optimize the performance of mutations at a given site, random mutagenesis may occur at the target codon or region, and the expressed mutant proteins may be screened for the optimal combination of desired activities. Techniques for creating substitution mutations at predetermined sites in DNA having a known sequence are well known, for example, mutagenesis by M13 primer and mutagenesis by PCR.

  Amino acid substitutions are typically single residues; insertions are usually made in units of about 1-20 amino acids, although fairly large insertions may be tolerated. Deletions can be larger, but range from about 1 to about 20 residues.

Substitutions, deletions, insertions or any combination thereof may be used to arrive at the final derivative. In general, these changes are made on a small number of amino acids to minimize molecular modifications. However, larger changes can be tolerated in certain situations. Substitutions are generally made according to Chart 1 if minor modifications are desired for the characteristics of the mutated protein.

  Substantial changes in function or immunological identity are made by selecting substitutions that are less conservative than those shown in Chart I. For example, more influential substitutions can be made: they can be polypeptide backbone structures of the region to be modified, such as alpha-helix or beta-sheet structures; molecular charge or hydrophobicity at the target site; or side chain It is a size. The substitutions that are generally expected to produce the greatest changes in the properties of the polypeptide are: (a) a hydrophilic residue such as seryl or threonyl, a hydrophobic residue such as leucyl, isoleucyl, phenylalanyl, valyl, or Substituting alanyl (or vice versa), (b) replacing cysteine or proline with any other residue (or vice versa), (c) a positively charged side chain such as lysyl, arginyl, or histidyl Is replaced with a negatively charged side chain such as glutamyl, aspartyl (or vice versa), or (d) a residue with a large side chain such as phenylalanine is replaced with a residue without a side chain such as glycine Yes (or vice versa).

  Typically, mutants exhibit the same biological activity qualitatively, but the immune response may be altered from that of the original candidate mutant protein, if desired. Alternatively, the mutant can be designed such that the biological activity of the candidate mutant protein is altered. For example, glycosylation sites can be modified or removed. Similarly, biological functions can be altered.

  In addition, in some embodiments, it is desirable to obtain candidate mutant proteins with altered immunogenicity that are more stable than the target protein. Preferably, it is desirable to obtain a protein exhibiting oxidative stability, alkali stability, and thermal stability.

  The change in oxidative stability is manifested by an increase in the activity of the mutant protein by at least about 20%, more preferably at least about 50% compared to that of the wild-type protein when exposed to various oxidizing conditions. Is done. Oxidative stability is measured by known methods.

  The change in alkali stability indicates that the half-life of the mutant protein activity is increased or decreased by at least about 5% or more (preferably compared to that of the wild type protein) when exposed to increasing or decreasing pH conditions. Is increased). In general, alkali stability is measured by known methods.

  Changes in thermal stability indicate that the half-life of the mutant protein activity is increased or decreased by at least about 5% or more compared to that of the wild-type protein when exposed to relatively high temperatures and neutral pH ( (Preferably increase). In general, thermal stability is measured by known methods.

  Candidate mutant proteins and nucleic acids of the invention can be made by a number of methods. Individual nucleic acids and proteins can be made as known in the art and as outlined below. Alternatively, a library of candidate mutant proteins can be created for testing.

  In a preferred embodiment, the candidate mutant protein library is generated from a probability distribution table. Produce probability distribution tables, including PDA® techniques, sequence alignment, force field calculations such as self-consistent mean field (SCMF) calculations, etc. as outlined herein There are various ways to make it happen. In addition, a probability distribution can be used to generate entropy core information for each position as a measure of the mutation frequency observed in the library.

  In this embodiment, the frequency of each amino acid residue at each variable position in the list is identified. Each frequency can be a threshold when all mutation frequencies below the cutoff are set to zero. This cut-off is preferably about 1%, 2%, 5%, 10% or 20%, with about 10% being particularly preferred. These frequencies are then incorporated into the candidate mutant protein library. That is, as described above, these variable positions are collected and any possible combinations are generated, but the amino acid residues that “fill” the candidate mutant protein library are utilized on a frequency basis. Thus, in a non-frequency-based candidate mutant protein library, one variable position with five possible residues is about 20% of proteins containing that variable position with a potential first residue. The second would have 20%, and so on. However, in a frequency-based candidate mutein library, one variable position with 5 possible residues, with a frequency of about 10%, 15%, 25%, 30% and 20%, respectively, is likely Will have 10% of the protein containing that variable position with the first residue, 15% of the protein with the second residue, 25% of the protein with the second residue, and so on. As will be apparent to those skilled in the art, the actual frequency may vary depending on the method actually used for protein production; for example, the exact frequency may be possible when the protein is synthesized. However, when using the frequency-based primer system outlined below, the actual frequency at each position varies as described below.

  As will be appreciated by those skilled in the art and as outlined herein, probability distribution tables can be generated in a variety of ways. In addition to the methods outlined herein, a self-consistent mean force field (SCMF) method can be used in the direct generation of probability tables. SCMF is a deterministic computerized method of calculating energy using a description of the mean force field of rotamer interactions. A probability table formed in this way can be used to create a candidate mutant protein library as described herein. SCMF can be used in three ways: list amino acid and rotamer frequency of each amino acid for each position; probability is determined directly from SCMF (Delarue et la. Pac., Which is incorporated herein by reference). Symp. Biocomput. 109-21 (1997)). In addition, highly variable positions and non-variable positions can be identified.

  Alternatively, another method is used to determine which sequence to jump to during the sequence space search; SCMF is used to obtain the correct energy for that sequence; this energy is then ranked And used to create a ranked list of sequences (similar to a Monte Carlo sequence list). A probability table showing the frequency of amino acids at each position is then calculated from this list (Koehl et al., J. Mol. Biol. 239: 249 (1994); Koehl et al., Nat. Struc. Biol. 2: 163 (1995); Koehl et al., Curr. Opin.Struct. Biol. 6: 222 (1996); Koehl et al., J. Mol. Bio. 293: 1183 (1999); Koehl et al., J. Mol. Biol. 293: 1161 (1999); Lee J. Mol. Biol. 236: 918 (1994); and Vasquez Biopolymers 36: 53-70 (1995); all of which are specifically incorporated herein by reference. And) .

  Similar methods include OPLS-AA (Jorgensen, et al., J. Am. Chem. Soc. (1996), v 118, pp 11225-11236; Jorgensen, WL; BOSS, Version 4.1; Yale University: New Haven. , CT (1999)); OPLS (Jorgensen, et al., J. Am. Chem. Soc. (1988), v 110, pp 1657ff; Jorgensen, et al., J Am. Chem. Soc. (1990), v 112, pp 4768ff); UNRES (United Residue Forcefield; Liwo, et al., Protein Science (1993), v 2, pp 1697-1714; Liwo, et al., Protein Science (1993), v 2, pp1715- 1731; Liwo, et al., J. Comp. Chem. (1997), v 18, pp 849-873; Liwo, et al., J. Comp. Chem. (1997), v 18, pp 874-884; Liwo, et al., J. Comp. Chem. (1998), v 19, pp 259-276); Forcefield for Protein Structure Prediction (Liwo, et al., Proc. Natl. Acad. Sci. USA (1999), v 96, pp 5482-5485); ECEPP / 3 (Liwo et al., J Protein Chem 1994 May; 13 (4): 375-80); AMBER 1.1 force field (Weiner, et al., J. Am. Chem) Soc. V 106, pp 765-784); AMBER 3.0 force field (UC Singh et al., Proc. Natl. Acad. Sci. USA. 82: 755-759); CHARMM and CHARMM22 (Brooks, et al., J. Comp. Chem. v4, pp 187-217); cvff3.0 (Dauber-Osguthorpe, et al., (1988) Proteins: Structure, Function and Genetics, v 4, pp 31-47); Including, but not limited to, CFF91 (Maple, et al., J. Comp. Chem. V 15, 162-182); and DISCOVER (cvff and cff91) and AMBER force fields are INSIGHT molecules Used in the modeling package (Biosym / MSI, San Diego California) and HARM is a QUANTA molecular modeling package (Biosym / MSI, San Diego California) (all references are hereby incorporated by reference) used.

  In addition, as outlined herein, a preferred method of generating a probability distribution table is through the use of a sequence alignment program. In addition, the probability table is obtained by a combination of sequence alignment and computational approach. For example, amino acids found in homologous sequence alignments can be added to the results of computer processing. Preferably, wild-type amino acids matching the probability table can be added if it is not found by computer processing.

  As will be apparent, candidate mutant protein libraries created by recombination of variable positions and / or residues at variable positions are not in rank order. In some implementations, the entire list need only be created and tested. Alternatively, in a preferred embodiment, this secondary library is also in a ranked list format. This can be done for several reasons, including that the size of the secondary library is still too large to generate experimentally, or for predictive purposes. This can be done in several ways. In one embodiment, the secondary library is ranked or filtered using the PDA® scoring or filtering function that ranks library members. Alternatively, statistical techniques can be used. For example, this secondary library may be ranked or filtered by frequency score; that is, it may be possible to rank proteins that contain the majority of high frequency residues, etc. This can also be done by adding or multiplying the frequency at each variable position to generate a numerical score. Similarly, various positions in the secondary library can be weighted and then the proteins can be scored; for example, those containing certain residues can be arbitrarily ranked or filtered.

  In a preferred embodiment, different protein members of the candidate mutant library can be chemically synthesized. This is particularly useful when the designed protein is short, preferably 150 amino acids or less, more preferably 100 amino acids or less, particularly preferably 50 amino acids or less, but as known in the art, Longer proteins can be prepared chemically or enzymatically. See, for example, Wilken et al, Curr. Opin. Biotevhnol. 9: 412-26 (1998), which is incorporated herein by reference.

  In a preferred embodiment, especially for longer proteins or proteins for which large samples are desired, candidate mutant sequences such as DNA that encode member sequences and can be cloned into host cells and optionally expressed and analyzed. Used to generate nucleic acid. Thus, nucleic acids, particularly DNA, encoding each member protein sequence can be created. This is done using known methods. The selection of codons, suitable expression vectors, and suitable host cells will vary depending on a number of factors and can be easily optimized as needed.

  In a preferred embodiment, multiple PCR reactions using pooled oligonucleotides are performed as generally shown in FIG. In this embodiment, overlapping oligonucleotides corresponding to the full-length gene are synthesized. These oligonucleotides may also represent all of the different amino acids at each mutation position or subset.

  In a preferred embodiment, these oligonucleotides are pooled in equal proportions and multiple PCR reactions are performed to create a full-length sequence containing the combination of mutations defined in the secondary library. In addition, this can be done using error-prone PCR (error-prone PCR) methods.

  In a preferred embodiment, the various oligonucleotides are added in relative amounts corresponding to the probability distribution table. Thus, multiple PCR reactions result in a full-length sequence having the desired mutation combination in the desired proportion.

  The total number of oligonucleotides required is a function of the number of positions to be mutated and the number of mutations considered at those positions. (Number of oligos for invariant positions) + M1 + M2 + M3 +... Mn = (total number of necessary oligos), where Mn is the number of mutations considered at position n of the sequence.

  In a preferred embodiment, each overlapping oligonucleotide contains only one position to be mutated; in another embodiment, the mutation positions are too close to each other to make this possible, and multiple oligonucleotides per oligonucleotide Mutations are used to complete all possible recombination. That is, each oligo can contain a single position to be mutated or one or more codons to be mutated. The multiple positions to be mutated must be close together in the sequence to prevent the oligo length from becoming impractical. By including or excluding an oligonucleotide encoding a particular combination of mutations for multiple mutation positions on an oligonucleotide, the combination can be included or excluded in the library. For example, as discussed herein, there may be a correlation between the variable positions; that is, if position X is a certain residue, position Y must be a certain residue (or Must not be). These sets of variable positions are sometimes referred to herein as “clusters”. When a cluster contains residues in close proximity to each other and can therefore be present on a single nucleotide primer, the cluster can be set to a “good” correlation to eliminate bad combinations that can reduce the effectiveness of the library. However, if the cluster residues are separated in the sequence and will therefore be present on another oligonucleotide for synthesis, the residues are set to “good” correlation or variable residues. It would be desirable to eliminate it entirely as a group.

  In another embodiment, the library is created in several stages so that cluster mutations appear exclusively. This method, i.e. identifying a mutant cluster and placing it on the same oligonucleotide or removing it from the library or generating the library in several steps while preserving the cluster The library can be made very rich in properly folded proteins. Cluster identification can be done, for example, using known pattern recognition methods, comparing mutation frequencies, or analyzing the energy of experimentally generated sequences (eg, if the interaction energy is high, the position is correlated) It can be implemented in a number of ways such as use. These correlations are both positional correlations (eg, positions 1 and 2 always change together or never change together), even if they are sequence correlations (eg, if there is a residue A at position 1, it will always be at position 2). Residue B).

  Pattern discovery in Biomolecular Data: Tools, Techniques, and Applications; edited by Jason TL Wang, Bruce A. Shapiro, Dennis Shasha.New York: Oxford University, 1999; Andrews, Harry C. Introduction to mathematical techniques in pattern recognition; New York , Wiley-lnterscience [1972]; Applications of Pattern Recognition; Editor, KS Fu. Boca Raton, Fla. CRC Press, 1982; Genetic Algorithms for Pattern Recognition; edited by Sankar K. Pal, Paul P. Wang. Boca Raton: CRC Press, c1996; Pandya, Abhijit S., Pattern recognition with neural networks in C ++ / Abhijit S. Pandya, Robert B. Macy. Boca Raton, Fla .: CRC Press, 1996; Handbook of pattern recognition & computer vision / edited by CH Chen, LF Pau, PSP Wang. 2nd ed. Singapore; River Edge, NJ: World Scientific, c1999; Friedman, Introduction to Pattern Recognition: Statistical, Structural, Neural, and Fuzy Logic Approaches; River Edge, NJ: World Scientific, c1999 , Series title: Series in machine perception and artificial intelligence; vol. 32, all of which are hereby incorporated by reference. In addition, the program used to search for common motifs can be used as well.

  In addition, correlation and mixing can be fixed or modified by modifying the design of the oligonucleotide, that is, by defining where the oligonucleotide (primer) starts and stops (eg where to “cleave” the sequence). Can be optimized. Oligo start and stop sites can be set to maximize the number of clusters that appear in a single oligonucleotide, thereby enriching the library with higher-scoring sequences. Computerized modeling of various oligonucleotide start and stop site options, ranked according to the number of clusters represented on a single oligo or according to the percentage of sequences generated that match the predicted sequence library Can be attached or filtered.

  The total number of oligonucleotides required increases when multiple mutable positions are encoded by a single oligonucleotide. The annealed region remains constant, i.e. has the sequence of the reference sequence.

  Oligonucleotides with inserted or deleted codons can be used to create libraries that express proteins of different lengths. In particular, computerized sequence screening for insertions or deletions can yield secondary libraries that define proteins of different lengths, which are pooled libraries of oligonucleotides of various lengths Can be expressed.

  In a preferred embodiment, the secondary library is performed by mixing families (eg, a set of variants); that is, with or without error-prone PCR (when using a ranked list). In addition, several sets of superordinate sequences can be mixed. “Mixing” in this context means recombination of related sequences, which is generally performed randomly. US Pat. Nos. 5,830,721; 5,811,238; 5,605,793; 5,837,458 and PCT US / 19256 (all of which are incorporated by reference) “Mixed” as defined and exemplified in the specification of which is hereby incorporated by reference). This set of sequences can be an artificial set; for example, derived from a probability table (eg, generated using SCMF) or a Monte Carlo set. Similarly, a “family” can be the top 10 sequences and the bottom 10 sequences, the top 100 sequences, and the like. This can also be done by error-prone PCR.

  Accordingly, in a preferred embodiment, in silico mixing is performed using the computer processing methods described herein. That is, starting with either two libraries or two sequences, random sequence recombination can be generated and evaluated.

  In a preferred embodiment, error-prone PCR is performed to generate a secondary library. See US Pat. Nos. 5,605,793, 5,811,238, and 5,830,721, all of which are hereby incorporated by reference. This can be done on optimal sequences or members of the library, or on other artificial sets or families. In this embodiment, a gene for the optimal sequence found in the computerized screening of the primary library can be synthesized. Error-prone PCR is then performed on the optimal sequence gene in the presence of oligonucleotides (bias oligonucleotides) that encode mutations at the mutation positions in the secondary library. The addition of oligonucleotides creates a bias that favors the introduction of mutations in the secondary library. Alternatively, only a set of oligonucleotides for certain mutations can be used to bias the library.

  In a preferred embodiment, error-prone PCR gene mixing is performed on optimally sequenced genes in the presence of biased oligonucleotides, reflecting the percentage of each mutation found in the secondary library. A DNA sequence library can be created. The selection of biased oligonucleotides can be done in various ways; they can be selected based on their frequency, i.e., they can use oligonucleotides that encode positions with high mutation frequency; To increase, the oligonucleotide containing the most susceptible position can be used; if the secondary library is ranked, some of the top score positions can be used to generate a biased oligonucleotide. A random location can be selected; a small number of top scores and a small number of low scores can be selected; and so on. What is important is to generate a new sequence based on the preferred mutation position and sequence.

  In a preferred embodiment, PCR with a wild type gene or target gene can be used, as outlined in FIG. In this embodiment, a starting gene is used; although it is not a requirement, generally the gene is a wild type gene. It can be a gene that encodes a global optimal sequence or any other sequence in the list. In this embodiment, oligonucleotides corresponding to the mutation positions and containing various amino acids of the secondary library are used. As is well known in the art, PCR is performed using PCR primers at the ends. This has two advantages; first, fewer oligonucleotides and fewer errors. In addition, when using a wild-type gene, there is an experimental advantage that there is no need to synthesize.

  In addition, there are several other techniques that can be used, as illustrated in FIGS. 2-5. In a preferred embodiment, PCR product ligation is performed.

  In a preferred embodiment, various additional steps can be performed on one or more candidate mutation secondary libraries; for example, further computer processing can be performed and the candidate mutation secondary libraries are recombined. Cut-offs from different candidate mutation secondary libraries can be combined. In a preferred embodiment, the candidate mutant secondary library can be re-engineered by computer processing to form additional secondary libraries (sometimes referred to herein as “tertiary libraries”). For example, any candidate mutant secondary library sequence can be selected for the second round of PDA® by freezing or fixing some or all of the altered positions in the first secondary library. Alternatively, only the changes found in the last probability distribution table are allowed. Alternatively, the stringency of the probability table may be altered by either increasing or decreasing the included cutoff. Similarly, a candidate mutation secondary library can be experimentally recombined after the first round; for example, the best gene / gene group from the first screen is taken and gene assembly is performed again (techniques outlined below, multiple Using PCR, error-prone PCR, mixing, etc.). Alternatively, a fragment from one or more good gene (s) to change the probability at several positions. This biases the search for areas of sequence space found in the first round of computational and experimental screening.

  In a preferred embodiment, a tertiary library can be generated from combining candidate mutant secondary libraries. For example, a probability distribution table is generated from a candidate mutation secondary library and recombined computationally or experimentally as outlined herein. PDA® technology candidate mutation secondary libraries may be combined with sequence alignment libraries, either recombined (again, computationally or experimentally) or simply combined with a cut-off from each A new tertiary library may be created. Upper sequences from several libraries can be recombined. Primary and secondary libraries can be combined in a similar manner. Sequences from the top of a library can be combined with sequences from the bottom of the library to sample a wider sequence space, or only sequences away from the top of the library can be combined. Candidate mutant secondary libraries that have analyzed various parts of the protein can be combined into a tertiary library that handles the combined parts of the protein.

  In a preferred embodiment, correlations in candidate mutant secondary libraries can be used to generate tertiary libraries. That is, residues at the first variable position can be correlated to residues at the second variable position (or similarly correlated to residues at further positions). For example, two variable positions can interact sterically or electrostatically, such that if the first residue is X, the second base must be Y. This can be either positive or negative correlation.

  A variety of expression vectors are generated using the nucleic acids of the invention encoding candidate mutant library members. Expression vectors can be either self-replicating extrachromosomal vectors or vectors that integrate into the host genome. In general, these expression vectors contain transcriptional and translational regulatory nucleic acids operably linked to nucleic acids encoding library proteins. The term “control sequence” refers to a DNA sequence necessary for the expression of a coding sequence operably linked in a particular host organism. The control sequences that are suitable for prokaryotes, for example, include a promoter, optionally an operator sequence, and a ribosome binding site. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

  A nucleic acid is “operably linked” when it is placed in a situation where it is in a functional relationship with another nucleic acid sequence. For example, DNA related to a precursor sequence or secretory leader is operably linked to the DNA of the polypeptide when expressed as a precursor protein involved in the secretion of the polypeptide; a promoter or enhancer is responsible for transcription of the sequence. If it affects, it is operably linked to the coding sequence; or the ribosome binding site is operably linked to the coding sequence if it is positioned to facilitate translation. In general, “operably linked” means that the DNA sequences to be bound are contiguous and, in the case of a secretory leader, contiguous and in the reading phase. means. However, enhancers do not have to be contiguous. Binding is achieved by ligation at convenient restriction sites. If such sites do not exist, synthetic oligonucleotide adapters or linkers are used according to conventional methods. Transcriptional and translational regulatory nucleic acids are generally appropriate for the host cell used for expression of the library protein, as will be appreciated by those skilled in the art; for example, transcriptional and translational regulatory nucleic acid sequences from Bacillus are preferably Bacillus. Used for the expression of library proteins. Various types of suitable expression vectors, and suitable regulatory sequences are known in the art for a variety of host cells.

  In general, transcriptional and translational regulatory sequences can include, but are not limited to, promoter sequences, ribosome binding sites, transcriptional start and stop sequences, translational start and stop sequences, and enhancer or activator sequences. In a preferred embodiment, the regulatory sequences include a promoter and transcription start and stop sequences.

  Promoter sequences include constitutive or inducible promoter sequences. The promoter can be a naturally produced promoter, a hybrid or a synthetic promoter. Hybrid promoters that combine one or more promoter elements are also known in the art and are useful in the present invention.

  In addition, the expression vector may contain additional elements. For example, an expression vector may have two replication systems and can be maintained in two organisms, such as mammalian or insect cells for expression and a prokaryotic host for cloning and amplification. Furthermore, in the case of an integrative expression vector, the expression vector comprises at least one sequence homologous to the host cell genome, and preferably two homologous sequences that abut both ends of the expression construct. The integrating vector can be directed to a specific locus in the host cell by selecting appropriate homologous sequences for inclusion in the vector. Integration vector constructs and appropriate selection and screening protocols are well known in the art, e.g., Mansouret all, Cell, 51: 503 (1988) and Murray, Gene Transfer and Expression Protocols, Methods in Molecular Biology, Vol. 7 ( Clifton: Humana Press, 1991).

  Further, in a preferred embodiment, the expression vector contains a selection gene that allows for selection of transformed host cells containing the expression vector, and cells that do not contain the vector are generally common, especially for mammalian cells. The stability of the vector is ensured. Selection genes are well known in the art and will vary with the host cell used. As used herein, “selection gene” means any gene that encodes a gene product that confers resistance to a selection agent. Suitable selective agents include, but are not limited to, neomycin (or its analog G418), blasticidin S, histinidol D, bereomycin, puromycin, hygromycin B and other drugs.

  In a preferred embodiment, the expression vector contains RNA splicing sequences upstream or downstream of the expressed gene to increase gene expression levels (Barret et al., Nucleic Acids Res. 1991; Groos et al., Mol. Cell. Biol. 1987; and Budiman et al., Mol. Cell. Biol. 1988).

  A preferred expression vector system is Mann et al., Cell, 33: 153-9 (1993); Pear et al., Proc. Natl. Acad. Sci. USA, 90 (18): 8392-6 (1993); Kitamura et al., Proc. Natl. Acad. Sci. USA, 92: 9146-50 (1995); Kinsella et al., Human Gene Therapy, 7: 1405-13; Hofmann et al., Proc. Natl. Acad. Sci. USA, 93: 5185-90; Choate et al., Human Gene Therapy, 7: 2247 (1996); generally described in PCT / US97 / 01019 and PCT / US97 / 01048, and references cited therein. Retroviral vector systems, all of which are incorporated herein by reference.

  The candidate mutant library protein of the invention is suitable for inducing or causing expression of a library protein in a host cell transformed with an expression vector containing a nucleic acid, preferably a nucleic acid encoding the library protein. Produced under culture conditions. Appropriate conditions for expression of the candidate mutant library protein will vary with the choice of the expression vector and the host cell, and will be readily ascertained by one skilled in the art through routine experimentation. For example, using constitutive promoters in expression vectors requires optimization of host cell growth and proliferation, while using inducible promoters requires growth conditions suitable for induction. Furthermore, in some embodiments, the timing of recovery is important. For example, the baculovirus system used in insect cell expression is a lytic virus, so the choice of recovery time is critical to product yield.

  As will be apparent to those skilled in the art, the cell types used in the present invention vary widely. In principle, a wide variety of suitable host cells can be used, including yeast, bacteria, archaea, fungi and animal cells including insect and mammalian cells. Particularly advantageous are Drosophila melangaster cells, Saccharomyses cerevisiae and other yeasts, E. coli, Bacillus subtilis, SF9 cells, C129 cells, 293 cells, Neurospora, BHK, CHO, COS and HeLa cells, fibroblasts, Schwannoma cell lines, immortalized mammalian myeloid and lymphoid cell lines, Jurkat cells, mast cells and other endocrine and exocrine cells, and neuronal cells. See the ATCC cell line catalog, which is incorporated herein by reference. Furthermore, expression of secondary libraries in phage display systems as is well known in the art is particularly preferred, especially when the secondary library contains random peptides. In certain embodiments, the cells can be genetically engineered, that is, contain an exogenous nucleic acid, eg, contain a target molecule.

  In a preferred embodiment, the candidate mutant library protein is expressed in mammalian cells. Any mammalian cell may be used, with mouse, rat, primate and human cells being particularly preferred. As will be appreciated by those skilled in the art, modification of the system by pseudotype allows the use of all eukaryotic cells, preferably higher eukaryotes. As described in more detail below, the screening is set up to show a selectable phenotype in the presence of random library members. As detailed further below, as long as an appropriate screen can be designed to allow selection of cells exhibiting an altered phenotype as a result of the presence of library members in the cell, a wide variety of Cell types associated with various disease symptoms are particularly useful.

  Thus, suitable mammalian cell types include, but are not limited to, all types of tumor cells (especially melanoma-like, myeloid leukemia, lung carcinoma, breast carcinoma, ovarian carcinoma, colon carcinoma, kidney carcinoma, Prostate cancer, pancreatic carcinoma and testicular carcinoma), cardiomyocytes, endothelial cells, epithelial cells, lymphocytes (T cells and B cells), mast cells, eosinophils, intima cells, hepatocytes, mononuclear leukocytes Stem cells (used in screening for differentiation and dedifferentiation factors), osteoclasts, chondrocytes and other connective tissue cells, keratinocytes, melanocytes, liver, such as leukocytes, hematopoiesis, nerves, skin, lungs, kidneys, liver and muscle stem cells Includes cells, kidney cells and adipocytes. Suitable cells also include known research cells, including but not limited to Jurkat T cells, NIH3T3 cells, CHO, Cos and the like. See the ATCC cell line catalog, which is incorporated herein by reference.

  Mammalian expression systems are also known in the art and include retroviral systems. A mammalian promoter is any DNA sequence capable of binding mammalian RNA polymerase and initiating downstream (3 ′) transcription of a sequence encoding a library protein into mRNA. A promoter has a transcription initiation region, usually located near the 5 'end of the coding sequence, and a TATA box using 25-30 base pairs located upstream of the transcription initiation site. The TATA box is thought to direct RNA polymerase II to initiate RNA synthesis at the correct site. Mammalian promoters also contain an upstream promoter element (enhancer element), typically located within 100 to 200 base pairs upstream of the TATA box. The upstream promoter element determines the rate of transcription initiation and can act in either orientation. Since viral genes are often highly expressed and have a wide host range, mammalian virus gene-derived promoters are particularly useful as mammalian promoters. Examples include the SV40 early promoter, mouse mammary tumor virus LTR promoter, adenovirus major late promoter, herpes simplex virus promoter, and CMV promoter.

  Typically, transcription termination and polyadenylation sequences recognized by mammalian cells are regulatory regions located 3 'to the translation stop codon, and thus are flanked by coding sequences along with promoter elements. The 3 ′ end of the mature mRNA is formed by site-specific post-transcriptional cleavage and polyadenylation. Examples of transcription terminators and polyadenylation signals include those derived from SV40.

  Methods for introducing exogenous nucleic acid into mammalian hosts as well as other hosts are well known in the art and will vary with the host cell used. Techniques include dextran mediated transfection, calcium phosphate precipitation, polybrene mediated transfection, protoplast fusion, electroporation, viral infection, encapsulation of polynucleotide (s) in liposomes, and direct microinjection of DNA into the nucleus.

  In a preferred embodiment, the candidate mutant library protein is expressed in a bacterial system. Bacterial expression systems are well known to those skilled in the art.

  A suitable bacterial promoter is any nucleic acid sequence capable of binding bacterial RNA polymerase and initiating downstream (3 ′) transcription of a sequence encoding a library protein into mRNA. Bacterial promoters usually have a transcription initiation region located near the 5 ′ end of the coding sequence. This transcription initiation region typically includes an RNA polymerase binding site and a transcription initiation site. Sequences encoding metabolic pathway enzymes provide particularly useful promoter sequences. Examples include promoter sequences from sugar metabolizing enzymes such as galactose, lactose and maltose, and sequences from biosynthetic enzymes such as tryptophan. Promoters from bacteriophages can also be used and are known to those skilled in the art. In addition, synthetic and hybrid promoters are also useful; for example, the tac promoter is a hybrid of the trp and lac promoter sequences. In addition, bacterial promoters can include naturally produced promoters of non-bacterial origin that have the ability to bind bacterial RNA polymerase and initiate transcription.

  In addition to a functional promoter sequence, an effective ribosome binding site is desirable. In the case of E. coli, the ribosome binding site is called the Shine-Dalgarno (SD) sequence and includes a 3-9 nucleotide long sequence located 3-11 nucleotides upstream of the start codon and start codon.

  The expression vector can also include a signal peptide sequence that provides for secretion of the library protein in bacteria. The signal sequence typically encodes a signal peptide containing hydrophobic amino acids that directs protein secretion from the cell, as is well known to those skilled in the art. Proteins are secreted into either the growth medium (gram-positive bacteria) or the peripheral space (gram-negative bacteria) between the inner and outer membranes of the cell.

  Bacterial expression vectors can also include a selectable marker gene that allows selection of transformed bacterial strains. Suitable selection genes include genes that confer resistance to drugs such as ampicillin, chloramphenicol, erythromycin, kanamycin, neomycin and tetracycline. Selectable markers also include biosynthetic genes such as those in the histidine, tryptophan and leucine biosynthetic pathways.

  These components are assembled into an expression vector. Expression vectors for bacteria are well known in the art and include vectors for Bacillus subtilis, E. coli, Streptococcus cremoris and Streptococcus lividans, and the like.

  Bacterial expression vectors are transformed into bacterial host cells using techniques well known in the art such as calcium chloride treatment, electroporation and the like.

  In certain embodiments, the candidate mutant library protein is produced in insect cells. Expression vectors for insect cell transformation and in particular baculovirus-based expression vectors are well known in the art and are described in, for example, O'Reilly et al., Baculovirus Expression Vectors: A Laboratory Manual (New York: Oxford University Press, 1994). Are listed.

  In a preferred embodiment, the candidate mutant library protein is produced in yeast cells. Yeast expression systems are well known in the art and include the expression vectors for Saccharomyces cerevisiae, Candida albicans and C. maltosa, Hansenula polymorpha, Kluyveromyces fragilis and K. lactis, Pichia guillerimondii and P. pastoris, Schizosaccharomyces pombe and Yarrowia lipoa . Preferred promoter sequences for expression in yeast include inducible GAL1, 10 promoter, alcohol dehydrogenase, enolase, glucokinase, glucose-6-phosphate isomerase, glyceraldehyde-3-phosphate dehydrogenase, hexokinase, phosphofructokinase, Included are promoters from 3-phosphoglycerate mutase, pyruvate kinase and acid phosphatase genes. Yeast selectable markers include ALG7 conferring resistance to ADE2, HIS4, LEU2, TRP1, and tunicamycin, the neomycin phosphotransferase gene conferring resistance to G418, and the CUP1 gene that grows yeast in the presence of copper ions. .

  Candidate variant library proteins of the invention can also be made as fusion proteins using techniques well known in the art. Thus, for example, for the creation of monoclonal antibodies, if the desired epitope is small, the library protein can be fused to a carrier protein to form an immunogen. Alternatively, library proteins can be made as fusion proteins to increase expression or for other reasons. For example, if the library protein is a library peptide, the nucleic acid encoding the peptide may be combined with other nucleic acids for expression purposes. Similarly, a rescue that allows the purification or isolation of any of the target sequences, library proteins, or nucleic acids that encode them, that allow the localization of library members to the intracellular or extracellular compartments of the cell. Sequences or purification tags; stable sequences that confer stability or degradation protection (eg, resistance to proteolysis) to library proteins or nucleic acids encoding library proteins, or combinations thereof, and linker sequences if necessary Other fusion partners such as can also be used.

  Thus, suitable target sequences include, but are not limited to, binding capable of binding the expression product to a predetermined molecule or class of molecules while retaining the biological activity of the expression product. A sequence (eg, targeting an associated enzyme class using an enzyme inhibitor or substrate sequence); a sequence that provides a signal for selective degradation of the protein bound to itself or together; and a candidate expression product, a) intracellular locations such as the Golgi apparatus, endoplasmic reticulum, nucleus, kidney, nuclear membrane, mitochondria, chloroplast, secretory vesicle, lysosome, and cell membrane, and b) the extracellular location via a secretion signal, A signal sequence capable of constitutive localization to a predetermined cell location. Either intracellular localization or extracellular localization via secretion is particularly preferred.

In a preferred embodiment, the candidate mutant library member comprises a rescue sequence. The rescue sequence is a sequence that can be used to purify or isolate either the candidate substance or the nucleic acid that encodes it. Thus, for example, peptide rescue sequences include, for example, purified sequences such as His ₆ tags for use with Ni affinity columns and epitope tags for detection, immunoprecipitation or FACS (fluorescence activated cell sorting). It is. Suitable epitope tags include myc (used with the commercially available 9E10 antibody), BSP biotinylated target sequence of the bacterial enzyme BirA, flag tag, lacZ and GST.

  Alternatively, the rescue sequence may be a unique oligonucleotide sequence that acts as a probe target site that allows for rapid and easy isolation of retroviral constructs via PCR, related techniques or hybridization.

In a preferred embodiment, the fusion partner is a stable sequence that confers stability to the library member or the nucleic acid encoding it. Thus, for example, according to Varahavsky's N-terminal rule, to protect a peptide that is ubiquitinated, the peptide is stabilized by incorporating glycine after the initiation methionine (MG or MGG0), thus in the cytoplasm. Can provide a long half-life. Similarly, the C-terminal two prolines result in peptides that are well resistant to the action of carboxypeptidases. The presence of two glycines before proline confers both variability and protective structure and propagates events in di-proline into the candidate peptide structure. Thus, the preferred stable sequence is: MG (X) _n GGPP, where X is any amino acid and n is an integer of at least 4.

  In certain embodiments, candidate variant library nucleic acids, proteins and antibodies of the invention are labeled. As used herein, “labeling” refers to at least one component, isotope or chemical attached to the nucleic acids, proteins and antibodies of the present invention to enable detection of the nucleic acids, proteins and antibodies of the present invention. Means having a compound. In general, labels are classified into three categories: a) isotope labels that can be radioactive or heavy isotopes; b) immunolabels that can be antibodies or antigens; and c) colored or fluorescent dyes. The label can be incorporated into the compound at any position.

  In a preferred embodiment, the candidate mutant library protein is purified or isolated after expression. Library proteins can be isolated or purified in a variety of ways known to those skilled in the art depending on what other components are present in the sample. Standard purification methods include electrophoresis, molecules, immunological methods and chromatographic techniques and chromatofocusing including ion exchange, hydrophobicity, affinity and reverse phase HPLC chromatography. For example, library proteins can be purified using standard anti-library antibody columns. Ultrafiltration and diafiltration techniques are also useful in the context of protein concentration. For general guidance on appropriate purification techniques, see Scopes, R., Protein Purification, Springer-Verlag, NY (1982). The required degree of purification depends on the use of the library protein. Purification may not be necessary.

  In a preferred embodiment, the candidate mutant library protein is purified or isolated after expression. The mutant protein can be isolated or purified in a variety of ways known to those skilled in the art depending on what other components are present in the sample. Standard purification methods include electrophoresis, molecules, immunological methods and chromatographic techniques and chromatofocusing including ion exchange, hydrophobicity, affinity and reverse phase HPLC chromatography. For example, the mutein can be purified using a standard anti-library antibody column. Ultrafiltration and diafiltration techniques are also useful in the context of protein concentration. For general guidance on appropriate purification techniques, see Scopes, R., Protein Purification, Springer-Verlag, NY (1982). The required degree of purification varies depending on the use of the mutant protein. Purification may not be necessary.

  Once expressed, and if necessary, purified candidate mutant library proteins and nucleic acids can be tested for immunogenic alteration. Suitable methods include measurement of MHC peptide complex binding to TCR, measurement of MHC / peptide interaction (Sidney, J., et al., In Current Protocols in Immunology (1998) 18.3.1-18.3.19 ), Testing for potential T cell epitopes in transgenic mice expressing human MHC molecules, potential T cells in mice reconstituted with human antigen presenting cells and T cells instead of endogenous cells Epitope testing (WO 98/52976; WO 00/34317), T cell proliferation and CTL assay (Hemmer, B., (1998) J. Immunol., 160: 3631-3636), stabilization assay, purified MHC molecule or Competitive inhibition assay for MHC with MHC (Brusic, V., et al., (1998) Nucleic Acids Res., 26: 368-371) and “i-mune assay” (Genecor; The Scientist, 15:14 , (2001)), including all references And is incorporated herein by reference).

  Once made, the candidate mutant proteins and nucleic acids of the invention are useful for a number of applications. In a preferred embodiment, candidate mutant proteins that are less immunogenic than the target protein are used as therapeutic proteins. For example, clinical and preclinical treatment studies have shown that exogenous proteins can be effective in vivo as artificial receptors for capturing radionuclides, as toxicants, or as catalysts for prodrug activation. (Meyer, DL., Et al. (2001) Protein Science, 10: 491-503). Other uses of therapeutic proteins with reduced immunogenicity include thrombolytic treatment of acute myocardial infarction (Laroche, Y., et al., (2000) Blood, 96: 1425-1432).

  In a preferred embodiment, candidate mutant proteins that are more immunogenic than the target protein are used in the development of vaccines and immunotherapeutic agents against autoimmune diseases and cancer. For example, insertion of linear amino acid sequence epitopes with increased affinity for MHC class I or class II molecules can create vaccines that are more effective in inducing immune responses (eg, Sarobe, P., et al. ( 1998) J. Clin. Invest., 102: 1239-1248; Thimme, R., et al. (2001) J. Virology, 75: 3984-3987; Roberts, C., et al., (1996) Aids Research and Human Retroviruses, 12: 593-610 (eg Sarobe, P., et al. (1998) J. Clin. Invest., 102: 1239-1248; Thimme, R., et al. (2001) J. Virology , 75: 3984-3987; Roberts, C., et al., (1996) Aids Research and Human Retroviruses, 12: 593-610; Kobayashi, H., et al., (2000) Cancer Res., 60: 5228 Keogh, E., et al., (2001) J. Immunology, 167: 787-796; Want, RF., (2001) Trends in Immunology, 22: 269-276; As a part of the book)).

  In a preferred embodiment, the insertion of at least one T cell epitope creates a vaccine that is more effective in inducing an immune response (de Lalla, C., et al., (1999) J. Immunology, 163: 1725 Kim and DeMars, (2001) Curr. Op Immunology, 13: 429-436; Berzofsky, JA, et al., European Patent Publication No. 0 273 716B1; all references are incorporated herein in their entirety. Part).

  In another embodiment, the insertion of a sequence encoding at least one structural three-dimensional epitope that interacts with a membrane-bound antibody on naive B cells creates a vaccine that is more effective in inducing an immune response (Criag , L., et al., (1998) J. Mol. Biol., 281: 183-201; Buttinelli, G., et al., (2001) Virology, 281: 265-271; Saphire, EO, et al ., (2001) Science, 293: 1155; Mascola and Nabel, (2001) Curr. Op. Immunology, 13: 489-495; all references are incorporated herein in their entirety. ).

  In still other embodiments, insertion of any combination of B cell epitopes, MHC class I binding motifs, MHC class II binding motifs and T cell epitopes creates vaccines that are more effective in inducing immune responses (eg, WO 01/41788 and US Pat. No. 6,037,135).

  An effective vaccine against allergens, bacterial pathogens, viral pathogens and tumors can be designed. See, for example, WO / 41788; US Pat. No. 6,322,789; US Pat. No. 6,329,505; WO 01/41799; WO 01/42267; WO 01/42270; and WO 01/45728.

  For example, a vaccine may be designed against one or more allergens, including but not limited to chemical allergens, food allergens, pollen allergens, fungal allergens, pet scales, mites, etc. (See Huby, RD et al., (2000) Toxicological Science, 55: 235-246, which is hereby incorporated by reference in its entirety).

  Preferably, hepatitis A, hepatitis B, hepatitis C, poliovirus, HIV, herpes simplex I and II, smallpox, human papilloma virus, cytomegalovirus, hantavirus, rabies, Ebola virus, yellow fever virus, rotavirus, Vaccines are made against viral pathogens including, but not limited to, rubella, measles virus, mumps virus, chickenpox (ie chicken pox), influenza, encephalitis, Lassa fever virus, and the like.

Preferably, Lyme disease, diphtheria, anthrax, botulism, pertussis, pertussis ^* , tetanus, cholera, typhoid fever, typhoid, plague, leprosy, tuberculosis (including multi-drug resistant forms), staphylococcal infection, streptococcal infection, Vaccines are made against bacterial pathogens including, but not limited to, causative agents such as Listeria, meningococcal meningitis, pneumococcal infection, Legionella disease, ulcers, and conjunctivitis.

  Other infections, including but not limited to dengue fever, malaria, African sleeping sickness, dysentery, rocky mountain spotted fever, schistosomiasis, diarrhea, West Nile fever, leishmaniasis, giardiasis Vaccines can also be made for substances.

  In other embodiments, the candidate variant protein is more immunogenic against a variety of cancers including solid tumors such as skin, breast, brain, cervical carcinoma, testicular carcinoma. In particular, cancers that can be treated by the compositions and methods of the present invention include: heart: sarcoma (angiosarcoma, fibrosarcoma, rhabdomyosarcoma, liposarcoma), myxoma, rhabdomyosarcoma, fibroma, lipoma and malformation. Lung: Bronchiogenic carcinoma (squamous cell, undifferentiated small cell, undifferentiated large cell, adenocarcinoma), alveolar (bronchiole) carcinoma, bronchial adenoma, sarcoma, lymphoma, cartilaginous hamartoma, mesothelioma Digestive organs: esophagus (squamous cell carcinoma, adenocarcinoma, leiomyosarcoma, lymphoma), stomach (carcinoma, lymphoma, leiomyosarcoma), pancreas (tubular adenocarcinoma, insulin-producing tumor, glucagon-producing tumor, gastrin-producing tumor, Carcinoid tumor, vipoma, small intestine (adenocarcinoma, lymphoma, carcinoid tumor, Kaposi's sarcoma, leiomyoma, hemangioma, lipoma, neurofibroma, fibroma), large intestine (adenocarcinoma, tubular adenoma, villous adenoma, Hamartoma, leiomyoma); urinary reproduction Pathology: Kidney (adenocarcinoma, Wilms tumor [nephroblastoma], lymphoma, leukemia), bladder and urethra (squamous cell carcinoma, transitional cell carcinoma, adenocarcinomas), prostate (adenocarcinoma, sarcoma), testis (seminal epithelioma) Teratoma, embryonal carcinoma, teratocarcinoma, choriocarcinoma, sarcoma, stromal cell carcinoma, fibroma, fibroadenoma, adenomatous tumor, lipoma); liver: hepatocellular carcinoma (hepatocellular carcinoma), cholangiocarcinoma, Hepatoblastoma, hemangiosarcoma, hepatocellular adenoma, hemangioma; bone: osteogenic sarcoma (osteosarcoma), fibrosarcoma, malignant fibrous histiocytoma, chondrosarcoma, Ewing sarcoma, malignant lymphoma (reticular cell sarcoma) , Multiple myeloma, malignant giant cell tumor chordoma, osteochronfroma (osteochondroma), benign chordoma, chordoblastoma, chordal mucinous fibroma, osteoid osteoma and giant cell tumor; nerve System: skull (osteomas, hemangiomas, granulomas, xanthomas, osteoarthritis), meninges (meningiomas, meninges) , Glioma), brain (astrocytoma, medulloblastoma, glioma, ependymoma, germinomas [pineomas], glioblastoma multiforme, oligodendroma, Schwann cell , Retinoblastoma, congenital tumor), spinal neurofibroma, meningioma, glioma, sarcoma); gynecology: uterus (endometrial carcinoma), cervix, (cervical carcinoma, pre-tumor cervix) Dysplasia), ovary (ovarian carcinoma [serous cystadenocarcinoma, mucinous cystadenocarcinoma, unclassifiable carcinoma], granulosa-capsular cell tumor, Sertoli-Leydig cell tumor, undifferentiated germ cell tumor, malignant teratoma) , Vulva (squamous cell carcinoma, carcinoma in situ, adenocarcinoma, fibrosarcoma, melanoma), vagina (clear cell carcinoma, squamous cell carcinoma, grape sarcoma [embryonic rhabdomyosarcoma), fallopian tube (carcinoma) ); Hematology: blood (myeloid leukemia [acute and chronic], acute lymphoblastic leukemia, chronic lymphocytic leukemia, Myeloproliferative disease, multiple myeloma, myelodysplastic syndrome), Hodgkin's disease, non-Hodgkin's lymphoma [malignant lymphoma]; skin: malignant melanoma, basal cell carcinoma, squamous cell carcinoma, Kaposi's sarcoma, mole-dysplastic mother Includes, but is not limited to, plaques, lipomas, hemangiomas, dermal fibromas, keloids, psoriasis; and adrenal gland: neuroblastoma.

  In a preferred embodiment, the vaccine comprises a tumor with p53, a melanoma antigen gene (MAGE; see WO 01/42267); a cancer embryo antigen (CEA; see WO 01/42270), a prostate specific antigen (PSA), Prostate cancer antigens (see WO 01/45728 and US Pat. No. 6,329,505), such as prostate specific membrane antigen (PSM), prostate acid phosphatase (PAP) and human kallikrein 2 (hK2 or HuK2), and breast cancer antigens (ie her2 / neu; see AU 2087401).

  In a preferred embodiment, a therapeutically effective amount of the candidate mutein is administered to a patient in need of treatment. As used herein, “therapeutically effective amount” means a dose that produces the effect for which it is administered. The exact dose will depend on the purpose of the treatment, and will be ascertainable by one skilled in the art using known techniques. In a preferred embodiment, a dose of about 5 μg / kg is used and is administered intravenously, intraperitoneally or subcutaneously. Candidate mutant proteolysis, systemic versus local delivery, and novel protease synthesis rates, as well as known in the art, and age, weight, general health, sex, diet, time of administration, drug interactions and severity of disease state Adjustment is necessary and can be confirmed by one skilled in the art by routine experimentation.

  “Patient” for the purposes of the present invention includes both humans and other animals, particularly mammals, and organisms. Thus, the method of the present invention is applicable to both human therapy and veterinary applications. In a preferred embodiment, the patient is a mammal, and in the most preferred embodiment, the patient is a human.

  The term “treatment” in the present invention is meant to include therapeutic treatment as well as means for preventing or suppressing a disease or disorder. Thus, for example, successful administration of a candidate mutant protein prior to the onset of the disease leads to “treatment” of the disease. As another example, successful administration of a mutein after clinical manifestation of a disease to combat disease symptoms includes “treatment” of the disease. “Treatment” also includes administering the mutant protein after the appearance of the disease to eradicate the disease. Successful administration of a substance after onset and after progression of clinical symptoms, with a possible reduction in clinical symptoms and possibly amelioration of the disease, includes “treatment” of the disease.

  Persons “in need of treatment” include mammals already having a disease or abnormality, as well as those prone to having the disease or abnormality, in whom the disease or abnormality should be prevented.

  Administration of the candidate mutant proteins of the invention, preferably in sterile aqueous solution form, includes oral, subcutaneous, intravenous, intranasal, transdermal, intraperitoneal, intramuscular, intrapulmonary, intravaginal, rectal or intraocular administration. Various methods can be used, including but not limited to these. In some cases, the candidate mutein may be applied directly as a solution or spray, for example in the treatment of injury or inflammation. Depending on the method of introduction, the pharmaceutical composition can be formulated in various ways. The concentration of the therapeutically active candidate mutein in the formulation can vary from about 0.1 to 100% by weight. In another preferred embodiment, the concentration of the candidate mutein is in the range of 0.003 to 1.0 molar and is 0.03, 0.05, 0.1, 0.2, and 0.0 per kilogram body weight. A dose of 3 mmol is preferred.

  The pharmaceutical composition of the invention comprises a candidate mutein in a form suitable for administration to a patient. In a preferred embodiment, the pharmaceutical composition is in a water-soluble form, such as present as a pharmaceutically acceptable salt, which is intended to include both acid and base addition salts. “Pharmaceutically acceptable acid addition salts” retain the biological effectiveness of the free base and are not biologically or otherwise inconvenient and include hydrochloric acid, hydrobromic acid, sulfuric acid, nitric acid. , Inorganic acids such as phosphoric acid, and acetic acid, propionic acid, glycolic acid, pyruvic acid, succinic acid, maleic acid, malonic acid, succinic acid, fumaric acid, tartaric acid, citric acid, benzoic acid, cinnamic acid, mandelic acid, methane It forms with organic acids, such as sulfonic acid, ethanesulfonic acid, p-toluenesulfonic acid, salicylic acid. "Pharmaceutically acceptable base addition salts" include those derived from inorganic bases such as sodium, potassium, lithium, ammonium, calcium, magnesium, iron, zinc, copper, manganese, aluminum salts and the like. Particularly preferred are ammonium, potassium, sodium, calcium and magnesium salts. Salts derived from pharmaceutically acceptable organic non-toxic bases include primary, secondary and tertiary amines, substituted amines such as naturally occurring substituted amines, cyclic amines and bases. Ionic ion exchange resins such as salts of isopropylamine, trimethylamine, diethylamine, triethylamine, tripropylamine and ethanolamine are included.

  The pharmaceutical composition also includes the following materials: carrier proteins such as serum albumin; buffers such as NaOAc; fillers such as microcrystalline cellulose, lactose, corn and other starches; binders; sweeteners and others Flavoring agents; coloring agents; and one or more of polyethylene glycols. Additives are well known in the art and are used in various formulations. See, for example, Goodman and Gilman, which is incorporated herein by reference.

  In a further embodiment, the candidate mutant protein is added to the micelle formulation; see US Pat. No. 5,833,948, hereby incorporated by reference in its entirety.

  Combinations of pharmaceutical compositions can be administered. For example, enhanced immunogenicity selected from the group consisting of variants of soluble proteins such as zinc-alpha 2-glycoprotein, human serum albumin, immunoglobulin G (IgG) and other non-immunogenic proteins. A pharmaceutical composition comprising a mixture of the variant proteins shown can be administered to a patient. In addition, the composition may be administered in combination with other therapeutic agents.

  In certain embodiments provided by the present invention, methods known in the art are used to generate antibodies to muteins, including but not limited to monoclonal and polyclonal antibodies. Part, see Soren, M., et al., EP 0 752 886). In preferred embodiments, these anti-variant antibodies are used for immunotherapy. Accordingly, a method of immunotherapy is provided. “Immunotherapy” refers to the treatment of autoimmune diseases that involve the production of self-proteins. In particular, to make a self-vaccine, the self-protein is linked to a T cell epitope (see, eg, WO 95/05849 and WO 00/20027; both are hereby incorporated by reference. ) Self-proteins for use in the present invention include TNFα and γ-interferon for the treatment of cancer, IgE for the treatment of allergies, TNFα, TNFβ and interleukin 1 for the treatment of chronic inflammatory diseases.

  As used in the present invention, immunotherapy can be passive or active. As defined herein, passive immunotherapy is the passive delivery of antibodies to a recipient (patient). Active immunization is the induction of antibody and / or T cell responses in a recipient (patient). Induction of the immune response may be the result of giving the recipient a mutated protein antigen that includes a T cell epitope and a self protein, against which an antibody is generated. As will be appreciated by those skilled in the art, a mutant protein antigen is a mutant protein antigen that is injected into a recipient with a mutant polypeptide that is desired to produce antibodies, or under conditions to express a mutant TNFα protein antigen. A nucleic acid encoding a mutein capable of expressing is provided by contacting the recipient.

  In a preferred embodiment, the candidate mutein is administered as a therapeutic agent and can be formulated as outlined above. Similarly, candidate mutant genes (including both full-length sequences, partial sequences, or mutant coding region regulatory sequences) can be administered in gene therapy applications as is known in the art. As will be appreciated by those skilled in the art, these mutant genes include antisense applications, either as gene therapy (ie, for integration into the genome) or as an antisense composition.

  In a preferred embodiment, nucleic acids encoding candidate mutant proteins can also be used in gene therapy. In gene therapy applications, genes are introduced into cells to achieve in vivo synthesis of therapeutically effective gene products, eg, for defective gene replacement. “Gene therapy” includes both conventional gene therapy that achieves a sustained effect with a single treatment and administration of a gene therapy agent, including single or repeated administration of therapeutically effective DNA or mRNA. Is included. Antisense RNA and DNA can be used as therapeutic agents to block the expression of certain genes in vivo. Despite low intracellular concentrations due to limited uptake by the cell membrane, it has already been shown that short antisense oligonucleotides are introduced into cells where they act as inhibitors (Zamecnik et al., Proc. Natl. Acad. Sci. USA 83, 4143-4146 [1986]). For example, oligonucleotides can be altered by replacing negatively charged phosphodiester groups with uncharged groups to enhance uptake.

  A variety of techniques are available for introducing nucleic acids into viable cells. The technique depends on whether the nucleic acid is transferred in vitro into cultured cells or in vivo into the intended host cell. Suitable techniques for transferring nucleic acids into mammalian cells in vitro include the use of liposomes, electroporation, microinjection, cell fusion, DEAE-dextran, calcium phosphate precipitation, and the like. In vivo gene transfer techniques currently preferred include transfection using viral (typically retroviral) vectors, and viral coat protein-liposome-mediated transfection (Dzau et al., Trends in Biotechnology 11 , 205-210 [1993]). In certain situations, it may be desirable to provide a nucleic acid source having a substance that targets the target cell, such as a cell surface membrane protein or an antibody specific for the target cell or a ligand for a receptor on the target cell. When liposomes are employed, proteins that bind to cell surface membrane proteins with endocytosis can be used to target and / or promote uptake. For example, capsid proteins or fragments thereof having affinity for specific cell types, antibodies to proteins that undergo internalization during cycling, proteins that target intracellular localization and enhance intracellular half-life. Techniques for receptor-mediated endocytosis are described, for example, in Wu et al., J. Biol. Chem. 262,4429-2232 (1987): and Wagnernet al., Proc. Natl. Acad. Sci. USA 87, 3410-3414. (1990). For an overview of gene marking and gene therapy protocols, see Andersen et al., Science 256, 808-813 (1992).

  In a preferred embodiment, the candidate mutant gene is administered as a DNA vaccine. Either a single gene or a combination of candidate mutant genes. Bare DNA vaccines are generally known in the art; Brower, Nature Biotechnology 16: 1304-1305 (1998). Methods of using genes as DNA vaccines are well known to those skilled in the art and involve placing a candidate mutant gene or portion of a mutant gene under the control of a promoter for expression in a patient in need of treatment. The mutant gene used in the DNA vaccine can encode the full length of the mutant protein, but more preferably encodes a portion of the mutant protein, including peptides derived from the mutant protein. In a preferred embodiment, the patient is immunized with a DNA vaccine comprising a number of nucleotide sequences derived from the mutated gene. Similarly, it is possible to immunize a patient with a number of mutated genes or portions thereof, as defined herein. Without being limited by theory, expression of the polypeptide encoded by the DNA vaccine is followed by induction of cytotoxic T cells, helper T cells and antibodies that recognize and destroy or eliminate cells expressing the TNFα protein. Is done.

  In a preferred embodiment, the DNA vaccine includes a gene encoding an adjuvant molecule with the DNA vaccine. Such adjuvant molecules include cytokines that increase the immune response to the mutant polypeptide encoded by the DNA vaccine. Additional or alternative adjuvants are well known to those of skill in the art and are useful in the present invention.   All references cited herein are hereby incorporated by reference.

FIG. 1 depicts full-length gene synthesis and possible total mutation introduction by PCR.    FIG. 2 shows a preferred scheme for synthesizing the libraries of the present invention.    FIG. 3 shows the overlap extension method.    FIG. 4 shows the ligation of PCR reaction products to synthesize the library of the present invention.    FIG. 5 depicts blunt end ligation of PCR products.   

Claims (22)

  A method for producing a polypeptide exhibiting enhanced immunogenicity, comprising: a) entering the target backbone structure into the computer along with the variable residue positions; b) Any order   i) at least one computerized protein design algorithm; and   ii) At least one computerized immunogenic filter Applying, and c) identifying at least one variant protein with enhanced immunogenicity, Including methods.   A method for producing a polypeptide exhibiting reduced immunogenicity, comprising: a) entering the target backbone structure into the computer along with the variable residue positions; b) Any order   i) at least one computerized protein design algorithm; and   ii) At least one computerized immunogenic filter Applying, and c) identifying at least one mutant protein having reduced immunogenicity, Including methods.   A method of inducing an enhanced immune response in a patient comprising a) entering the target backbone structure into the computer along with the variable residue positions; b) Any order   i) at least one computerized protein design algorithm; and   ii) At least one computerized immunogenic filter Applying, c) identifying at least one mutant protein having enhanced immunogenicity; and d) administering the mutant protein to a patient; Including methods.   4. The method of claim 1, 2 or 3, wherein the computerized protein design algorithm is applied prior to the filter.   The method of claim 1, claim 2 or claim 3, wherein the computerized protein design algorithm is applied after the filter.   4. The method of claim 1, 2 or 3, wherein the computerized protein design algorithm includes the filter as a scoring function.   The target protein is selected from the group consisting of Zn-alpha 2-glycoprotein, human serum albumin, immunoglobulin G, and non-immunogenic protein, claim 1, claim 3, claim 4, claim 4, 7. A method according to claim 5 or claim 6.   8. The computerized immunogenic filter of claim 1, claim 2, claim 3, claim 4, claim 6 or claim 7, wherein the scoring function for MHC class I motifs. the method of.   Claim 1, Claim 2, Claim 3, Claim 4, Claim 5, Claim 7, or Claim 7, wherein the computerized immunogenic filter comprises a scoring function for the MHC class II motif. Item 9. The method according to Item 8.   The enhanced immunogenicity is due to the presence of at least one immunogenic sequence, claim 1, claim 3, claim 4, claim 5, claim 6, claim 7 10. A method according to claim 8 or claim 9.   11. A method according to claim 10, wherein the immunogenic sequences are identical.   11. The method of claim 10, wherein the immunogenic sequences are different.   13. The method according to claim 10, 11 or 12, wherein the immunogenic sequence is selected from the group consisting of a B cell epitope, a T cell epitope, an MHC class I motif and an MHC class II motif.   14. The method of claim 10, 11, 12, or 13, wherein the immunogenic sequence further comprises a specific cleavage motif.   The step of generating by the computer processing includes DEE computer calculation, claim 1, claim 3, claim 4, claim 5, claim 7, claim 8, claim 9. 15. A method according to claim 10, claim 11, claim 12, claim 13 or claim 14.   16. The method of claim 15, wherein the DEE computer calculation is selected from the group consisting of an original DEE and a Goldstein DEE.   Claim 1, Claim 2, Claim 3, Claim 4, Claim 5, Claim 7, wherein the set of primary variant amino acid sequences is optimized for at least one scoring function. The method according to claim 8, claim 9, claim 10, claim 11, claim 12, claim 13, claim 14, claim 15 or claim 16.   18. The method of claim 17, wherein the set of primary variant amino acid sequences that are optimized for at least one scoring function comprises a global optimal protein sequence.   The scoring function is selected from the group consisting of a van der Waals potential scoring function, a hydrogen bond potential scoring function, an atomic solvation scoring function, an electrostatic scoring function, and a secondary structure trend scoring function. 18. The method according to 17.   Claims 1, 2, 3, 4, 5, 6, 7, 8, 8, wherein the step of generating by computer includes the use of a Monte Carlo search. A method according to claim 10, claim 11, claim 12, claim 13, claim 14, claim 15, claim 16, claim 17, claim 18 or claim 19.   Claim 1, Claim 2, Claim 3, Claim 4, Claim 5, Claim 6, Claim 7, Claim 8, Claim 9, Claim 11, Claim 12, Claim A modified polypeptide exhibiting enhanced immunogenicity produced by the method of claim 13, claim 14, claim 15, claim 16, claim 17, claim 18, claim 19 or claim 20.   4. The method of claim 3, wherein the variant protein is selected from the group consisting of Zn-alpha 2-glycoprotein, human serum albumin, immunoglobulin G, non-immunogenic protein variants and mixtures thereof.

Images