RU2303264C2 - Method for identifying epitopes of t-cells and its using in preparing molecules of reduced immunogenicity - Google Patents

Abstract

FIELD: medicine, biochemistry.

SUBSTANCE: invention relates to a method for identifying epitopes of T-cells that induce the immune response in living organism-host. By this method immunogenicity-free biological compounds can be prepared or with reduced immunogenicity and in comparison with the corresponding unmodified object by decreasing or removing potential epitopes of T cells in limits of sequence of the parent immunogenic molecules. Also, invention relates to novel biological molecules being especially to proteins and antibodies prepared by the proposed invention. Invention provides reducing immunogenicity of biological compounds.

EFFECT: improved identifying method.

17 cl, 3 tbl, 8 dwg, 22 ex

Description

Область изобретения

Field of Invention

This invention relates to a new approach in identifying T-cell epitopes that elicit an immune response in a living host organism, which involves calculating potential T-cell epitopes for the binding sites of MHC class II molecules in a peptide using automated methods. In addition, the invention relates to methods for producing biological molecules, especially proteins and antibodies, which elicit an immune response when introduced into a host organism, preferably a human. Using this method, molecules that are non-immunogenic or have reduced immunogenicity when exposed to a particular type of immune system can be obtained, and compared with the corresponding unmodified object, by reducing or removing potential T-cell epitopes within the sequence of these initially immunogenic molecules. Thus, the invention also relates to new biological molecules, which are obtained by the method according to the invention.

BACKGROUND OF THE INVENTION

The therapeutic use of certain peptides, polypeptides and proteins is limited due to their immunogenicity in mammals, especially in humans. For example, when applying murine antibodies to patients who do not have immunosuppressants, most of these patients display immune responses to the introduced foreign object by producing human anti-mouse antibodies (NAMAs) (e.g. Schroff RW et al. (1985), Cancer Res. 45: 879 -885; Shawler DL et al. (1985) J. Immunol. 135: 1530-1535). There are two serious consequences. First, a patient’s anti-mouse antibody can bind and destroy a therapeutic antibody or immunoconjugate before it can bind, for example, a tumor and exert its therapeutic function. Secondly, the patient may develop allergic sensitivity to the mouse antibody and there may be a risk of anaphylactic shock with any administration of mouse immunoglobulin in the future.

Several methods were used to solve the NAMA problem and, thus, created the possibility of using human therapeutic monoclonal antibodies (see, for example, WO-A-8909622; EP-A-0239400; EP-A-0438310; WO-A-9106667 ) These recombinant DNA-based approaches typically attenuate mouse genetic information in the final antibody construct, thereby enhancing human genetic information in the resulting construct. Nevertheless, the obtained “humanized" antibodies nevertheless evoke in some cases an immune response in patients (Issacs JD (1990) Sem. Jmmunol. 2: 449, 456; Rebello, PR et al. (1999) Transplantation 68: 1417-1420 )

A common aspect in these methodologies is the introduction into a therapeutic antibody, usually from rodents, amino acid residues, even significant pathways of amino acid residue sequences identical to those found in human antibody proteins. For antibodies, this method is possible due to the relatively high degree of structural (and functional) conservatism among antibody molecules of various types. However, for potentially therapeutic peptides, polypeptides, and proteins where structural homology in host species (eg, humans) cannot exist for a therapeutic protein, such methods are not applicable. In addition, these methods suggest that the main introduction of a human amino acid residue sequence will render the remodeled antibody non-immunogenic. It is known that certain short peptide sequences (“T cell epitopes”) can be released during the destruction of peptides, polypeptides or proteins within cells and, therefore, take part in the presentation by molecules of the main histocompatibility complex (MHC) in order to initiate T- activation cells. For peptides represented by MHC class II, such activation of T cells can then elicit, for example, an antibody response by directly stimulating B cells to produce such antibodies. Accordingly, it would be desirable to eliminate potential T cell epitopes from a peptide, polypeptide or protein. Even proteins of human origin and those that have human amino acid sequences can induce an immune response in humans. Illustrative examples include the therapeutic use of colony-stimulating macrophage granulocyte factor (Wadhwa M. et al. (1999) Clin. Cancer. Res. 5: 1353-1361) and interferon alfa-2 (Russo D. et al. (1996) Bri. J. Haem. 94: 300-305; Stein R. et al. (1998) New Engl. J. Med. 318: 1409-1413).

Elimination of T cell epitopes from a protein has been previously disclosed (see, for example, WO 98/52976, WO 00/34317). The main methods disclosed in the prior art include the following steps:

(a) Determination of the amino acid sequence of a polypeptide or part thereof.

(b) Identification of one or more potential T cell epitopes within the amino acid sequence of the protein by any method, including determining the binding of peptides to MHC molecules using in vitro or in silico methods or biological assays.

(c) Construction of new sequence variants with one or more amino acids within the identified potential T-cell epitopes, thus modified to substantially reduce or eliminate the activity of T-cell epitopes, as determined by binding of peptides to MHC molecules using in vitro or in silico or biological tests. Such sequence variants are designed in such a way as to avoid the formation of new potential T-cell epitopes by changing the sequences until such new potential T-cell epitopes, in turn, are modified in this way to significantly reduce or eliminate the activity of T- epitopes cells.

(d) constructing such sequence variants using recombinant DNA techniques and testing said variants in order to identify one or more variants with desired properties.

Other methods using soluble complexes of recombinant MHC molecules in combination with synthetic peptides that are capable of binding T-cell clones to human peripheral blood samples or experimental animals are used in the art (Kern F. et al. (1998) Nature Medicine 4: 975-978; Kwok, WW et al. (2001) TRENDS in Immunology 22: 583-588) and can be used in an epitope identification strategy.

Potential T-cell epitopes are usually defined as any sequence of amino acid residues that has the ability to bind to MHC class II molecules. Such T cell epitopes can be measured to establish MHC binding. Of course, “T-cell epitope” means an epitope which, when bound to MHC molecules, can be recognized by the T-cell receptor and which can, at least in principle, trigger the activation of these T-cells. This, however, is commonly understood as the fact that certain peptides that are such that bind to MHC class II molecules can be held in the protein sequence, since such peptides are recognized as "intrinsic" to the body into which the final protein is introduced.

The aim of the present invention was to overcome the practical reality that soluble proteins introduced into the autologous host organism for therapeutic purposes can initiate an immune response leading to the development of host antibodies that bind the soluble protein. Among other examples is interferon alfa-2, to which some patients develop antibodies despite the fact that this protein is produced endogenously [Russo D. et al. (1996) Bri. J. Haem. 94: 300-305; Stein R. et al. (1998) New Engl. J. Med. 318: 1409-1413].

MHC class II molecules are a group of highly polymorphic proteins that play a central role in the selection of helper T cells and activation. The human leukocyte antigenic group DR (HLA-DR) is the predominant isotype of this group of proteins and is the basis of this invention. However, the HLA-DQ and HLA-DP isotypes perform similar functions, thus the invention is equally applicable to these two isotypes. MHC molecules of class II DR are understood as homodimers in which each "half" is a heterodimer consisting of α and β chains. Each heterodimer has a ligand-binding domain that binds to peptides and has a length ranging from 9 to 20 amino acids, despite the fact that the binding groove may include a maximum of 9-11 amino acids. The ligand-binding domain includes amino acids 1 to 85 of the α chain, and amino acids 1 to 94 of the β chain. DQ molecules have recently been shown to have a homologous structure and the DP protein family is also expected to be very similar. About 70 different allotypes of DR isotype are known in humans, 30 different allotypes are known for DQ, and 47 different allotypes for DP. Each individual carries from two to four DR alleles, two DQ alleles and two DP alleles. The structure of a number of DR molecules has been studied, and these structures indicate a binding groove of the peptide with an open end structure with a series of hydrophobic pockets that involve hydrophobic residues (pocket residues) of the peptide (Brown et al. Nature (1993) 364: 33, Stern et al. ( 1994) Nature 368: 215). Polymorphism that identifies various class II allotypes of molecules contributes to a wide variety of different peptide binding surfaces within the peptide binding groove and at the population level provides maximum flexibility regarding the ability to recognize foreign proteins and elicit an immune response to pathogenic organisms.

There is significant polymorphism within the ligand-binding domain with distinct “families” within different geographical populations and ethnic groups. This polymorphism affects the binding characteristics of the peptide binding domain, and thus, different “families” of DR molecules will have specificity for peptides with different sequence properties, although they may overlap to some extent. This specificity determines the recognition of Th-cell epitopes (class II T-cell response), which are ultimately responsible for stimulating the production of antibodies to B-cell epitopes that are present in the same protein from which the Th-cell epitope originates. Thus, the immune response to a protein in an individual human body is largely influenced by the recognition of the T-cell epitope, which is a function of the peptide-binding specificity of such an individual HLA-DR allotype. Thus, in order to identify T-cell epitopes within a protein or peptide in the context of a global population, it is desirable to consider the binding properties of as wide a set of HLA-DR allotypes as possible, thus covering as high a percentage of the world population as possible.

The immune response to a therapeutic protein, such as the protein of the invention, goes through the presentation path of the MHC class II peptide. Thus, in order to eliminate or reduce immunogenicity, it is desirable to identify and remove T cell epitopes from the protein.

Unmodified biological molecules can be obtained by the recombinant method, which per se is well known in the art using a number of different types of host cells.

However, there is a need for analogues of these biological molecules with improved properties. Desirable improvements include alternative regimens and modalities for the expression and purification of the indicated therapeutic agent, as well as in particular the improvement of the biological properties of the protein. There is a particular need to improve in vivo performance when administered to humans. In this regard, it is especially desirable to provide selected biological molecules with a reduced or absent potential for inducing an immune response in the human body. It is expected that such proteins will show increased circulation time in the human body and will be particularly useful in the treatment of chronic or recurring manifestations of the disease, such as in the case of some indications for the specified biological molecule.

SUMMARY AND DESCRIPTION OF THE INVENTION

The present invention therefore relates to two main aspects, such as:

(a) a convenient and efficient computational method for identifying and calculating T-cell epitopes for universally different numbers of MHC class II molecules and, based on this knowledge, for developing and constructing new variants of sequences of biological molecules with improved properties, and

(b) new biologically active molecules for use especially in humans and, in particular, for therapeutic use; these biological molecules according to this invention are immunogenically modified polypeptides, proteins and immunoglobulins (antibodies) obtained according to the method of this invention, whereby the modification leads to a reduced tendency of the biological molecule to elicit an immune response when administered to humans.

In particular, the invention relates to the modification of several generally known proteins and antibodies with a high therapeutic advantage of human or non-human origin, obtained by the method according to this invention, which leads to the production of proteins that are substantially non-immunogenic or less immunogenic than any of their unmodified equivalent when used in vivo. Molecules modified according to the new method of the present invention are expected to exhibit an increased circulation time in the human body and will be particularly useful in treating chronic or recurring manifestations of the disease, such as in the case of some indications. The present invention provides, both in a specific embodiment and for demonstrating the effectiveness of the method of the invention, modified forms of said molecules which are expected to exhibit improved in vivo properties. These molecules with modified immunogenicity, that is, with reduced immunogenic potential, can be used in pharmaceutical compositions. Such modified molecules in this description are called "immunogenically" modified.

A method for identifying T-cell epitopes in part by computational means can be used to calculate theoretical values of T-cell epitopes and, thus, to identify potential binding peptides of a MHC class II molecule within a protein; where the binding site includes a sequence of amino acid sites within the protein. The identified peptides can then be modified without a significant decrease and, possibly, an increase in the therapeutic value of the protein. This computational method involves selecting a region of a protein having a known sequence of amino acid residue by sequentially testing overlapping segments of amino acid residues (windows) of a predetermined constant length and consisting of at least three amino acid residues from a selected region, calculating the binding coefficient of the MHC class II molecule for each selected segment and identifying at least one of the selected segments suitable for modification based on the calculated display MHC molecules bind the body of class II for this segment. The total binding rate of the MHC class II molecule for the peptide can then be changed without significantly reducing the therapeutic value of the protein.

The binding rate of a MHC class II molecule for a selected segment of an amino acid residue in one aspect of the present invention is calculated by summing certain values for each hydrophobic amino acid residue of a side chain present in a selected segment of the amino acid residue of a peptide. To produce a graphical overview, the value of this sum can then be assigned to a single amino acid residue at approximately the midpoint of the segment. This procedure is repeated for each of the overlapping segments (windows) in the region of interest or regions of the peptide. The determined value for each aromatic side chain present is approximately half the determined value for each hydrophobic amino acid residue of the side chain. The hydrophobic amino acid residues of the side chain are those that are present in valine, leucine, isoleucine and methionine. Aromatic side chains are those that are present in phenylalanine, tyrosine and tryptophan. The preferred specific value for the aromatic side chain is approximately 1 and for the hydrophobic amino acid residue of the side chain is approximately 2. However, other values may be used.

Thus, in a first aspect, the invention provides a computational method suitable for identifying one or more potential T cell epitopes of peptides within an amino acid sequence of a biological molecule by steps involving determining the binding of said peptides to MHC molecules using in vitro or in silico methods or biological assays , the specified method includes the following stages:

(a) selecting a portion of the peptide that has a known sequence of amino acid residues;

(b) sequential selection of overlapping segments of amino acid residues with a predetermined equal size and consisting of at least three amino acid residues of the selected site;

(c) calculating the binding index of the MHC class II molecule for each indicated selected segment by summing certain values for each hydrophobic amino acid residue of the side chain that is present in each selected segment of the amino acid residue; and

(d) identifying at least one of these segments suitable for modification based on the calculated binding rate of the MHC class II molecule of this segment to change the overall binding rate of MHC class II for the peptide without significantly reducing the therapeutic usefulness of the peptide.

In a specific embodiment, the invention relates to a method in which step (c) is performed using a Bohm counting function modified to include a 12-6 ligand-van der Waals ligand-protein complex repulsive energy component and a ligand conformational energy component by:

(1) providing a first database of MHC class II molecule models;

(2) providing a second database of a possible peptide skeleton for these models of MHC class II molecules;

(3) selecting a model from said first database;

(4) selecting a possible peptide skeleton from a second database;

(5) identification of the side chains of amino acid residues that are present in each selected segment;

(6) determining the value of the binding affinity for all side chains that are present in each selected segment; and optional

(7) repeating steps (1) to (5) for each indicated model and each indicated skeleton.

In a further embodiment, the binding rate for each selected sequence is calculated by (i) providing a first database of MHC class II molecule models; (ii) providing a second database of a possible peptide skeleton for molecular models of said MHC class II; (iii) providing a third database of possible side chain amino acid conformations for each of these twenty amino acids at each position of each skeleton; (iv) selecting a model from said first database; (v) selecting a possible peptide skeleton from a second database; (vi) identifying side chains of amino acid residues that are present in each selected segment along with their possible conformations from said third database; (vii) determining the optimal binding affinity for all side chains that are present in each selected segment in each possible conformation; (viii) repeating steps (v) to (vii) for each indicated skeleton and determining an optimal binding index; and (ix) repeating steps (iv) to (viii) for each indicated model.

It should be understood that the three databases described above can be combined into one database, or any two databases can be combined to provide a combined database.

The length of the segments of the amino acid residue to be selected may vary. Preferably, the selected amino acid residue segments are comprised of from about 10 to about 15 amino acid residues, more preferably about 13 amino acid residues.

Selected segments of amino acid residues may overlap to varying degrees. Preferably, the selected segments of amino acid residues overlap substantially. Most preferably, consecutively selected segments of amino acid residues overlap with all but one amino acid residue. That is, in an amino acid residue segment having n residues, n-1 residues overlap with the next sequentially selected amino acid residue segment.

Thus, in more detail, the invention also relates to the following preferred embodiments:

- according to the specified method, in which the specified value for each aromatic side of the chain is approximately half the specified value for each hydrophobic aliphatic side of the chain;

- according to the specified method, in which the selected segment of the amino acid residue is composed of 13 amino acid residues;

- according to the specified method, in which successive selected segments of the amino acid residue overlap one to five amino acid residues;

- according to the specified method, in which successive selected segments of the amino acid residue overlap significantly;

- according to the specified method, in which all but one of the amino acid residues in consecutive selected segments of the amino acid residue overlap.

In a second basic aspect, the present invention provides modified forms of various biological molecules with one or more deleted T cell epitopes, where said modification can be achieved by the methods described above and in the claims. Molecules can also be prepared by methods as described in the aforementioned prior art, however, molecules obtained by the methods of the present invention show improved properties. In prior art methods, predicted T-cell epitopes are isolated using appropriate amino acid substitutions in the primary sequence of a therapeutic antibody or protein that is not an antibody of either human or non-human origin.

The present invention provides modified forms of proteins and immunoglobulins that are expected to exhibit improved in vivo properties.

Therefore, the aim of the invention is to provide a method for producing an immunogenically modified biological molecule derived from a parent molecule, wherein the modified molecule has an amino acid sequence different from the sequence of the specified parent molecule, and shows reduced immunogenicity relative to the parent molecule when exposed to an immune system of a particular kind; said method includes: (i) determining the amino acid sequence of a parent biological molecule or part thereof; (ii) identifying one or more potential T cell epitopes within the amino acid sequence of the protein by any method, including determining the binding of peptides to MHC molecules using in vitro or in silico methods or biological assays; (iii) the development of new variants of the sequence by changing at least one amino acid residue within the originally identified sequences of T-cell epitopes, these variants are modified in such a way that they significantly reduce or eliminate the activity or a certain number of sequences of T-cell epitopes and / or a certain number of MHC allotypes capable of binding peptides derived from the specified biological molecule, as determined by binding the peptide to MHC molecules using the use of in vitro or in silico methods or biological assays or the binding of peptide-MHC complexes to T cells, (iv) constructing such sequence variants using recombinant DNA techniques and testing these variants in order to identify one or more variants with desired properties; and (v) optionally repeating steps (ii) to (iv), where the identification of T cell epitope sequences according to step (ii) is achieved by the method as described above and below.

Specific embodiments of step (iii) of the invention relate to the following summarized steps:

- according to the specified method, in which 1-9 amino acid residues in any of the initially present epitopes of T cells are changed;

- according to the specified method, in which one amino acid residue in any of the initially present epitopes of T cells is changed;

- according to the specified method, in which the amino acid change is performed in relation to the homologous sequence of the protein and / or in accordance with in silico modeling methods.

- according to the specified method, in which the change in amino acid residues is the replacement, deletion or addition of the initially present (s) amino acid (s) residue (s) to another (s) amino acid (s) residue (s) in a specific position (s) )

- according to the specified method, in which further further change, preferably by substitution, addition or deletion of certain amino acid (s), is carried out to restore the biological activity of the specified biological molecule.

With the exception of step (ii), other steps of the disclosed method can be achieved by methods and methods that are well known to those skilled in the art. Since modified biological molecules are obtained, preferably by recombinant methods, of the corresponding DNA structures that were deduced from the amino acid sequence after completion of the exchange of amino acid residues identified by the method of step (i). Recombinant methods used herein are known in the art (e.g., Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory Press, NY, USA).

The biological molecule obtained according to the invention is preferably a peptide, protein, antibody, antibody fragment or fusion protein. In addition, the invention includes modifications, variants, mutations, fragments, derivatives, non-, partially or fully glycosylated forms of these molecules having the same or similar biological and / or pharmacological activity.

Although the method disclosed in this invention is not limited to specific biological molecules, a specific embodiment of the invention is to provide preferably molecules that are known in the art and to demonstrate therapeutic usefulness and value. Thus, an additional object of the invention is to provide an immunogenically modified biological molecule derived from a parent molecule, wherein the modified molecule has an amino acid sequence different from the sequence of the specified parent molecule, and shows reduced immunogenicity relative to the parent molecule when exposed to an immune system of a particular kind obtained by the method according to invention, as described in detail above and below.

Biological molecules of particular interest obtained by this method are selected from the groups:

(a) monoclonal antibodies:

anti-40kD glycoprotein antigen KS 1/4 antibody;

anti-GD2 antibody 14.18;

anti-Her2 antibody 4D5 (murine) and a humanized version (Herceptin®);

anti-Her1 (EGFR) antibody c225 and h425;

anti-IL-2R (anti-TAC) antibody (Zenapax®);

anti-CD52 antibody (SAMPATH®);

anti-CD20 antibodies (C2B8, Rituxan®; Bexxar®);

an antibody directed to the C5 complement human protein;

(b) human proteins:

sTNF-R1, sTNF-R2, sTNFR-Fc (Enbrel®);

protein C, acr30, ricin A, CNTFR ligands;

subtilisin, GM-CSF, human follicle-stimulating hormone (h-fsh);

β-glucocerebrosidase, GLP-1, apolipoprotein A1;

leptin (human obesity protein), KGF, GM-CSF;

BDNF, EPO, I1-1R antagonist.

A third main aspect of the present invention relates to sequences of T cell epitopes that are derived from parental immunogenically unmodified biological molecules. These epitopes are preferably 13-dimensional peptides. Within these peptides, sequences having 9 consecutive amino acid residues are preferred. Thus, another object of the invention is to provide access to such epitopes and sequences. In more detail, the invention relates to:

- the use of a potential T-cell epitope of the peptide within the amino acid sequence of the parent immunogenically unmodified biological molecule identified according to any of the described methods for producing a biological molecule with reduced immunogenicity having the same biological activity;

- appropriate use of a potential T-cell epitope of a peptide, wherein said T-cell epitope is a 13-dimensional peptide;

- the use of a peptide sequence consisting of at least 9 consecutive amino acid residues of a 13-dimensional T cell epitope, as described above, to obtain a biological molecule with reduced immunogenicity having the same biological activity compared to the parent unmodified molecule.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 is a flowchart illustrating a first aspect of the present computational method;

FIGURE 2 is a block diagram illustrating database generation for a computational method embodied by the present invention;

FIGURE 3 is a flowchart illustrating a database query for creating a peptide template for potential T cell epitopes;

FIGURE 4 is an additional flowchart illustrating a computational method.

FIGURE 5 is a graph of the probability index of the T-cell epitope versus coordinates (positions) of the amino acid residue of glutamic acid decarboxylase (MW: 65000) isoforms (GAD 65);

FIGURE 6 is a graph of the probability index of the epitope of T cells versus coordinates (positions) of the amino acid residue of erythropoietin (EPO);

FIGURE 7 is a graph of the probability index of the T-cell epitope versus coordinates (positions) of the amino acid residue of the light chain of a humanized anti-A33 monoclonal antibody; and

FIGURE 8 is a graph of the probability index of the T-cell epitope versus coordinates (positions) of the amino acid residue of the heavy chain of a humanized anti-A33 monoclonal antibody.

In the above FIGURES 5-8, the solid line (-) represents the T-cell epitope index calculated by the computational method in accordance with the block diagram shown in FIGURE 1, and the dashed line (.....) represents the estimated number of T- epitopes cells, calculated in accordance with the computational method in accordance with the flowchart shown in FIGURE 3, in accordance with another aspect of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The term "T cell epitope" means, in accordance with the understanding of the present invention, an amino acid sequence that is capable of binding to acceptable effective class II MCH molecules (or their equivalents in non-human species), capable of stimulating T cells and / or also binding (without the inevitable a certain degree of activation) with T cells in combination with MCH class II.

The term "peptide", as used in the context of this application and the attached claims, is a compound that includes two or more amino acids. Amino acids bind together using a peptide bond (defined below). There are 20 different natural amino acids involved in the biological production of peptides, and any number of them can bind in any order to form a peptide chain or ring. The naturally occurring amino acids used in the biological production of peptides all have an L configuration. Synthetic peptides can be obtained using acceptable synthesis methods using L-amino acids, D-amino acids, various combinations of amino acids of two different configurations. Some peptides contain only a few amino acid units. Short peptides, for example, having less than ten amino acid units, are sometimes called "oligopeptides." Other peptides contain a large number of amino acid residues, for example up to 100 or more, and are called "polypeptides". By agreement, a "polypeptide" can be considered any peptide chain that contains three or more amino acids, while an "oligopeptide" is usually considered a particular type of "short" polypeptide. Thus, as used in the context of this application, it is understood that any reference to a “peptide” also includes an oligopeptide. In addition, any reference to a “peptide” includes polypeptides, oligopeptides and proteins. Each different distribution of amino acids forms different polypeptides or proteins. The number of polypeptides, and hence the number of different proteins that can be formed, is virtually unlimited.

The term “reduced or decreased immunogenicity”, used earlier, and hereinafter, is a relative term and refers to the immunogenicity of the corresponding original parent molecule when it is exposed in vivo to the same species compared to a molecule modified according to the invention.

The term "modified protein", as used in the context of this invention, describes a protein that has a reduced number of T-cell epitopes and therefore exhibits reduced immunogenicity compared to the parent protein when exposed to the immune system of certain species. The term "unmodified protein", as used in this invention, describes a "parent" protein, which in comparison with a "modified protein" has a greater number of T-cell epitopes and therefore increased immunogenicity relative to the modified protein when exposed to the immune system of certain species.

"Alpha carbon (Cα)" is the carbon atom of a hydrocarbon (CH) component that is in the peptide chain. A “side chain” is a suspended group to Cα, which may include a simple or complex group or residue, and which has physical dimensions that may differ significantly from the size of the peptide.

T cell epitopes can be identified by the computational method of the present invention by considering amino acid residues that are important for binding a specific T cell epitope to MHC class II molecules. Once identified, potential T cell epitopes can be isolated or removed from the amino acid residue sequence by altering, such as mutating, the key amino acid residues in that sequence. Any modification made to a peptide sequence in a region that is likely to contain T cell epitopes by deletion, addition, or substitution leading to a relatively lower overall binding index will result in a less immunogenic amino acid residue sequence. In some cases, it may be desirable to increase the binding of certain peptides to MHC class II molecules. For example, it has been suggested that tolerance to certain autoantigens can be restored in individuals suffering from an autoimmune disease if such individuals are treated with peptide analogues of autoantigen regions that are known to contain T-cell epitopes. The natural epigon usually has a moderate affinity for MHC class II molecules, while the peptide analog is made so that it has a relatively higher affinity for MHC class II molecules. This high affinity is important both for promoting immune surveillance for the purification of such T cells representing this high affinity for the epitope, and for them to become anergic. Such a modification in T cell epitopes can also be done at the protein level of the peptide, and the whole protein is used as a therapeutic agent.

There are a number of factors that play an important role in determining the overall structure of a protein or polypeptide. First, the peptide bond, that is, the bond that binds the amino acids together in a chain, is a covalent bond. This bond is flat in structure, essentially substituted by an amide. "Amide" is any group of organic compounds containing a moiety

A flat peptide bond connecting the Cα of neighboring amino acids can be represented as shown below:

Since O = C and C-N atoms lie in a relatively motionless plane, free rotation does not occur around these axes. Therefore, the plane shown schematically by the dashed line is sometimes referred to as the "amide" or "peptide plane", which contains the atoms of oxygen (O), carbon (C), nitrogen (N) and hydrogen (H) of the peptide skeleton. In opposite corners of this amide plane, Cα atoms are located. Since there is no significant rotation around the O = C and C-N atoms in the peptide or amide plane, the polypeptide chain thus includes a series of planar peptide bonds connecting the Cα atoms.

The second factor that plays an important role in determining the overall structure or conformation of a polypeptide or protein is the angle of rotation of each amide plane around the common Cα bond. The terms "rotation angle" and "twist angle" are used hereinafter as equivalent terms. Assuming that the O, C, N, and H atoms remain in the amide plane (which is usually the real assumption, although there may be some slight deviations from the plane of these atoms for some conformations), these rotation angles determine the N and R conformation of the polypeptide skeleton , that is, the structure in the form in which it exists between adjacent residues. These two scaffolds are known as φ and ψ. The set of angles φ ₁ and ψ ₁ , where index ₁ represents a separate residue of the polypeptide chain, thus, effectively defines the secondary structure of the polypeptide. The conventions used in determining the φ and ψ angles, i.e., the base points at which the amide plane forms an angle equal to zero degrees, and the definition of what is the angle φ and what is the angle ψ for a given polypeptide is given in the literature. See, for example, Ramachandran et al. Adv.Prot.Chem. 23: 283-437 (1968), namely, on pages 285-94, which are incorporated herein by reference.

This method can be applied to any protein and is based in part on the discovery of the fact that in humans the anchor position of the primary pocket 1 of the binding grooves of the MHC class II molecule has a clearly corresponding specificity for certain amino acid side chains. The specificity of this pocket is determined by the amino acid identity at position 86 of the beta chain of the MHC class II molecule. This site is located at the bottom of pocket 1 and determines the size of the side chain that can be accommodated by this pocket (Marshall, K.W., (1994), J. Immmunol., 152: 4946-4956). If this residue is glycine, then all hydrophobic aliphatic and aromatic amino acids (hydrophobic aliphatic amino acids are: valine, leucine, isoleucine, methionine, and aromatic are phenylalanine, tyrosine and tryptophan) can be accommodated in your pocket, aromatic side chains are preferred. If this pocket residue is valine, then the side chain of this amino acid extends into the pocket and limits the size of the side chains of the peptide that can undergo accommodation in such a way that only hydrophobic aliphatic side chains can undergo accommodation. Thus, in the sequence of amino acid residues where an amino acid with a hydrophobic aliphatic or aromatic side chain is found, there is potential for the present limited epitope of T-cells of class II MHC. Even if the side chain is hydrophobic aliphatic, nevertheless, it is approximately two times more associated with the T-cell epitope than the aromatic side chain (even assuming an approximate distribution of pocket 1 types in the general population).

A computational method for embodying the parameters of the present invention regarding the likelihood of peptide sites that contain T-cell epitopes is as follows:

(1) The primary sequence of the peptide segment of a predetermined length is examined and all hydrophobic aliphatic and aromatic side chains present are identified. (2) Hydrophobic aliphatic side chains are rated greater than aromatic side chains; approximately twice as large as for aromatic side chains, for example, a value of 2 for a hydrophobic aliphatic side chain and a value of 1 for an aromatic side chain. (3) The determined values are summed for each overlapping segment of amino acid residues (window) of a predetermined constant length within the peptide, and the total value for the private segment (window) is determined for one amino acid residue in the intermediate position of the segment (window), preferably for the residue that located in the center of the selective segment (window). This procedure is repeated for each selective overlapping segment of amino acid residues (windows). Thus, each amino acid residue of a peptide is determined by a value that is related to the probability of the T-cell epitope that is present in this particular segment (window). (4) Calculated and determined as described in Step 3, the values can be plotted against the amino acid coordinates of the whole amino acid sequence that is being analyzed. (5) All parts of the sequence that have a predetermined value count, for example, a value of 1, are probably considered to be those that contain an epitope of T cells and can be modified if desired. This particular aspect of the invention provides a general method by which portions of a peptide probabilistically containing T cell epitopes can be described. Peptide modifications in these regions have the potential to modify the binding characteristics of MHC class II.

In accordance with another aspect of the present invention, T cell epitopes can be predicted with greater accuracy using a more sophisticated computational method that takes into account interactions of peptides with class II MHC alleles.

Computational prediction of T cell epitopes present within a peptide in accordance with a particular aspect of the invention considers the construction of models of at least 42 class II MHC alleles based on the structures of well-known class II MHC molecules and the method for using these models in computational identification T-cell epitopes, constructs of peptide skeleton libraries for each model in order to be able to known variability in the related positions of the alpha-carbon (Cα) peptide of the skeleton, the design of the libraries of amino acid side chain structures for each skeleton with each model for each of the 20 amino acid alternatives in critical positions for each interaction between the peptide and the MHC class II molecule, the use of these skeleton libraries and side chain structures in combination with the counting function to select the optimal skeleton and structure of the side chain for a private peptide docked with a private class II MHC molecule, and obtaining the binding value from the results of this interaction.

Models of MHC class II molecules can be obtained by homologous modeling from among similar structures found in the Brookhaven Protein Data Bank (“PDB”). This can be done using semi-automatic homologous modeling software (Modeller, Sali A. & Bhmdell TL., 1993. J. Mol. Biol. 234: 779-815), which introduces the reproducing annealing function in conjunction with the CHARMm force field to minimize energy (available from Molecular Simulations Inc., San Diego, CA). Alternative modeling techniques may also be used.

The present method is significantly different from other computational methods that use libraries of experimentally obtained data regarding the binding of each amino acid alternative at each position of the binding groove for a small set of MHC class II molecules (Marshall, KW et al., Biomed. Pept. Proteins Nucleic Acids, 1 ( 3): 157-162) (1995)) or even other computational methods that use simple experimental binding data to determine the binding characteristics of particular types of binding pockets within grooves, again using relatively small subsets of MHC class II molecules and then “mixing and selecting” pocket types from this pocket library to artificially create further “virtual” MHC class II molecules (Sturniolo T. et al., Nat. Biotech. 17_ ( 6): 555-561 (1999)). Both previous methods suffer from the main disadvantages, which, due to the complexity of the analyzes and the need to synthesize a large number of peptide variants, make it possible to experimentally study only a small number of MHC class II molecules. Thus, the first prior art method can only make it possible to predict properties for only a small number of MHC class II molecules. The second method of the prior art also makes it possible to make an assumption that a pocket coated with similar amino acids in one molecule will have the same binding characteristics in the context of different class II alleles and also suffer from the same disadvantages as those class II MHC molecules that can be virtually created and which contain pockets contained within a library of pockets. Using the modeling approach described in this application, the structure of any number and type of studied class II molecules of MHCs can be deduced, and thus alleles can be specifically selected as representatives of the general population. In addition, the number of MHC class II molecules studied can be increased by creating additional models to obtain additional data through complex experiments.

The use of a skeleton library provides for variations in the positions of the Cα atoms of various studied peptides when docked with MHC class II private molecules. This again differs from the alternative prior art computational methods described above, which rely on the use of simplified peptide skeletons to study amino acid binding in private pockets. These simplified skeletons are probably not representative skeleton structures that are found in “real” peptides, which leads to errors in predicting peptide binding. This skeleton library is created by superimposing the skeletons of all peptides associated with the MHC class II molecules found within the protein database and by fixing the initial value of the deviation area (RMC) between the Cα atoms of each of the eleven amino acids located within the binding groove. Since this library can be obtained from a small number of acceptable mouse or human structures (currently 13), in order to allow for even greater variability, the quantitative RMS data for each C "-α position is increased by 50%. Then, the average value is determined the position Cα of each amino acid and draw a sphere around this point, the radius of which is equal to the deviation of the RMS at this position plus 50%. This sphere represents all possible Cα positions. Using Cα with at least a deviation RMS (those amino acids in pocket 1, as mentioned above, which are equivalent to the position 2 of eleven residues in the binding groove), the sphere has a spatial reference, and each vertex within the lattice is then used as a possible location for the Cα amino acid. The resulting amide plane corresponding to the peptide, associated with the subsequent amino acid, moves to each of these Cα, and the angles φ and ψ are rotated step by step through a series of intervals in order to move subsequent Cα. If consecutive Cα falls within the “sphere of permissible positions” for these Cα, then the orientation of the dipeptide is accepted, while if it falls outside the sphere, the dipeptide is rejected. This process is then repeated for each subsequent Cα position so that the dipeptide grows from the initial Cα pocket 1, until all 9 subsequent Cα are positioned from all possible permutations of the previous Cα. The process is then repeated once again for one Cα of the previous pocket 1 to create a library of Cα skeleton positions located within the binding groove. The number of skeletons obtained depends on several factors: the size of the "sphere of possible positions"; binding qualities of the "primary sphere" in the position of pocket 1; qualities of stepwise rotation of the angles φ and ψ used to position subsequent Cα. Using this process, a large skeleton library can be created. For a larger skeleton library, it is more likely that it will be such that the optimal fit is found for the private peptide within the groove of the binding of the MHC class II molecule. Since all skeletons will not be acceptable for docking with all models of MHC class II molecules due to disharmony with the amino acids of the binding domains, a subset of the library is created for each allele, including skeletons that can be accommodated by these alleles. Using a library of skeletons in combination with models of MHC class II molecules creates an exhaustive database of possible side chain structures for each amino acid at each position of the binding groove for each MHC class II molecule that fits with every possible skeleton. This dataset is obtained using the simple steric overlap function, where a MHC class II molecule joins the skeleton and the amino acid side chain moves to the skeleton at the desired position. Each of the side chain bonds capable of rotation rotates step by step through a set of intervals, and the obtained atom positions depend on the recorded bond. The interaction of the atom with the atoms of the side chains of the connecting groove is recorded and the positions are accepted or rejected depending on the following criteria: the total sum of the overlap of all atoms positioned to date, should not exceed a predetermined value. Thus, the stiffness of the conformational search is a function of the interval used in stepwise rotation of the bond and a predetermined boundary for general overlap. Thus, the latter value may be small if it is known that a particular pocket is stiff, but stiffness may be weakened if it is known that the positions of the side chains of the pocket are relatively flexible. Thus, an assumption can be made to simulate flexibility variations within the pockets of the bonding groove. This conformational search is then repeated for each amino acid in each position of each skeleton when docked with each of the MHC class II molecules to create a comprehensive database of side chain conformations.

An acceptable mathematical expression is used to estimate the binding energy between models of MHC class II molecules in combination with conformations of petid ligands, which are empirically obtained by examining the large database of skeleton / side chain conformations described above. Thus, the protein is tested for potential T-cell epitopes by examining each possible peptide length, which varies between 9 and 20 amino acids (despite the fact that the length remains constant for each study) for the following calculations: a class II molecule of MHC is selected together with the peptide skeleton possible for this molecule and side chains corresponding to the desired peptide sequence to which they are transferred. Atomic identity and interatomic distance data related to a particular side chain at a specific position in the skeleton are collected for each possible conformation of this amino acid (obtained from the database described above). This is repeated for each side chain along the skeleton and peptide values are obtained using the count function. The best value for this skeleton is remembered and the process is repeated for each possible skeleton for the selected model. The values of all possible skeletons are compared, and the highest value is considered as the peptide value for the desired peptide in this MHC class II model. This process is then repeated for each model with each possible peptide that is derived from the test protein, and the values for the peptides against the models are shown.

In the context of this invention, each ligand presented for counting binding affinity is an amino acid segment selected from the group of peptides or proteins discussed above. Thus, a ligand is a selected amino acid chain that contains from about 9 to 20 amino acids in length and which is derived from a peptide, polypeptide or protein of a known sequence. The terms "amino acids" and "residues" in this application are used as equivalent terms. The ligand in the form of consecutive amino acids of the peptide that is being studied is transferred to the skeleton from the skeleton library, positioned in the binding groove of the MHC class II molecule from the MHC class II molecule model library through the C ”coordinates of the peptide skeleton atoms, and the possible structures for each side chain are selected from databases of possible conformations Identities of the relevant atoms and interatomic distances are also extracted from this database and used to calculate the peptide binding index. MNCs are indicated as candidates for site-directed mutagenesis by the affective affinity for the class II binding pocket. Amino acid substitutions are made in the marked ligands (and thus in the protein that is of interest and which is then re-tested using the counting function to determine the changes which reduce binding affinity below a predetermined value.) These changes can then be introduced into the protein of interest to remove T cell epitopes. The binding between the peptide ligand and the groove of binding of class II MHC molecules involves non-covalent interactions, including but not limited to the following: hydrogen bonds, electrostatic interactions, hydrophobic (lipophilic) interactions, and Van der Waals interactions. This is included in the peptide counting function as described in more detail below. It is clear that the hydrogen bond is a non-covalent bond that can form between polar or charged groups and consists of a hydrogen atom divided between two other atoms. The hydrogen of the hydrogen donor has a positive charge, while the hydrogen acceptor has a partial negative charge. For the purposes of peptide / protein interactions, the hydrogen bonding donors may be either nitrogen attached or hydrogen attached to oxygen or nitrogen. The hydrogen bond acceptor may be oxygen, not attached to hydrogen, nitrogen without attached hydrogen and one or two bonds, or sulfur with only one bond. Certain atoms, such as oxygen attached to the hydrogen or nitrogen of an imine (for example, C = NH) can be both acceptors and hydrogen donors. The hydrogen bond energy ranges from 3 to 7 Kcal / mol and is stronger than the van der Waals bonds, but weaker than the covalent bonds. Hydrogen bonds are also highly directed and are most strong when the donor atom, the hydrogen atom and the acceptor atom are colinear. Electrostatic bonds are formed between oppositely charged pairs of ions, and the interaction force is inversely proportional to the square of the distance between the atoms in accordance with the Coulomb law. The optimal distance between the pairs of ions is approximately 2.8 Å. In protein / peptide interactions, electrostatic bonds can form between arginine, histidine or lysine and aspartate or glutamate. The strength of the bond will depend on the pKa of the ionizing groups and the dielectric constant of the medium, although they are approximately similar in strength to the hydrogen bonds.

Lipophilic interactions are favorable hydrophobic-hydrophobic contacts that arise between a protein and a peptide digand. Usually they arise between the hydrophobic side amino acid chains of the peptide hidden inside the pockets of the binding groove so that they are not exposed to the solvent. The susceptibility of hydrophobic residues to the action of the solvent is extremely unfavorable, since the surrounding solvent molecules affect the hydrogen bond with other cell-like clathrate-forming structures. The resulting decrease in entropy is extremely unfavorable. Lipophilic atoms can be sulfur, which is neither a polar nor a hydrogen acceptor, and carbon atoms that are not polar.

Van der Waals bonds are nonspecific forces existing between atoms that are at a distance of 3-4Å. They are weaker and less specific than hydrogen and electrostatic bonds. The distribution of the electron charge around the atom changes with time, and in any case, the charge distribution is not symmetrical. This short-term asymmetry in the electron charge induces a similar asymmetry in neighboring atoms. The forces of attraction between the atoms resulting from this peak at a contact distance of Van der Waals, but very quickly decrease at a distance of from about 1 Å to about 2 Å. Conversely, as soon as atoms become separated by less than contact distance, to a large extent strong repulsive forces become dominant, while the external electron clouds of atoms overlap. Although attractive forces are relatively weaker than electrostatic and hydrogen bonds (approximately 0.6 Kcal / mol), repulsive forces, in particular, may be more important in determining whether a peptide ligand will successfully bind to a protein.

In one embodiment, the Böhm exponent function (SCORE 1 approach) is used to evaluate binding contact (Böhm HJ, J.Comput AidedMol. Des 8 (3): 243-256 (1994), reference incorporated herein in its entirety). In another embodiment, an indicator function (SCORE2 approach) is used to evaluate binding affinity as an indicator of a ligand-containing T cell epitope (Böhm HJ, J.Comput Aided Mol. Des. 12 (4): 309-323 (1998), link cited in this application in its entirety). However, the Böhm evaluation function, as described in the above references, is used to evaluate the binding affinity of a ligand with a protein, where it is known in advance that the ligand successfully binds to a protein and the protein / ligand complex has a specific structure, a specific structure is presented in the protein database ("PDB "). Thus, the counting function was further developed with the advantage of known positive binding data. In order to be able to distinguish between positive and negative binders, the repulsive component may not necessarily be added to equality. In addition, a more satisfactory estimate of the binding energy is achieved by calculating lipophilic interactions in a pairwise manner, more accurately than using an area based on the energy component of the reduced Böhm functions. Thus, in a preferred embodiment, the binding energy is estimated using the modified Böhm counting function. In the modified Böhm counting function, the binding energy between the protein and the ligand (ΔG _bind ) is estimated by taking into account the following parameters: a decrease in the binding energy due to the absolute loss of translational and rotational entropy of the ligand (ΔG ₀ ); contributions of ideal hydrogen bonds (ΔG _hb ), where at least one partner is neutral; contributions of unperturbed ionic interactions (ΔG _ionic ); lipophilic interactions between lipophilic ligand atoms and lipophilic acceptor atoms (ΔG _lipo ); loss of binding energy due to solidification of the internal degrees of freedom in the ligand, that is, freedom of rotation around each CC bond is reduced (ΔG _rot ); the interaction energy between a protein and a ligand (E _VdW ). The analysis of these components gives equality 1:

(ΔG _bind) = (ΔG _{0) + (ΔG hb × N hb} ) + (ΔG ionic × N ionic) + (ΔG lipo × N lipo) + (ΔG rot + N rot) + (E VdW).

where N represents the number of established interactions for a specific component and in one embodiment, ΔG ₀ , ΔG _hb , ΔG _ionic , ΔG _lipo and ΔG _rot are constants that have values: 5.4, -4.7, -4.7, -0.17, and 1.4, respectively.

Component Nhb is calculated in accordance with equality 2:

N _hb = Σ _h-bond f (ΔR, Δα) × f (N _neighb ) × f _pcs

f (ΔR, Δα) is a penalty function that is calculated for large deviations of hydrogen bonds from ideal bonds and is calculated in accordance with equality 3:

f (ΔR, Δ-α) = f1 (ΔR) × f2 (Δα)

Where: f1 (ΔR) = 1 if ΔR <= TOL

or = 1- (ΔR-TOL) / 0.4, if ΔR <= 0.4 + TOL

or = 0 if ΔR> 0.4 + TOL

And: f2 (Δα) = 1 if Δα <30 °

or = 1- (Δα-30) / 50, if Δα <= 80 °

or = 0 if Δα> 80 °

TOL represents the permissible deviation in the length of the hydrogen bond = 0.25Å.

ΔR represents the deviation of the length of the H-O / N hydrogen bond from the ideal value = 1.9 Å.

Δα is the deviation of the angle of the hydrogen bond ∠ _{N / OH..O / N} from the ideal value of 180 °.

f (N _neighb ) distinguishes between the concave and curved parts of the surface of the protein and, thus, sets a greater weight for the polar interactions found in the pockets, more than those for those found on the surface of the protein. This function is calculated in accordance with equality 4 below:

f (N _neighb ) = (N _neighb / N _{neighb, 0} ) ^α _, where α = 0.5

N _neighb is the number of non-hydrogen atoms of a protein that are closer than 5Å to any protein atom.

N _neighb , ₀ is a constant equal to 25.

f _pcs is a function that is valid for the area of the polar contact surface per hydrogen bond and, thus, distinguishes between strong and weak hydrogen bonds, its value is determined in accordance with the following criteria:

f _pcs = β when A _polar / NHB <10Å ²

or f _pcs = 1 when A _polar / N _HB > 10Å ²

A _polar is the size of the area of the polar protein-ligand contact

N _HB is the number of hydrogen bonds

β is a constant whose value = 1.2

To implement the modified Böhm counting function, the contributions of ionic interactions, ΔG _ionic, are calculated in the same way as for the hydrogen bonds described above, since the same geometric dependence is assumed.

Component N _{lipo is} calculated in accordance with equality 5 below:

N _lipo = Σ _1L f (r _1L )

f (r _1L ) is calculated for all lipophilic ligand atoms, 1, and all lipophilic protein atoms, L, in accordance with the following criteria:

f (r _1L ) = 1 when r _1l <= R _1f (r _1L ) = (r _1l -R1) / (R2-R1) when R2 <r _1l > R1

f (r _1L ) = 0 when f _1l > = R2

where: R1 = r1 ^VdW + rL ^VdW +0.5

and R2 = R1 + 3.0

and: r ₁ ^VdW is the van der Waals radius of atom 1

and r _L ^VdW is the van der Waals radius of the atom L

The N _rot component is the number of rotatable amino acid side chain bonds and is taken to represent the number of acyclic sp ³ -sp ³ and sp ³ -sp ² bonds. Rotations of the terminal end —CH ₃ or —NH _{3 are} not taken into account.

The final component E _VdW is read in accordance with equality 6 below:

E _VdW = ε ₁ ε ₂ ((r1 ^VdW + r2 ^VdW ) ¹² / (r1 ^VdW + r ₂ ^VdW ) ⁶ / r ⁶ , where

where ε ₁ and ε ₂ are constants depending on the identity of the atoms

r ₁ ^VdW + r ₂ ^VdW are the atomic radii of Van der Waals

r represents the distance between a pair of atoms.

As for equality 6, in one embodiment the constants ε ₁ and ε ₂ have the following meanings for atoms: C: 0.245, N: 0.283, O: 0.316, S: 0.316, respectively (i.e., carbon, nitrogen, oxygen and sulfur atoms, respectively) . As for equalities 5 and 6, the Van der Waals radii have the following meanings for atoms: C: 1.85, N: 1.75, O: 1.60, S: 2.00 Å.

It is clear that all the predefined values and constants given in the equalities above are defined within the limited conditions for understanding protein-peptide interactions in a particular connection with the type of calculation that is carried out in this application. Thus, it is possible that this counting function will be further improved as a result of progress in the field of modeling molecular interactions, these values and constants can change, therefore, any acceptable numerical value that gives the desired results in the components of the estimation of the binding energy of a protein with a ligand can be used and therefore falls within the scope of this invention.

As described above, the counting function is applied to data retrieved from the database of side chain conformations, identical atoms, and interatomic distances. For the purposes of this description, the number of MHC class II molecules included in this database is 42 models plus 4 established structures. From the above description it is clear that the modular nature of the construction of the computational method in accordance with this invention means that new models can simply be added and explored together with a library of peptide skeletons and a function for searching for side chain conformations to create additional data sets that can be processed using peptide counting function as described above. This makes it possible to easily increase the set of studied class II MHC molecules or structures and associated data, which are replaced if the data is available to create more accurate models of existing alleles.

It is clear that although the counting function described above is relatively simple compared to some complex methodologies that are available, the calculations are extremely fast. It is understood that the goal is not to calculate the actual binding energy per se for each peptide that mates with the binding groove of the selected MHC class II protein. The main objective is to obtain binding energy data for applicants as a goal for predicting the placement of T-cell epitopes based on the primary structure (i.e., amino acid sequence) of the selected protein. The relatively high binding energy or binding energy of the selected value suggests the presence of T-cell epitopes in the ligand. The ligand can then be subjected to at least one circle of amino acid substitutions and recalculated binding energy. Due to the speed of counting, this processing of the peptide sequence can be carried out using computer technology within the user interface of the program on effective accessible computer technology. The main investment in computer technology, therefore, is not required.

Any person skilled in the art will recognize that other available software can be used for the same purpose. In particular, more sophisticated software that is capable of linking ligands to protein-binding sites can be used in conjunction with energy minimization. Examples of docking software products are: DOCK (Kuntz et al., J. Mol. Biol., 161: 269-288 (1982)), LUDI (Böhm HJ, J. Comput Aided Mol. Des., 8: 623-632 (1994) and FLEXX (Rarey M. et al. ISMB, 3: 300-308 (1995)) Examples of molecular modeling and manipulation software include: AMBER (Tripos) and CHARMm (Molecular Simulations Inc.). Using these computational methods will strictly limit the capacity of the method in accordance with this invention due to the length of processing time required for the necessary calculations. th, that such methods may be used as a 'secondary screen' to obtain more accurate calculations of binding energy for peptides which are defined as 'positive binders' by using the method according to the invention.

The limitation of processing time for complex molecular-mechanical or molecular-dynamic calculations is one of the reasons that is determined by the software and the capabilities of modern computer equipment technology. It can be foreseen that in the future, with writing more rational code and a continuing increase in the speed of computer processors, it will become possible to carry out such calculations within a more manageable time frame. Further information regarding energy functions is applicable to macromolecules and a discussion of the various interactions that take place within the assembled protein structure can be found in: Brooks B.R. et al. J.Comput Chem., 4: 187-217 (1983) and further information regarding general protein-ligand interactions can be found in: Dauber-Osguthrpe et al., Proteins 4 (I): 31-47 ( 1988), these references are incorporated into this application in their entirety. Useful information can also be found, for example, in Fasman G. D. ed .. Prediction of Protein Structure of Protein Conformation, Plenum Press, New York, ISBN: 0-306, 4313-9.

The present prediction method can be calibrated in accordance with a data set including a large number of peptides whose affinity for various MHC class II molecules was previously experimentally determined.

According to a preferred embodiment of the method, any of the specific prediction methods described herein, go to any other computational method for predicting peptide-MHC class II interactions that results in numerical indicators for each peptide / MHC class II pair, is calibrated in accordance with a large data set the number of peptides whose affinity for various MHC class II molecules was previously experimentally determined. By comparing the calculated data and the experimental data, as defined above, it is possible to reduce the number of values if it is known that all experimentally determined epitopes of T cells were correctly predicted.

In particular, a computer-generated numerical indicator is calculated for each peptide / MHC class II pair in the data set. The rate is calculated such that a higher rate represents an increased likelihood of binding. The lowest computer-derived rate for the peptide / MHC class II pair, which was found experimentally for binding, was taken as the boundary. All computer-derived metrics that are significantly below this boundary metric are believed to represent non-binding peptide / MHC class II pairs, while computer-based metrics above the limiting metric represent a potentially binding peptide / MHC class II pair. In general, for this computer-based evaluation algorithm, there will be some peptide / MHC class II combinations that give indicators above the boundary indicator, but which are actually non-binding. Thus, this preferred embodiment of the method can produce false positives, but will never or only rarely produce false negatives.

This embodiment of the method based on the boundary value is especially useful when the goal is to eliminate by mutation the largest number or all of the T cell epitopes from the protein. In particular, according to a more preferred embodiment of the method of the invention, the largest number or all of the epitopes of T cells are removed from the protein as follows. The protein sequence is scanned by a computer algorithm for potential T-cell epitopes. Each potential T-cell epitope is assigned an indicator with increasing rates correlated with a higher probability of binding to MHC class II. Each peptide segment with an indicator greater than the boundary indicator is mutated so that the indicator of the mutated segment is less than the boundary indicator. Mutations are preferably chosen so as not to reduce the activity of the protein below the activity necessary for this purpose. A multiply mutated protein is developed that lacks mainly or generally computer-predicted T-cell epitopes. Such a multiply mutated protein is called "Deimmunized protein".

The de-immunized protein is synthesized by standard methods. For example, artificial DNA sequences encoding a De-immunized protein are assembled from synthetic oligonucleotides ligated into an expression vector and functionally linked to elements that promote the expression of the De-immunized protein. The de-immunized protein is then purified by standard methods. The resulting De-immunized protein contains amino acids mutated in such a way that the true T-cell epitopes are eliminated. In addition, a de-immunized protein will often contain mutated amino acids in the segments, which, as predicted by the algorithm, are T-cell epitopes, but which are not actually T-cell epitopes. However, mutations in falsely predicted epitopes do not result in significantly harmful consequences, because the mutations are chosen so that they have little effect on protein activity. In addition, the harmful effects do not follow from the possible introduction of new B-cell epitopes into the protein, because the lack of T-cell epitopes prevents the B-cell response to the modified protein.

The application of the above methodology to various peptides that can be considered for de-immunization, for modifications to improve the binding of MHC class II for therapeutic purposes, is illustrated below.

The invention can be applied to any biological molecule having a specific biological and / or pharmacological activity with substantially similar primary amino acid sequences as disclosed herein, and therefore will include molecules obtained by genetic engineering or other methods. The term "biological molecule" is used here for molecules that have a biological function and cause a biological, pharmacological or pharmaceutical effect or activity. Preferably, the biological molecules of the invention are peptides, polypeptides, proteins. Below proteins, immunoglobulins are preferred. The invention also includes variants and another modification of a particular polypeptide, protein, fusion protein, immunoglobulin, which in principle have the same biological activity and similar (reduced) immunogenicity. In addition, antibody fragments such as sFv, Fab, Fab ', F (ab') 2 and Fc, and biologically effective protein fragments are included. Antibodies of human origin or humanized antibodies show per se reduced immunogenicity or lack of immunogenicity in humans and do not have or have a lower number of immunogenic epitopes compared to non-human antibodies. However, there is also a need to de-immunize such molecules, some of which have shown that they elicit a significant immune response in humans. In addition, antigens that cause an undesired and too strong immune response can be modified according to the method of the invention and lead to antigens that have reduced immunogenicity, which, however, is strong enough to use an antigen, for example, as a vaccine.

Some molecules, such as leptin, such as those identified from other mammalian sources, have basically many of the peptide sequences of the present invention and generally have many peptide sequences with substantially similar sequences to those disclosed in the list. Therefore, such protein sequences are equally within the scope of the present invention.

The invention relates to analogues of biological molecules according to the invention, in which substitutions of at least one amino acid residue have been made in positions, which leads to a significant reduction in the activity or elimination of one or more potential T-cell epitopes from the protein.

One or more amino acid substitutions at specific points within any of the potential MHC class II ligands identified in the example tables can lead to a molecule with reduced immunogenic potential when used as a therapeutic agent in humans.

Preferably, amino acid substitutions are made at appropriate points within the peptide sequence predicted to achieve a significant reduction or elimination of T cell epitope activity. A substantially corresponding point will preferably be equated to the binding of an amino acid residue within one of the hydrophobic pockets provided within the MHC class II binding groove. Amino acid residues in the peptide at positions that are equivalent to binding within other pocket regions within the MHC binding gap are also contemplated and fall within the scope of the present invention.

It is understood that individual amino acid substitutions within a given potential T cell epitope are the most preferred way in which the epitope can be eliminated. Substitution combinations within a single epitope may be considered and, for example, may be particularly acceptable where individually defined epitopes overlap. In addition, amino acid substitutions, either separately within a given epitope, or in combination within a single epitope, can be made at positions that do not amount to “pocket residues” with respect to the MHC class II binding groove, but at any point within the peptide sequence. All such substitutions fall within the scope of the present invention.

Amino acid substitutions other than those within the peptides identified above can be especially considered when made in combination with substitution (s) made within the peptide listed. For example, a change may be provided to restore the structure or biological activity of the variant molecule. Such compensatory changes and changes, including the deletion or addition of specific amino acid residues from a molecule according to the invention, leading to a variant with the desired activity, and in combination with changes in any of the disclosed peptides fall within the scope of the present invention.

In another aspect, the present invention relates to the encoding of nucleic acids of said biological molecules having reduced immunogenicity. Methods for creating gene constructs and gene products are well known in the art. In a final aspect, the present invention relates to pharmaceutical compositions comprising these biological molecules, which can be obtained by the methods disclosed in the present invention, and methods of therapeutic treatment of humans using modified molecules and pharmaceutical compositions.

As can be seen from the following examples, the computational methods described above provide a very good indicator that T cell epitopes are likely to be found in any peptide. Thus, this allows the identification of regions of amino acid residue sequences, which, if altered by one or more changes in amino acid residues, has the effect of removing T-cell epitopes and, thus, increases the therapeutic value of the peptide. By this method, biological molecules, such as peptides, proteins, immunoglobulins and fusion proteins and the like, having improved properties and pharmacological value can be obtained.

The above description and examples are intended to be illustrative and should not be construed as limiting. However, other variations within the spirit and scope of this invention are possible and will easily be presented by those skilled in the art.

EXAMPLE 1

This example shows the probability profile of the T cell epitope of the decarboxylase isoform of the glutamic acid autoantigen (GAD 65; MW: 65.000), which is involved in the development of type 1 diabetes. This specific protein may be a potential target in order to increase the affinity of T cell epitopes, and also providing a good example for demonstrating the probability index of the T-cell epitope, since it is a relatively long protein (585 amino acid residues), and therefore provides a relatively large sample size for profiling.

The probability profile of the T-cell epitope for GAD 65 is shown in FIGURE 5. The solid line represents the calculated T-cell epitope index using the computational method shown in FIGURE 1, and the dashed line represents the predicted T-cell epitope index using the computational method shown in FIGURE 3 and 4.

EXAMPLE 2

This example shows the probability profile of the erythropoietin T cell epitope (EPO), a cytokine consisting of 193 amino acid residues, widely used as an intravenous (IV) drug used to increase red blood cell counts. It provides a good example of a biological drug with therapeutic value, but which could trigger inappropriate or undesirable immune responses, especially when used intravenously, and which may therefore benefit from de-immunization after potential T-cell epitopes are identified.

The probability profile of the T-cell epitope for EPO is shown in FIGURE 6. The solid line represents the calculated T-cell epitope index using the computational method shown in FIGURE 1, and the dashed line represents the predicted T-cell epitope index using the computational method shown in FIGURE 3 and 4.

EXAMPLE 3

FIGURES 7 and 8 show the T cell epitope index for the heavy and light chains of a mouse humanized monoclonal antibody directed against the A33 antigen. The latter is a transmembrane glycoprotein, present on the surface of> 95% of cancers of the intestine and therefore has the potential of a therapeutic anti-cancer agent.

In FIGURES 7 and 8, the solid line represents the calculated T-cell epitope index using the computational method shown in FIGURE 1 and the dashed line represents the predicted T-cell epitope index using the computational method shown in FIGURE 3 and 4.

EXAMPLE 4 (leptin):

One of these therapeutically valuable molecules is a human obesity protein called leptin. Leptin is a secreted signaling protein consisting of 146 amino acid residues involved in homeostatic mechanisms supporting fat mass (e.g., WO 00/40615, WO 98/28427, WO 96/05309). Protein (and its antagonists) presents significant therapeutic potential for the treatment of diabetes, high blood pressure and cholesterol metabolism. The protein can be obtained by recombinant methods using a number of different types of host T cells. The amino acid sequence of leptin (represented in a one-letter code) is as follows:

VPIQKVQDDTTKTLIKTIVTRINDISHTQSVSSKQKVTGLDFIPGLHPILTLSKMDQTLAVYQQILTSMPSRNVIQISNDLENLRDLLHVLAFSKSCHLPWASGLETLDSLGGVLEASGYSTLVLSVLVLSSLVLSVVL

Amino acid sequence, which is part of the sequence of immunogenically unmodified obesity protein (leptin) of a person and which has the potential binding activity of MHC class II, is selected from the following group identified according to the method of the present invention:

Any of the above sequences can be used for modification by replacing one or more amino acids to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 5 (IL-1R antagonist):

IL-1 is an important inflammatory and immune modulation cytokine with a pleiotropic effect on various tissues, but can lead to pathologies that are associated with rheumatoid arthritis and other diseases that are associated with local tissue damage. An IL-1 receptor antagonist capable of inhibiting the effect of IL-1 has been purified and the gene cloned [Eisenburg S.P. et al. (1990) Nature, 343: 341-346; Carter, D.B. et al. (1990) Nature, 344: 633-637]. Others provided IL-IRa molecules [eg, US 5075222].

The amino acid sequence of IL-IRa (represented in the one-letter code) is as follows:

RPSGRKSSKMQAFRIWDVNQKTFYLRNNQLVAGYLQGPNVNLEEKIDVVPIEPHALFLGIHGGKMCLSCVKSGDETRLQLEAWITDLSENRKQDKRFAFIRSDSGPTTSFESAACPGWFDTTVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTFVTLTFTTVTA

The amino acid sequence that is part of the immunogenically unmodified IL-IRa sequence and which has potential MHC class II binding activity is selected from the following group:

Any of the above sequences can be used to modify, by replacing one or more amino acids, to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 6 (BDNF):

Another therapeutically valuable molecule is human brain neurotrophic factor (BDNF). BNDF is a glycoprotein of the nerve growth factor protein family. Mature 119 amino acid glycoprotein is isolated from a larger precursor to produce a neutrophic factor that promotes the survival of neural cell populations [Jones KL.R. & Reichardt, L.F. (1990) Proc. Natl. Acad. Sci U.S.A. 87: 8060-8064]. Such neural cells are all located either in the central nervous system or directly connected to it. Recombinant BNDF preparations have revealed the therapeutic potential of the protein for promoting nerve regeneration and the treatment of degenerative diseases.

The amino acid sequence of human brain neurotrophic factor (BDNF) (represented in the one-letter code) is as follows:

HSDPARRGELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKE

GCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR.

Other methods provide modified BDNF molecules [US 5770577], and are aimed at improving the commercial production of a recombinant BDNF molecule [US 5986070].

Amino acid sequence, which is part of the sequence of immunogenically unmodified cerebral neurotrophic factor (BDNF) of a person and which has potential MHC class II binding activity, is selected from the following group:

Any of the above sequences can be used for modification by replacing one or more amino acids to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 7 (EPO):

Another therapeutically valuable molecule is erythropoietin (EPO). EPO is a hormone glycoprotein involved in the development of erythroid progenitor cells in red blood cells. Natural EPO is produced by the liver during fetal life and the kidney of adults and circulates in the blood to stimulate the production of red blood cells in the bone marrow. Anemia is almost always a consequence of kidney failure due to decreased production of EPO by the kidneys. Recombinant EPO is used to effectively treat anemia due to chronic renal failure. Recombinant EPO (expressed in mammalian cells) has the amino acid sequence of 1-165 human erythropoietin [Jacobs, K. et al. (1985) Nature, 313: 806-810; Lin, F. - K. et al. (1985) Proc. Natl. Acad. Sci. U.S.A. 82: 7580-7585] containing three N-linked and one O-linked oligosaccharide chains having terminal sialic acid residues. The latter is important in order to enable EPO to avoid rapid clearance from the blood circulation by the hepatic asialoglycoprotein binding protein.

The amino acid sequence of erythropoietin (EPO) (represented in the single-letter code) is as follows:

APPRLICDSRVLERYLLEAKEAENITTGCAEHCSLNENITVPDTKVNFYAWKRMEVGQQAVEVWQGLALLSEAVLRGQALLVNSSQPWEPLQLHVDKAVSGLRSLTTLLRALGAQKEAISPPDAASAAPLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTLRTL

Amino acid sequence, which is part of the sequence of immunogenically unmodified human erythropoietin (EPO) and which has potential MHC class II binding activity, is selected from the following group:

EXAMPLE 8 (G-CSF):

Colony-stimulating granulocytic factor G-CSF is an important hematopoietic cytokine currently used in the treatment of indications where an increase in blood neutrophils will be beneficial. Which include cancer therapy, various infectious diseases, and related conditions such as sepsis. G-CSF is used alone or in combination with other compounds and cytokines in ex vivo expansion of hematopoietic cells for bone marrow transplantation. Two forms of human G-CSF are commonly recognized for this cytokine. The first is a protein of 177 amino acids, the second is a protein of 174 amino acids [Nagata et al. (1986), EMBO J. 5: 575-581], the 174 amino acid form has been found to have a large specific biological activity in vivo. Recombinant DNA methods allowed the production of G-CSF on an industrial scale using both eukaryotic and prokaryotic host cell expression systems.

The amino acid sequence of colony-stimulating granulocytic factor (G-CSF) (represented in the one-letter code) is as follows:

TPLGPASSLPQSFLLKCLEQVRKIQGDGAALQEKLCATYKLCHPEELVLLGHSLGIPWAPLSSCPSQALQLAGCLSQLHSGLFLYQGLLQALEGISPELGPTLDTLQLDVADFATTIWQQMEELGMAPALQPAQLFRVLFAMPLFRVFAMPLF

Other analogs of the polypeptide and fragments of the G-CSF peptide have been previously disclosed, including forms modified by site-specific amino acid substitutions and or by modification by chemical adducts. Thus, US 4,810,643 discloses analogues with specific Cys residues replaced by another amino acid, and G-CSF with an Ala residue in the first (N-terminal) position. EP 0335423 discloses a modification of at least one amino group in a polypeptide having G-CSF activity. EP 0272703 discloses derivatives of G-CSF in which an amino acid is substituted or deleted at the N terminus. EP 0459630 discloses derivatives of G-CSF in which Cys 17 and Asp 27 are replaced by Ser residues. EP 0243153 discloses G-CSF modified by inactivation of at least one site of the yeast-processing KEX2 protease to increase production in recombinant products, and US 4,904,584 discloses lysine altered proteins. WO 90/12874 further discloses Cys altered variants and Australian Patent Document AU-A-10948/92 discloses the addition of amino acids to either end of a G-CSF molecule to facilitate folding of the molecule after prokaryotic expression. AU-763 80/91 discloses G-CSF variants at positions 50-56 for the 174 amino acid form of G-CSF and at positions 53-59 for the 177 amino acid form. Additional changes in His specific residue have also been described.

The amino acid sequence that is part of the human colony-stimulating granulocyte factor (G-CSF) sequence and which has potential MHC class II binding activity is selected from the following group:

Any of the above sequences can be used for modification by replacing one or more amino acids to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 9 (KGF):

Another therapeutically valuable molecule is keratinocyte growth factor (KGF). KGF is a member of the fibroblast growth factor (FGF) / heparin-binding growth factor protein family. This is a secreted glycoprotein secreted mainly in the lung, promoting healing of damage, stimulating the growth of keratinocytes and other epithelial cells [Finch et al. (1989), Science 24: 7 52-755; Rubin et al. (1989), Proc. Natl. Acad. Sci. U.S.A. 86: 802-806]. The developed (transformed) form of the glycoprotein includes 163 amino acid residues and can be isolated from the conditioned medium of the following culture of specific cell lines [Rubin et al. (1989) ibid.], Or obtained using recombinant methods [Ron et al. (1993) J. Biol. Chem. 268: 2984-2988]. The protein has therapeutic value to stimulate the growth of epithelial cells in a number of serious diseases and conditions for the regeneration of damage. This description, in particular, relates to human KGF protein, which is a developed (transformed) form of 163 amino acid residues.

Other methods provide KGF molecules [for example, US 6,008,328; WO 90/08771;] including modified KGF [Ron et al. (1993) ibid; WO 95/01434]. However, these methods do not recognize the importance of T-cell epitopes for the immunogenic properties of a protein; they do not consider the direct effect of these properties on a specific and controlled pathway in accordance with the scheme of this invention.

The amino acid sequence of keratinocyte growth factor (KGF) (represented in the one-letter code) is as follows:

MCNDMTPEQMATNVNCSSPERHTRSYDYMEGGDIRVRRLFCRTQWYLRIDKRGKVKGTQEMKNNYNIMEIRTVAVGIVAIKGVESEFYLAMNKEGKLYAKKECNEDCNFKELILENHYNTYASAKWTHNGKKVKKLKGKK

Amino acid sequence, which is part of the sequence of human keratinocyte growth factor (KGF), and which has the potential binding activity of MHC class II, is selected from the following group:

Any of the above sequences can be used for modification by replacing one or more amino acids to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 10 (soluble TNF RI):

sTNF-RI (soluble type I tumor necrosis factor receptor) is a derivative of the human tumor necrosis factor receptor described previously [Gray, P.W. et al. (1990) Proc. Nat. Acad. Sci. U.S.A. 87: 7380-7384; Loetschere, N. et al. (1990) Cell 61: 351-359; Schall, TJ. et al. (1990) Cell 61; 361-370], containing the extracellular domain of the intact receptor and having an approximate molecular weight of 30KDa. Additional soluble TNF inhibitors and in particular form 40 KDa are also known [US 6143866]. Soluble forms are able to bind alpha tumor necrosis factor with high affinity and inhibit cytokine cytokine activity in vitro. Recombinant sTNF-RI preparations have significant therapeutic value for the treatment of diseases where an excess level of tumor necrosis factor causes a pathogenic effect. Indications such as cachexia, sepsis and autoimmune disorders, and in particular rheumatoid arthritis, and others may be targets for the administration of such therapeutic drugs sTNF-RI. Others, including Brewer et al., US 6143866, provided modified sTNF-RI molecules.

Any of the sequences shown can be used to modify by replacing one or more amino acids to obtain a sequence that is non-immunogenic or has reduced immunogenicity.

EXAMPLE 11 (soluble TNF-R2):

The soluble tumor necrosis factor 2 receptor (sTNF-R2) is a derivative of the human tumor necrosis factor 2 receptor described previously [Smith, C.A. et al. (1990) Science 248: 1019-1023; Kohno, T. et al. (1990) Proc. Nat. Acad. Sci. U.S.A. 87: 8331-8335; Beltmger, C.P. et al. (1996) Genomics 35: 94-100] containing the extracellular domain of an intact receptor. Soluble forms are able to bind tumor necrosis factor with high affinity and inhibit cytokine cytokine activity in vitro. Recombinant sTNF-R2 preparations have significant therapeutic value for the treatment of diseases where an excessive level of tumor necrosis factor causes a pathogenic effect. A specific recombinant drug called ethanercept has received clinical approval for the treatment of rheumatoid arthritis, and this and other similar agents may be useful in treating other indications, such as cachexia, sepsis and autoimmune disorders. Etanercept is a dimeric fusion protein containing the extracellular domain of a human TNFR2 molecule in combination with the Fc domain of a human IgGl molecule. The dimeric molecule includes 934 amino acids [US 5395760; US 5605690; US 5945397, US RE36.755].

EXAMPLE 12 (β-GCR):

Beta-glucocerebrosidase (b-D-glucosyl-N-acylsphingosine glucohydrolase, E.C.3.2.1.45) is a monomeric glycoprotein consisting of 497 amino acid residues. The enzyme catalyzes the hydrolysis of glucocerebroside glycolipid to glucose and ceramide. Lack of GCR activity leads to lysosomal storage disease, known as Gaucher disease. The disease is characterized by the accumulation of macrophages of tissue poured by glucocerebroside, which accumulate in the liver, spleen, bone marrow and other organs. The disease has various degrees of severity: from type 1 - a disease with hematological problems, but without a neuronal lesion, to type 2 - a disease manifesting shortly after birth with an extensive neuronal lesion and general progression with a fatal outcome within 2 years. Some classifications also recognize the third type of disease, which is also characterized by neuronal damage. Previously, the only useful therapy for Gaucher’s disease was the use of GCR obtained from human placenta (known as aglucerase), but later the pharmaceutical preparations of recombinant GCR (“ceredase” and “ceresim”) showed efficacy in the treatment of the first type of disease [Niederau, C. et al. (1998) Eur. J. Med. Res. 3: 25-30].

EXAMPLE 13 (Protein C):

Protein C is a vitamin K-dependent serine protease involved in the regulation of blood coagulation. The protein is activated by thrombin to produce activated protein C, which in turn inhibits Factors Va and VIII in the coagulation cascade. Protein C is produced in the liver as a single-chain precursor and undergoes a series of treatments that end with a molecule containing light and heavy chains joined together by a disulfide bond. Protein C is activated by cleaving the tetradecapapeptide from the N-terminus of the heavy chain with thrombin. Protein C pharmaceutical preparations in native or activated form are valuable in treating patients with vascular diseases and / or acquired deficiency in Protein C. Such patients therefore include individuals suffering from thrombotic stroke or Protein C deficiency associated with sepsis, transplant procedures, pregnancy, serious burns, radical surgery, or other serious injuries. Protein C is also used in the treatment of individuals with a hereditary deficiency in Protein C. This description particularly applies to a protein C of a person in a mature (processed) form containing a light chain of 155 amino acid residues and a heavy chain of 262 amino acid residues [Foster, D.C. et al. (1985) Proc. Nad. Acad. Set. U.S.A. 82: 4673-4677; Beckman, R.J. et al. (1985) Nucleic Acids Res. 13: 5233-5247]. Others have obtained protein C molecules, including activated protein C compositions, and methods of use [US 6159468; US 6156734; US 6037322; US 5618714].

EXAMPLE 14 (subtilisins)

Subtilisins are a class of protease enzyme with significant economic and industrial importance. They can be used as components of detergents or cosmetics or in the manufacture of textiles and other industries. When exposed to some people, bacterial subtilisins can cause an undesirable hypersensitivity reaction. Therefore, there is a need for subtilisin analogues with improved properties and especially improved biological properties of the protein. In this regard, it is very desirable to develop subtilisins with a reduced or absent potential to induce an immune response in relation to humans. Subtilisin proteins identified from various sources, including bacterial, fungal, or vertebrate, as well as mammals and humans, have in general many of the protein sequences of the present disclosure and have in general many protein sequences with substantially the same sequence as those listed. Therefore, such protein sequences equally fall within the scope of the present invention. Others have produced subtilisin molecules, including modified subtilisins [US 5700676; US 4914031; US 5,397,705; US 5972682].

Example 15 (CNTF ligands):

The present invention provides modified forms of protein subunits containing a heterodimeric ligand for the ciliary neurotrophic factor receptor complex (CNTF) in humans. The receptor complex is activated by at least two ligands, including CNTF and a heterodimeric complex containing a cardiotrophin-like cytokine (CLC) and a soluble cytokine-like factor 1 receptor (CLF) [Elson G.CA. et al. (2000) Nature Neuroscience 3: 867-872]. CLC is a protein of the IL-6 family of cytokines, also known as the novel neurotrophin-1 / B cell-stimulating factor-3 [Senaldi, G. et al. (1999) Proc. Nat. Acad. Sci. USA 96: 11458-11463, US 5741772]. CLF is homologous to type I cytokine receptor family proteins [Elson, G.C.A. et al. (1998) Journal of Immunol. 161: 1371-1379] and was also identified as NR6 [Alexander W.S. et al. (1999) Curr. Biol. 9: 605-608]. The heterodimers formed by the association of CLC and CLF showed the ability to directly interact with CNTFR, and thus the formed trimeric complex is able to stimulate signaling of events within cells expressing other recognized components of the CNTFR complex of the gp130 and LIFR type [Elson G.C.A. et al. (2000), ibid.].

EXAMPLE 16 (follicle-stimulating hormone)

The present invention provides modified forms of human hFSH with one or more deleted T cell epitopes. hFSH is a dimeric hormone glycoprotein containing two glycoprotein subunits. Protein is therapeutically used in the treatment of human infertility and recombinant forms of the protein have been the subject of many clinical trials [Out, H.J. et al. (1995) Hum. Reprod. 10: 2534-2540; Hedon, B. et al. (1995) Hum. Reprod. 10: 3102-3106; Recombinant Human FSH study Group (1995) Fertil. Steril. 63: 7786; Prevost, R.R. (1998) Pharmacotherapy 18: 1001-1010].

EXAMPLE 16 (ricin A)

The present invention provides modified forms of a ricin toxin A chain (RTA) with one or more deleted T cell epitopes. Ricin is a cytotoxin originally isolated from the seeds of a castor plant and is an example of a type II ribosome inactivating protein (RIP). A native mature protein is a RTA-containing heterodimer that consists of 267 amino acid residues, on a disulfide bond with the ricin B chain, which consists of 262 amino acid residues. The B chain is pectin with a binding affinity for galactosides. The native protein is able to bind cells through the B-chain and enter the cell by endocytosis. Inside the cell, RTA is isolated from the B-chain by the restoration of the disulfide bond and leaves the endosome in the cytoplasm by unknown mechanisms. In the cytoplasm, the toxin destroys the ribosomes, acting as a specific N-glycosylase, which quickly leads to the cessation of protein conversion and cell death. The extreme cytotoxicity of RTA and other RIPs has led to their use in experimental therapies for the treatment of cancer and other diseases that require ablation of a specific cell population. Immunotoxin molecules containing antibody molecules in connection with RTA were obtained and used in numerous clinical trials [Ghetie, M.A. et al. (1991) Cancer Res. 51: 5876-5880; Vitetta, E.S. et al. (1991) CancerRes. 51: 4052-4058; Arnlot, P.L. et al. (1993) Bloody. 2624-2633; Conry, R.M. et al. (1995) J. Immunotlier. Emphasis Tumor Immunol. 18: 231-241; Schnell, R. et al. (2000) Leukaemia 14: 129-135]. In an immunotoxin, an antibody domain provides binding to the surface of a desired target cell, and communication with RTA can be achieved through chemical cross-linking or as a recombinant fusion protein.

EXAMPLE 17 (adipocyte complement associated protein):

The present invention provides modified forms of Acrp30 human or mouse with one or more deleted T-cell epigones. Acrp30 is an excess serum protein with a molecular weight of approximately 30 kDa, secreted exclusively by adipocyte (fat) cells [Scherer, P.E. et al. (1995) J. Biol. Chez. 270: 26746-26749]. The protein sequence of the human Acrp30 gene is described, for example, in US Pat. No. 5,869,330. Insulin increases protein secretion and protein levels are reduced in obese patients. Protein is involved in regulating the energy balance, and especially in regulating the metabolism of fatty acids. The four domain sequences identified in the mouse and human proteins contain a cleaved N-terminal signal, a region with unrecognized homology to other proteins, a collagen-like domain, and a globular domain. The globular domain can be removed from the mouse protein by protease treatment to obtain gAcrp30. Mouse gAcrp30 preparations have pharmaceutical properties and show the ability to reduce increased levels of free fatty acids in mouse serum after eating high fatty foods or intravenous lipid injections [Fruebis, J. et al. (2001) Proc. Natl. Acad. Sci. U.S.A. 98: 2005-2010].

EXAMPLE 18 (anti-C5 antibody):

The present invention provides modified forms of monoclonal antibodies with binding specificity directed to C5 complement human protein. The invention provides modified antibodies with one or more deleted T cell epitopes. Antibodies with binding specificity directed at the C5 complement protein block the activation of cleavage of the C5 convertase and thus inhibit the production of the pro-inflammatory components C5a and C5b-9. Activation of the complement system is a significant contributing factor in the pathogenesis of many acute and chronic diseases, and inhibition of the complement cascade at the C5 level offers a promising therapeutic approach for some of them [Morgan B.P. (1994) Eur. J. Clin. Invest. 24: 219-228]. Several anti-C5 antibodies and methods for their therapeutic use have been described in the art [Wurzner R. et al. (1991) Complement Inflamm. 8: 328-340; Thomas, T.C. et al. (1996) Molecular Immunology 33: 1389-14012; US 5853722; US 607464]. Antibody designated as 5G1.1 [Thomas, T.C. et al. (1996), ibid.] and the single-chain humanized variant were clinically tested for several indications of diseases, including extracorporeal circulation [Fitch, J.C.K. et al. (1999) Circulation 100: 2499-2506] and rheumatoid arthritis. The invention discloses sequences identified within the anti-C5 antibody designated as 5G1.1 [Thomas, T.C. and etc.]. The described sequences are obtained from the domains of the variable regions of the heavy and light chains of the antibody sequence, which are potential T-cell epitopes based on the binding potential of MHC class II. The description further defines the potential epitopes within the protein sequences of the single-stranded and "humanized" variants of the 5G1.1 antibody [Thomas, T.C. et al. (1996), ibid.].

EXAMPLE 19 (anti-CD20 antibodies):

The present invention provides modified forms of a monoclonal antibody with binding specificity directed at the human CD20 antigen. CD20 is a B cell specific surface molecule produced in pre-B and mature B cells, including more than 90% of non-Hodgkin's lymphoma B cells (NHL). Monoclonal antibodies and radioimmunoconjugates attacking CD20 have emerged as new approaches for the treatment of NHL. Illustrative examples include monoclonal antibodies 2B8 [Reff, ME et al. (1994) Blood 83: 435-445] and B1 [US 6090365]. 2B8 variable region domains were cloned and combined with human constant region domains to produce a chimeric antibody designated C2B8, sold under the name Rituxan ^™ in the USA [US 5776456] or MabThera ^R (rituximab) in Europe. C2B8 is recognized as a valuable therapeutic agent for the treatment of NHL and other B-cell diseases [Maloney, DG et al. (1997) J. Clin. Oncol. 15: 3266-3274; Maloney, DG and up. (1997) Blood 90: 2188-2195]. The B1 antibody was similarly registered for use as an NHL therapeutic agent, although in this case the molecule (Bexxar ^TM ) is a ¹³¹ I radioimmunoconjugate, although the native (non-conjugated) antibody is useful in ex vivo purification modes for autologous bone marrow transplantation treatment for lymphomas and intractable leukemia [Freedman, AS et al. (1990), J. Clin. Oncol. 8: 784]. Despite the success of antibodies of type C2B8 (rituximab) and Bexxar ^™, there is a need for anti-CD20 analogues with improved properties.

EXAMPLE 20:

The present invention provides modified forms of a monoclonal antibody with binding specificity for the human IL-2 receptor. The monoclonal antibody designated as anti-TAC and the modified form have one or more deleted T cell epitopes. The anti-TAC antibody binds with high specificity to the alpha subunit (p55 alpha, CD25, or TAC) of the human high affinity IL-2 receptor, expressed on the surface of T and B lymphocytes. A binding antibody blocks the ability of IL-2 to bind to the receptor and achieve T cell activation. The ability of an anti-TAC antibody to act as an IL-2 antagonist has significant clinical potential for the treatment of organ transplant rejection. Clinical studies using mouse antibodies have shown some initial utility for patients who underwent a kidney transplant, although the long-term benefits of conventional immune suppression have not been achieved due to the development of a NAMA response in most patients [Kirkham, RL et al. (1991) Transplantation 51 : 107-113]. A “humanized” anti-TAC antibody has been developed in which the essential components of a protein are designed to contain a protein sequence identified from the human antibody gene [Queen, C. et al. (1989) Proc. Natl. Acad. Sci. (USA) 86: 10029-10033; US 5530101; US 5585089; US 6013256]. The “humanized” anti-TAC (Zenapax ^™ or daclizumab) has been clinically tested as an immunosuppressive agent to control acute host transplantation disease and suppress renal transplant rejection [Anasetti, C. et al. (1994). Blood 84: 1320-1327; Anasetd, S. et al. (1995) Blood 86: Supplement 1: 62a; Eckhoff, DE et al. (2000) Transplantation 69: 1867-1872; Ekberg, N. et al. (1999) Transplant Proc. 31: 267-268].

EXAMPLE 21 (14.18 antibody):

Unless otherwise indicated, all amino acids in the variable heavy and light chains are numbered, as in Kabat et al., 1991 (Sequences of Proteins of Immunological Interest, US Department of Health and Human Services). Potential T-cell epitopes are numbered with the linear number of the first amino acid of the epitope, counting from the first amino acid of the heavy and light chains.

1. Comparison with Mouse Subgroup Structures

The amino acid sequences of murine 14.18 VH and VK were compared with sequences of common typical structural elements for Kabat murine subgroups of the heavy and light chains (Kabat et al., 1991). 14.18 VH can be assigned to subgroup IIA of the mouse heavy chains. The sequence of 14.18 VH is shown on SEQ. No. 1. Comparison with the sequence of common typical structural elements for this subgroup shows that histidine at position 81 (usually glutamine), lysine at position 82a (usually serine or asparagine), valine at position 93 (usually alanine) and serine at position 94 (usually arginine) are unusual for this subgroup. Residues at positions 19, 40 and 66 are also infrequent in this subgroup, but are not believed to have a significant effect on the structure and binding of the antibody. 14.18 VK can be assigned to subgroup II of kappa mouse chains. Comparison with the sequence of common typical structural elements for this subgroup shows that histidine at position 49 is unusual for this subgroup. The most common residue for this position is tyrosine.

2. Comparison with Human Structures

The amino acid sequences of the murine 14.18 VH and VK were compared with the sequences of the human germline (VH) germline line (Tomlinson et al., (1992) J. Mol. Biol. 227: 776-798) and VK (Sokh et al., 1994 ) Eur. J. Immunol. 1-4-: 827-83b) sequences, as well as with the region of human J germline sequences (Routledge et al., In Protein Engineering of Antibody Molecules for Prophylactic and Therapeutic Applications in Man, Clark, M ed. Academic Titles, Nottingham, UK, pp 13-44, 1991). The human reference structure selected for 14.18 VH was DP25 with human JH6. This germline sequence was found in the rearranged mature antibody gene with no amino acid changes. For structure 3, the sequence of mature human antibody 29 was used. This sequence is identical to the murine sequence directly adjacent to CDR3. The human reference structure selected for 14.18VK was DPK22. This germline sequence was found in the rearranged mature antibody gene with no amino acid changes. For structure 2, the sequence of the mature human antibody 163.5 was used. This sequence is identical to the murine sequence directly adjacent to CDR2. The sequence of the J region was human JK2 (Routledge et al., 1993).

3. Development of “veneered” sequences

After identification of the base sequences of the reference human structure, some non-identical amino acid residues within the 14.18 VH and VK structures were changed to the corresponding amino acid in the base sequences of the human structure. Residues that are believed to be critical for antibody structure and binding were excluded from this process and did not change. The mouse residues that were retained at this stage were largely non-surface, internal residues, except for residues at the N-ends, for example, which are close to the CDRs in the target antibody. This process leads to a sequence that is definitely similar to a “veneered” antibody, since surface residues are mainly human and internal residues, such as in the original mouse sequence.

4. Peptide Losing Analysis

The murine and venous 14.18 VH and VK sequences were analyzed using the method of the invention. The amino acid sequences were divided into all possible thirteen-dimers. Thirteen-dimensional proteins are sequentially presented to models of the binding groove of the HLA-DR allotypes, and binding values correlated with each peptide for each allele. The conformational value for each pocket-linked side chain of the peptide was calculated. This value is based on steric superposition, potential hydrogen bonds between the protein and the residues in the binding groove, electrostatic interactions, and favorable contacts between the protein and pocket residues. The conformation of each side chain was changed, the value was re-calculated. Having determined the highest conformational value, the binding value was recalculated based on the groove-bound hydrophobic residues, the non-grooved hydrophilic residues and the amount of residues that fit into the binding groove. Peptides, which are known binding class II MHC of a person, achieve high binding values with almost no incorrect results. Thus, peptides that achieve significant binding in the current assay are believed to be potential T cell epitopes. The results of the peptide piercing assay are shown in Table 1.

Table 1: Potential T-cell epitopes in murine and venous 14.18 sequences Sequence The number of potential T-cell epigones The location of the potential epigope Mouse 14.18 VH eleven 3 (17), 9 (15), 30 (5), 35 (17), 39 (15), 43 (9), 58 (12), 62 (11), 81 (11), 84 (16), 101 (7) Veneered 14.18 VH 5 43 (9), 58 (12), 62 (11), 81 (11), 84 (16) Mouse 14.18 VK 7 7 (7), 13 (11), 27 (15), 49 (11), 86 (17), 97 (11), 100 (4) Veneered 14.18 VK 5 27 (15), 49 (11), 86 (17). 97 (11), 100 (17)

5. Removal of potential T-cell epitopes

Potential T cell epitopes are removed by amino acid substitution in the specific peptide that makes up the epitope. Substitutions were made by the introduction of amino acids with possibly similar physicochemical properties. However, in order to remove some potential epitopes, amino acids of different size, charge or hydrophobicity may need to be replaced. If changes must be made within the CDRs that could have a binding effect, it is necessary to create a variant with and without specific amino acid substitution. Amino acid residues for substitution are given with the Kabat number indicated in parentheses. Potential T cell epitopes are indicated by the linear number of the first thirteen-dimensional residue.

The necessary amino acid changes to remove T-cell epitopes from the 14.18 venous variable region of the heavy chain were as follows:

1. The replacement of isoleucine with proline at residue 41 (Kabat number 41), combined with the replacement of leucine with alanine at residue 50 in CDR2, removes the potential epitope at position 43.

2. Alternative to (1), replacing threonine with leucine at residue 45 (Kabat number 45) with proline at position 41 (Kabat number 41) also removes the potential epitope at position 43.

3. Replacing serine with glycine at residue 66 (Kabat number 65) in CDR2 and valine with alanine at residue 68 (Kabat number 67) removes the potential epitope at position 58. The series was found at this position in human and mouse antibody sequences.

4. Replacing isoleucine with leucine at residue 70 (Kabat: 69) reduces the number of MHC allotypes that bind to a potential epitope at position 62 from 11 to 4.

5. Replacing alanine with valine at position 72 (Kabat number 71) removes the potential epitope at position 62. The amino acid size at this position is critical and alanine is hydrophobic and similar in size to valine.

6. Replacing threonine on the series at residue 91 (Kabat number 87) removes potential epitopes at positions 81 and 84.

The necessary amino acid changes to remove potential T-cell epitopes from the 14.18 venous variable region of the light chain were as follows:

1. Replacing serine with arginine at residue 32 (Kabat number 27e) removes the potential epitope at position 27. This residue is within CDR2, however, the series are often found at tWs position in mouse and human antibodies. There is no change with CDR that eliminates a potential T cell epitope.

2. Replacing tyrosine with histidine at position 54 (Kabat number 49) eliminates the potential epitope at position 43. Tyrosine is the most common amino acid found at position 49 in mouse and human antibodies.

1. An alternative substitution to (2) to eliminate the potential epitope at position 43 is the substitution of methionine for leucine at residue 51 (Kabat number 46). Methionine is similar in size and hydrophobicity to leucine.

2. Replacing methionine with leucine at residue 88 (Kabat number 83) eliminates the potential epitope at position 86.

1. The replacement of threonine with leucine at residue 102 (Kabat number 96) in CDRH3 in combination with the replacement of glutamine with glycine at position 105 (Kabat number 100) reduces the number of MHC allotypes that bind to the potential epitope at position 97 from 11 to 5.

2. An alternative substitution to (5), which eliminates the potential epitope at position 97, is the replacement of pralin with leucine at residue 102 (Kabat number 96).

3. Replacing valine with leucine at residue 110 (Kabat number 104) removes, eliminates the potential epitope at position 100.

6. Development of de-immunized sequences

The deimmunized heavy and light chain sequences were designed with the substitutions required to remove potential T-cell epitopes and analyze the structure of residues that could be critical for antibody structure and binding. In addition to the de-immunized sequences that are based on the venous sequence, an additional sequence was developed for each VH and VK based on the mouse sequence, called the Mouse Protein Stitched (Mo RT) version. For this version, substitutions were made directly in the mouse sequence to eliminate T-cell epitopes, but changes were made only with CDRs that are not considered undesirable for binding. No attempt was made to remove surface (B-cells) epitopes for this version of the immunized sequence.

Primary de-immunized VH includes the substitutions 1, 3, 4, 5, and 6 described in section 5 above and does not include any potential T cell epitopes. An additional 4 de-immunized VH sequences were designed to test the effect of the necessary different substitutions on antibody binding. Version 2 is an alternative to Version 1, in which an alternative replacement (2 in Section 2.5 above) was used to remove the same potential T-cell epitope. Cumulative changes were made for the first de-immunized sequence (14.18DIVH1) and the remaining potential T-cell epitopes are shown in Table 2. A mouse stitched version is included for comparison.

Table 2: Amino Acid Substitutions and Potential Epitopes in a Dimmunized 14.18 VH Option Cumulative Balance Changes Potential epitopes (No. of potential MHC binders of 18 tested) 14.18DIVH1 no no 14.18DIVH2 411 → P, 45L → T, 50L → A A no 14.18DIVH3 65S → G 58 (8) 14.18DIVH4 71A → V 58 (8), 62 (4) 14.18DIVH5 45T → L, 41P → 1 43 (9), 58 (8), 62 (4) 14.18MoPTVH NA 43 (9), 58 (12), 62 (11)

Primary de-immunized VK includes substitutions 1, 2, 4, 6, and 7 described in section 5 above and does not include any potential T-cell epitopes. An additional 5 de-immunized VK sequences were designed to test the effect of the necessary various substitutions on antibody binding. Version 2 is an alternative to Version 1, which used an alternative substitution to remove the same potential T-cell epitope at position 43. Version 3 includes an alternative substitution (6 in Section 2.5 above), this replacement reduces the number of MHC allotypes that bind to potential epitope at position 97 from 11 to 5. Aggregate changes were made for the primary de-immunized sequence (14.18DIVK1) and the remaining potential T-cell epitopes are shown in Table 3. A mouse stitched version is included for comparison I am.

Table 3: Amino acid substitutions and potential epitopes in a de-immunized 14.18 VK Option Total balance changes * Potential epitopes (No. of potential MHC binders of 18 tested) 14.18DIVK1 no no 14.18DIVK2 46LS → M, 49Y → H no 14.18DIVK3 96P → T, 100Q → G 97 (5) 14.18DIVK4 96T → L 97 (11) 14.18DIVK5 27e S → R 27 (15), 97 (11) 14.18DIVK6 46M → L 27 (15), 49 (11), 97 (11) 14.18MOPTVK NA 27 (15), 49 (11), 97 (11), 100 (4)

EXAMPLE 22 (KS antibody)

1. Comparison with Mouse Subgroup Structures

The amino acid sequences of murine KS VH and VK were compared with sequences of common typical structural elements for Kabat murine subgroups of the heavy and light chains (Kabat et al., 1991). KS VH cannot be assigned to any of the subgroups, but is closest to subgroup II (A) and V (A). Atypical residues are found at position 2, which is usually valine, 46, which is usually glutamic acid, and 68, which is usually threonine. Residue 69 is more commonly leucine or isoleucine. In 82b, series are most often found. Mouse KS VK can be assigned to Subgroup VI (Figure 2). Unusual residues are found at 46 and 47, which are usually both leucine. Residue 58 is not typical with either leucine or valine, usually in this position.

2. Comparison with Human Structures

The amino acid sequences of murine KS VH and VK were compared with the sequences of the germline human germline (VH (Tomlinson et al., (1992)) and VK (Cox et al., (1994)) sequences, as well as with the region of the germline sequences J human (Routledge et al., (1993)). The human reference structure selected for KS VH was DP 10 with human JH6. This germline sequence was found in the rearranged mature antibody gene with no amino acid changes. The human reference structure selected for KS VK was B1. For structure 2, the sequence of the mature human IMEV antibody was used (Kabat et al., 1991). This sequence is identical to the murine sequence directly adjacent to CDR2. The sequence of the J region was human JK2 (Routledge et al., 1993). This germ line sequence was not found as a rearranged mature antibody light chain.

3. Development of “veneered” sequences

After identification of the base sequences of the reference human structures, some non-identical amino acid residues within the 425 VH and VK structures were changed to the corresponding amino acid in the human reference sequence. Residues that are believed to be critical for antibody structure and binding were excluded from this process and did not change. The mouse residues that were retained at this stage were largely non-surface, internal residues, except for residues at the N-ends, for example, which are close to the CDRs in the target antibody. This process leads to a sequence that is definitely similar to a “venated” antibody, since surface residues are mainly human and internal residues are those of the original murine sequence.

4. Peptide Piercing Analysis

Mouse and venous KS VH and VK sequences were analyzed using the method of the invention. The amino acid sequences were divided into all possible thirteen-dimers. Thirteen-dimensional proteins are sequentially presented to models of the binding groove of the HLA-DR allotypes, and binding values correlated with each peptide for each allele. The conformational value was calculated for each pocket-linked side chain of the peptide. This value is based on steric superposition, potential hydrogen bonds between the protein and the residues in the binding groove, electrostatic interactions, and favorable contacts between the protein and pocket residues. The conformation of each side chain was changed, the value was re-calculated.

Having determined the highest conformational value, the binding value was recalculated based on the groove-bound hydrophobic residues, the non-grooved hydrophilic residues and the amount of residues that fit into the binding groove. Peptides, which are known binding class II MHC of a person, achieve significant binding values with almost no incorrect results. Thus, peptides that achieve significant binding in the current assay are believed to be potential T cell epitopes. The results of the peptide piercing assay for murine and venous sequences are shown in Table 4.

Table 4: Potential T-cell epitopes in murine and venous KS sequences Sequence The number of potential T-cell epigones Location of potential epitopes (No. of potential MHC binders) Mouse KS VH 6 35 (11), 62 (17), 78 (12), 81 (12), 89 (6), 98 (15) Mouse KS VH 5 30 (7), 62 (15), 78 (11), 89 (6), 98 (15) Mouse KSVK b 1 (14), 2 (5), 17 (5), 27 (5), 51 (13), 72 (18) Veneered KS VK 3 1 (17), 27 (5), 51 (13)

5. Removal of potential T-cell epitopes

Potential T cell epitopes are removed by amino acid substitution in the specific peptide that makes up the epitope. Substitutions were made by the introduction of amino acids with possibly similar physicochemical properties. However, in order to remove some potential epitopes, amino acids of different size, charge or hydrophobicity may need to be replaced. If changes are to be made within the CDR-ob, which could have a binding effect, it is necessary to create a variant with and without specific amino acid substitution. Amino acid residues for substitution are numbered, as in Kabat. Potential T cell epitopes are indicated by the linear number of the first thirteen-dimensional residue.

The necessary amino acid substitutions to remove T-cell epitopes from the KS veneered variable region of the heavy chain were as follows:

1. replacing arginine with lysine at residue 38 (Kabat number 38) removes the potential epitope at residue number 30.

2. replacing alanine with leucine at residue 72 (Kabat number 71) and isoleucine with phenialanine at residue 70 (Kabat number 69) removes the potential epitope at residue 62. Isoleucine in Kabat number 69 and alanine in Kabat number 71 are found in human sequence VH germline, DP10.

3. replacing leucine with alanine at residue 79 (Kabat number 78), removes the potential epitope at residue number 78.

1. replacing threonine with methionine at residue 91 (Kabat number 87), removes the potential epitope at residue 89.

2. substituting methionine for isoleucine at residue 100 (Kabat number 96) in CDRH3 removes the potential epitope at residue 98. There is no change with CDRH3 that removes this potential epitope.

The necessary amino acid substitutions to remove T-cell epitopes from the KS venous variable region of the light chain were as follows:

1. Replacing isoleucine with methionine at residue 32 (Kabat number 33), removes the potential epitope at residue number 27. This residue is within CDR2. Isoleucine is usually found in this position in human antibodies.

2. The potential epitope at position 1 is removed by replacing the valine with the leucine residue (Kabat No. 3).

3. Replacing serine with alanine at residue 59 (Kabat number 60) removes the potential epitope at residue 51.

6. Development of de-immunized sequences

The deimmunized heavy and light chain sequences were designed to take into account the changes required to remove potential T-cell epithets and analyze the structure of residues that could be critical for antibody structure and binding. In addition to the de-immunized sequences that are based on the venous sequence, an additional sequence was developed for each VH and VK based on the mouse sequence, called the Mouse Protein Stitched (Mo RT) version. For this version, substitutions were made directly in the mouse sequence to eliminate T cell epitopes, but the changes were made only with CDRs that are not considered undesirable for binding. No attempt was made to remove surface (B-cells) epitopes for this version of the immunized sequence.

Primary de-immunized VH includes substitutions 1-5, described in section 5 above, and one additional change at residue 43 (Kabat number 43). The lysine found in the mouse sequence was replaced with glutamine from the human structure. Lysine is positively charged and therefore significantly different from glutamine; this area may be involved in VH / VL contacts. Primary de-immunized VH does not include any potential T cell epitopes. An additional 4 de-immunized VH sequences were designed to test the effect of the necessary different substitutions on antibody binding. Cumulative changes were made for the first de-immunized sequence (KSDIVHv1) and the remaining potential T-cell epitopes are displayed in Table 5.

Table 5: Amino Acid Substitutions and Potential Epitopes in Dimmunized KS VH Option Cumulative Balance Changes Potential epitopes (No. of potential MHC binders of 18 tested) KSDIVHv1 no no KSDIVHv2 96M → I 98 (15) KSDIVHvS 71A → L, 78L → A 62 (16), 78 (11), 98 (15) KSDIVHv4 38 R → K 30 (7), 62 (16), 78 (11), 98 (15) KSDIVHv5 68T → A, 69I → F 30 (7), 62 (17), 78 (11), 98 (15) KSMoPTVH NA 98 (15), 78 (12)

Primary de-immunized VK includes substitutions 1-3 described in section 5 above. An additional 3 de-immunized VK sequences were designed to test the effect of the necessary different substitutions on antibody binding. Cumulative changes were made for the primary de-immunized sequence (KSDIVKv1) and the remaining potential T-cell epitopes are displayed in Table 6.

Table 6: Amino Acid Substitutions and Potential Epitopes in a Dimmunized KS VK Option Cumulative Balance Changes Potential epitopes (No. of potential MHC binders of 18 tested) KSDIVKv1 no no KSDIVKv2 331 → M 27 (5) KSDIVKvS 3V → L 1 (17), 27 (5) KSDIVKv4 60S → A 1 (17), 27 (5), 5 (13) KSMoPTVK NA no

Claims (17)

1. A method suitable for identifying one or more potential T cell epitopes of peptides within the amino acid sequence of a biological molecule through steps including determining the binding of these peptides to the molecules of the major histocompatibility complex class II (MHC class II) using in vitro methods or biological assays , the specified method includes the following stages: (a) selecting a portion of a peptide that has a known sequence of amino acid residues; (b) sequential selection of overlapping segments of amino acid residues with a predetermined equal size and consisting of at least three amino acid residues of the selected site, and the specified selected segment of amino acid residues consists of 13 amino acid residues (c) calculating the binding index of the MHC class II molecule for each specified selected segment by summing certain values for each hydrophobic amino acid residue of the side chain that is present in the specified selected segment of the amino acid residue; and (d) identifying at least one of these segments suitable for modification based on the calculated binding rate of the MHC class II molecule of this segment to change the overall binding rate of MHC class II for the peptide without significantly reducing the therapeutic usefulness of the peptide. 2. The method according to claim 1, in which stage (c) is performed using the Bohm counting function modified to include 12-6 Van der Waals repulsive energy components of the ligand-protein complex and a ligand conformational energy component by (1) obtaining a computer library of MHC class II molecule models; (2) obtaining a computer library of a possible peptide backbone for these models of MHC class II molecules; (3) selecting a model from said first database; (4) selecting a possible peptide backbone from a second database; (5) identification of amino acid residues of the side chains that are present in each selected segment; (6) determining the value of the binding affinity for all side chains that are present in each selected segment; and optional (7) repeating steps (1) to (5) for each indicated model and each indicated side chain. 3. The method according to claim 1, in which the specified value for each aromatic side chain is approximately half the specified value for each hydrophobic aliphatic side chain. 4. The method according to claim 1, in which the selected segments of the amino acid residue overlap one to five amino acid residues. 5. The method according to claim 1, in which the consecutive selected segments of the amino acid residue substantially overlap. 6. The method according to claim 1, in which all but one of the amino acid residues in consecutive selected segments of the amino acid residue overlap. 7. A method for producing an immunogenically modified biological molecule derived from a parent molecule, wherein the modified molecule has an amino acid sequence different from the sequence of the specified parent molecule, and exhibits reduced immunogenicity relative to the parent molecule in a certain type of immune system, said method comprising (i) determining the amino acid sequence of a parent biological molecule or part thereof; (ii) the identification of one or more potential T-cell epitopes within the amino acid sequence of a protein by any method, including determining the binding of peptides to MHC molecules using in vitro methods or biological assays; (iii) the development of new variants of the sequence by changing 1-9 amino acid residues within the originally identified sequences of T-cell epitopes, these variants are modified in such a way as to significantly reduce or eliminate the activity or number of sequences of T-cell epitopes and / or the number of MHC allotypes, capable of binding peptides derived from a specified biological molecule, as determined by binding peptides to MHC molecules using in vitro methods or biologists chemical assays or binding of peptide-MHC complexes to T cells; (iv) constructing such sequence variants using recombinant DNA techniques and testing said variants in order to identify one or more variants with desired properties; and (v) optionally repeating steps (ii) to (iv), characterized in that the identification of sequences of T-cell epitopes according to stage (ii) is achieved by the method as indicated in any one of claims 1 to 7. 8. The method according to claim 7, in which one of the amino acid residues in any of the initially present sequences of T-cell epitopes is changed. 9. The method according to claim 7, in which the amino acid change is performed in relation to the homologous sequence of the protein. 10. The method according to claim 7, in which the change in amino acid residues is the replacement, division or addition of the initially present (s) amino acid (s) residue (s) with another (s) amino acid (s) residue (s) in a specific (s) position (s). 11. The method according to claim 7, in which further further change is carried out to restore the biological activity of the specified biological molecules. 12. The method according to claim 11, in which an additional further change is the substitution, addition or division of certain amino acid (s). 13. The method according to claim 7 for producing a polypeptide, protein, fusion protein, antibody or fragment thereof with reduced immunogenicity. 14. The method of claim 13, wherein said polypeptide, protein, fusion protein, or antibody is selected from the groups: (a) monoclinal antibodies: anti-40kD glycoprotein antigen antibody KS 1/4, anti-GD2 antibody 14.18, anti-Her2 antibody 4D5 (mouse) and the humanized version (Herceptin®), anti-IL-2R (anti-TAC) antibody (Zenapax®), anti-CD52 antibody (SAMRATH®), anti-CD20 antibodies (C2B8, Rituxan®; Bexxar®), antibody directed to C5 complement human protein (b) proteins: sTNF-R1, sTNF-R2, sTNFR-Fc (Enbrel®), protein C, acrp30, ricin A, CNTFR ligands, subtilisin, GM-CSF, follicle-stimulating human hormone (h-fsh), β-glycocerebrosidase, GLP-1, apolipoprotein A1. 15. An immunogenically modified biological molecule obtained from a parent molecule, where the modified molecule has an amino acid sequence different from the sequence of the specified parent molecule, and exhibits reduced immunogenicity relative to the parent molecule in a certain type of immune system, obtained by the method according to any one of claims 1-14. 16. The use of a potential T-cell epitope of a peptide within the amino acid sequence of a parent immunogenically unmodified biological molecule, identified according to any one of claims 1 to 6, to obtain a biological molecule with reduced immunogenicity and retained the desired biological activity, where the T-cell epitope is 13-dimensional peptide. 17. The use of a peptide sequence consisting of at least 9 consecutive amino acid residues of a 13-dimensional epitope of T cells, as described in clause 16, to obtain a biological molecule with reduced immunogenicity compared to the parent unmodified molecule, and having biological activity .

Images