WO1998014208A1

WO1998014208A1 - Reactive ligands and covalent ligand-protein complexes

Info

Publication number: WO1998014208A1
Application number: PCT/US1997/017483
Authority: WO
Inventors: Don C. Wiley; Stuart L. Schreiber; Robert J. Valentekovich; Gregory A. Weiss; Soroosh Shambayati
Original assignee: President And Fellows Of Harvard College
Priority date: 1996-09-30
Filing date: 1997-09-30
Publication date: 1998-04-09

Abstract

Ligands bind to proteins according to particular kinetic profiles. The reversible kinetics of binding lead to release of the ligand from the binding site, thereby affecting the recognition of the ligand-protein complex. By changing the kinetics of the reversible binding, it is possible to alter the response to the cellular immune system. Reactive ligands can be used to change the kinetics of binding by reacting with the protein when bound.

Description

REACTIVE LIGANDS AND COVALENT LIGAND-PROTEIN COMPLEXES

BACKGROUND OF THE INVENTION Ligand protein interactions influence many factors of biological systems. Ligands bind to proteins according to particular kinetic profiles. For example, antigenic peptides of T cells first bind to class I or class II major histocompatibility complex (MHC) molecules before interacting with T cell receptors on CD⁸⁺ T cells or CD⁴⁺ T cells to elicit immune responses. Stern, L.J. & Wiley, D.C. (1994), Structure 2, 245-251. In general, conservative residue substitutions in the middle portion of antigenic peptides of T cells are introduced to selectively inhibit activation of T cells without significantly altering binding affinity to the relevant MHC molecules. See, Sloan-Lancaster, J. , Evavold, B.D. & Allen, P.M. (1993) Nature 363, 156-159; De Magistris, M.T., Alexander, J. , Coggeshall, M., Altman, A., Gaeta, F.C.A., Grey, M.H. & Sette, A. (1992) Cell 68, 625-634; and Alexander, J., Snoke, K. , Ruppert, J., Sidney, J., Wall, M., Southwood, S., Oseroff, C, Arrhenius , T., Gaeta, F.C.A., Colon, S.M., Grey, H.M. & Sette, A. (1993) J. Immunol. 150, 1-7. Other designs of MHC-blocking peptides have included the replacement of several residues from the middle portion of antigenic peptides with moieties such as 4-aminobutyric acid, 6-aminohexanoic acid, and phenanthridine to act as spacers. E.g., see Rognan, D., Scapozza, L., Folkers, G. & Daser, A. (1995) Proc . Natl . Acad. Sci . USA 92, 753-757.

The reversible kinetics of binding lead to the release of the ligand from the binding site, thereby influencing recognition of the ligand-protein complex. By changing the kinetics of binding it is possible to alter the response to the cellular immune system. For example, covalent complexes of peptides with MHC molecules have been constructed previously by generating recombinant proteins with elongated N-termini that extended into the binding grooves of MHC class I (see, e.g., Mottez, E., Langlade-Demoyen, P., Gournier, H., Martinon, F., Maryanski , J. , Kourilsky, P. & Abastado, J.-P. (1995) Journal of Experimental Medicine 181, 493- 502) and MHC class II (see, e.g., Kozono, H. , White, J., Clements, J. , Marrack, P. & Kappler, J. (1994) Nature 369, 151-154) molecules. These molecules are active and have been suggested as possible therapeutics to induce T- cell tolerance in cases where the peptide autoantigens can be identified. See, for example, Kozono, H. , White, J., Clements, J. , Marrack, P. & Kappler, J. (1994) Nature 369, 151-154 and Sharma, S. D., Nag, B., Su, X.-M., Green, D., Spack, E., Clark, B.R. & Sriram, S. (1991) Proc. Natl. Acad. Sci . USA 88, 11465-11469.

SUMMARY OF THE INVENTION The present invention features compounds that bind as ligands to a binding region of a protein. The compounds react with residues in the binding region, selectively forming a covalent bond between the protein and the ligand. The formation of the covalent bond changes the reversibility of the binding kinetics between the ligand and the protein. In one aspect, the invention relates to a compound of the formula (I)

where each of R-. and R₂, independently, is hydroxy, halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, alkyl carboxylate, a Michael acceptor, amino, alkylamino, ammonium, or alkyl ammonium, or together R_x and R₂ are 0, NH, NH₂X or NZ, where Z is alkyl or aryl; X is C_x to C₆ alkyl, substituted aryl, or deleted; R₃ is H, alkyl, or aryl; R₄ is alkyl, aryl, -NH- W, or -C(=0)-Y, where W is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y is alkyl, alkoxy, amino, substituted amino, hydroxyl , a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group; and R₅ is H, alkyl, aryl, -NH-W' , or -C(=0)-Y', where W' is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y' is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group, or an amino acid residue bonded at its amino group .

An alkyl group is a branched or unbranched _λ to C₆ alkyl or benzyl, and aryl is a substituted or unsubstituted C₆ to C₁₄ carbocyclic or heterocyclic aromatic having one or two rings. A substituted aryl is an aryl bearing one or more Cj_. to C₆ substituents including alkyl, alkoxy, or acyl groups, or halogen, hydroxy, amino, or amido groups. A substituted amino is an amino group having one or two branched or unbranched C_x to C₆ alkyl or benzyl groups. A Michael acceptor (e.g., H₂C=CH-C (=0) - , and related unsaturated groups) is a group that undergoes conjugate addition with a nucleophile.

Preferably, R-_^ is halogen, alkyl sulfonate, carboxylate, or aryl sulfonate and R₂ is hydroxy, amino, alkylamino, ammonium, or alkyl ammonium, or together R_λ and R₂ are 0, NH₂ ⁺, or NH; X is a C₂ to C₄ alkyl; R₃ is H; R₄ is -NH-W, where W is an amino acid residue bonded at its carbonyl group or a peptide moiety bonded at its terminal carbonyl group; and R₅ is a -C(=0)-Y', where Y' is a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group.

In preferred embodiments, W is an amino acid residue bonded at its carbonyl group and Y' is a peptide moiety bonded at its terminal amino group. In other preferred embodiments, R_x is alkyl sulfonate and R₂ is an ammonium, or together R_x and R₂ are O, or NH; X is C₄ alkyl; and the amino acid residue is a glycine residue.

In preferred embodiments, a portion of the peptide moiety is identical to a portion of a selected peptide or an antigenic peptide. The peptide moiety can preferably contain between 2 and 10 amino acid residues. Most preferably, the compound is G(az)IDKPILK or G (az) AFVTIGK, where (az) is an unnatural amino acid residue where the side chain is a C₆ alkyl having a terminal aziridine group . A further aspect of this invention relates to a method of forming a ligand-protein complex. The method includes the steps of providing a protein having an arginine-specific binding region, the binding region including a nucleophilic residue, and adding a ligand to bind with the protein at the arginine-specific binding region, the ligand having a moiety which reacts with the nucleophilic residue. The moiety of the ligand and the nucleophilic residue of the protein form a covalent bond between the ligand and the protein. In preferred embodiments, the arginine-specific binding region includes a thiol, a thiol and a carboxyl, or two carboxyl groups, and the moiety is halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, carboxylate, a Michael acceptor, aziridine, or epoxide. Most preferably, the arginine-specific binding region includes a thiol and a carboxyl and the moiety is aziridine, or epoxide.

In other preferred embodiments, the step of providing the protein includes the step of altering a binding region of the protein to introduce the arginine- specific binding region.

Another aspect of this invention relates to a covalently linked ligand-protein complex including a protein having an arginine-specific binding region and a ligand covalently bonded to the binding region of the protein, said ligand having the formula

where one of R and R₂ is hydroxy, alkyl, alkyl carbonyl, amino, alkyl amino, ammonium, or alkyl ammonium, and the other of R_x and R₂ is the covalent bond to the protein; X is C_]_ to C₆ alkyl, substituted aryl, or deleted; R₃ is H, alkyl, or aryl; R₄ is alkyl, aryl, -NH-W, or -C(=0)-Y, where W is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group; and R₅ is H, alkyl, aryl, -NH-W', or -C(=0)-Y', where W' is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y' is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group, or an amino acid residue bonded at its amino group.

In preferred embodiments, the arginine-specific binding region includes a thiol, a thiol and carboxyl, or two carboxyl groups; R_x is the covalent bond between the ligand and the protein; R₂ is hydroxy, alkyl, alkyl carbonyl, amino, alkyl amino, ammonium, or alkyl ammonium; R₃ is H; and X is a C₂ to C₄ alkyl. Most preferably, R₂ is the covalent bond between the ligand and the protein (i.e., a thioether or ester linkange) . The invention includes one or more of the following advantages. The ligand selectively reacts with groups in the binding region of the protein. Thus, undesired reactions outside of the binding region are unlikely to occur. In addition, the covalent nature of the linkage between the ligand and the protein changes the kinetics of ligand binding dramatically by making the ligand binding essentially irreversible.

A "selected peptide" is any peptide or variant thereof that is associated with an undesired physical condition, such as, but not limited to, an immune system- selective peptide (i.e., a peptide that induces an immune response) . Note that "an antigenic peptide" mentioned herein can be either a naturally occurring peptide or a variant thereof (e.g., containing an acetylated alpha- amino group, an amidated alpha-carboxyl group, a D- and/or beta- or gamma-amino acid residue, or any suitable organic moiety) ; indeed, there are only two criteria, i.e., it is a peptide and it binds to a class I or class II MHC molecule in the manner described in Stern, L.J. & Don C. Wiley, Structure (1994) 2, 245-251.

Other features or advantages of the present invention will be apparent from the following drawings and detailed description of representative examples. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic diagram depicting the synthesis and structure of a compound according to the invention. Fig. 2 is a drawing of a model of the P2 pocket of HLA-B27 binding arginine (Fig. 2A) , a model of an aziridine-containing ligand (Fig. 2B) and a covalently bonded residue (Fig. 2C) .

Fig. 3 is a graph of three HPLC runs depicting the stability of the aziridine ligand to free thiols in solution.

Fig. 4 is an image of an SDS-PAGE (15%) of HLA-B27 complexed with the aziridine-containing peptide (G(az) IDKPILK) (lane 1) and the control peptide (GRIDKPILK) (lane 2) .

DETAILED DESCRIPTION OF THE INVENTION Ligands bind to proteins according to particular kinetic profiles. Short peptides containing arginine at peptide position 2 (P2) bind to arginine-specific binding regions of class I major histocompatibility complex (MHC) glycoproteins, such as HLA-B27. The HLA-B27/peptide complex is recognized by particular T cells. This recognition leads to both the development of the repertoire of T cells in the cellular immune system and the activation of cytotoxic T cells. The reversible kinetics of binding lead to release of the ligand from the binding site, thereby affecting the recognition of the ligand-protein complex. By changing the kinetics of the reversible binding, it is possible to alter the response to the cellular immune system. Reactive ligands can be used to change the kinetics of binding by reacting with the protein when bound.

Suitable reactive ligands for changing the kinetics of binding with protein substrates are compounds of the formula I. Each of R₁ and R₂, independently, is hydroxy, halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, carboxylate, a Michael acceptor, amino, alkylamino, ammonium, or alkyl ammonium. Alternatively, together R-. and R₂ are 0, NH, NH₂ ⁺, or NZ, where Z is alkyl, or aryl, thus forming an epoxide, aziridine or substituted aziridine ring. X, a spacer group, is C to C₆ alkyl, substituted aryl, or deleted. R₃ is H, alkyl, or aryl. R₄ is alkyl, aryl, -NH-W, or -C(=0)-Y, where W is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group. R₅ is H, alkyl, aryl, -NH-W', or -C(=0)-Y', where W' is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y' is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group, or an amino acid residue bonded at its amino group.

In particular, one of R_: and R₂ is a reactive group that will react with a nucleophile (e.g., -SH, -NH₂, -OH, or -C00χ . Thus, one of the groups is a leaving group or contains a leaving group. The leaving group can be a halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, alkyl carboxylate, ammonium, alkyl ammonium, epoxide, aziridine, or substituted aziridine, where the alkyl group is a branched or unbranched C_x to C₆ alkyl or benzyl and aryl is a substituted or unsubstituted C₆ to C₁₄ carbocyclic or heterocyclic aromatic having one or two rings. Alternatively, R_x and R₂ is a Michael acceptor (e.g., H₂C=CH-C (=0) -, and related unsaturated groups) which undergoes conjugate addition with a nucleophile. The reactions between the group at R_x or R₂ and the nucleophile take place at a binding region of a protein.

The compounds of the invention can be used to prepare reactive ligands, or are themselves reactive ligands that can be used to make covalent ligand-protein complexes in vi tro or in vivo . At R₄ and R₅, is a reactive group or the peptide moiety or amino acid residue is ultimately a group that binds selectively to a protein. In particular, a portion of the peptide moiety, the amino acid residue, or a combination thereof is a selected peptide or an antigenic peptide, which can be either a naturally occurring peptide or a variant thereof (e.g., containing an acetylated alpha-amino group, an amidated a-carboxyl group, a D- and/or beta- or gamma- amino acid residue, or any suitable organic moiety) . In particular, a portion of the peptide moiety, the amino acid residue, or their combination, is a peptide and it binds to a class I or class II MHC molecule in the manner described in Stern, L.J. & Wiley, D.C., Structure (1994) 2, 245-251. For example, short peptides containing arginine at peptide position 2 (P2) bind to arginine- specific binding regions of class I major histocompatibility complex (MHC) glycoproteins, such as HLA-B27. The HLA-B27/peptide complex is recognized by particular T cells.

Fig. 1 is a schematic diagram depicting the synthesis and structure of a compound according to the invention. Abbreviations are Boc, butoxycarbonyl ; Ph, phenyl; DEAD, diethyl azodicarboxylate; THF, tetrahydrofuran; DMS, dimethyl sulfide; TFA, trifluoroacetic acid; Gly, glycine; BOP, benzotriazol-1- yloxy-tris (dimethylamino)phosphonium hexafluorophosphate; HOBT, 1-hydroxybenzotriazole; DIEA, diisopropyl ethylamine; MCPBA, /π-chloroperoxybenzoic acid; EtOAc, ethyl acetate; Ms, mesyl; Me, methyl; HBTU 2-(lH- benzotriazol-1-yl) -1,1,3, 3-tetramethyluronium hexafluorophosphate; and DMF, dimethylformamide.

The first two steps shown in Fig. 1 can be modified to prepare variants of Compound 1, where the length of the alkenyl group can be varied, or other functionalities can be added using ordinary synthetic methods. Boc-amine (Compound 1) was prepared according to the methods described in Arnold, L.D., Kalantar, T.H. & Vederas, J.C. (1985) J. Am. Chem. Soc. 107, 7105-7109, and Arnold, L.D., Drover, J.C. G. & Vederas, J.C. (1987) J. Am. Chem. Soc. 109, 4649-4659 and the dipeptides were prepared according to the methods described in Arnold, L.D., Kalantar, T.H. & Vederas, J.C. (1985) J. Am. Chem. Soc. 107, 7105-7109, and Arnold, L.D., Drover, J.C.G. & Vederas, J.C. (1987) J. Am. Chem. Soc. 109, 4649-4659. The olefinic group of Compound 1 can be transformed by ordinary methods, e.g., addition reactions. The subsequent reactions of Compound 1 and its analogues can be carried out to link other natural or unnatural amino acid residues or peptide moieties.

A ligand-protein complex is formed by exposing a protein having an arginine-specific binding region which includes a nucleophilic residue to a ligand which binds with the protein at the arginine-specific binding region. The ligand is a compound of the invention, thus having a moiety which reacts with the nucleophilic residue. Upon binding, the reactive moiety of the ligand and the nucleophilic residue of the protein form a covalent bond between the ligand and the protein, resulting in a covalently linked ligand-protein complex where the protein and ligand become linked at R_x or R₂, changing the kinetics of release of the ligand from the binding site. As an example, this influences the recognition of the ligand-protein complex by the cellular immune system, making it possible to alter immune responses. Cytotoxic T-lymphocytes (CTLs) recognize antigens as short peptides bound to MHC class I molecules that consist of a heavy chain (H) and /3₂-microglobulin (jS₂-m) . For a review of CTLs, see Germain, R.N. & Margulies, D.H. (1993) A. Rev. Immun. 11, 403-450. The structure of HLA- B27 is known. See, Madden, D.R., Gorga, J.C, Strominger, J.L. & Wiley, D.C. (1991) Nature 353, 321- 325, Gorga, J.C, Madden, D.R., Prendergast, J.K., Wiley, D.C. &. Strominger, J.L. (1992) Proteins 12, 87-90, and Madden, D.R. , Gorga, J.C, Strominger, J.L. & Wiley, D.C. (1992) Cell 70, 1035-1048.

The involvement of HLA-B27 has been implicated in, for example, the autoimmune disorder ankylosing spondylitis. See, Brewerton, D.A., Caffrey, M., Hart, F.D., James, D.C.O., Nicholls, A. & Sturrock, R.D. (1973) Lancet 1, 904-907, Schlosstein, L., Terasaki, P.I., Bluestone, R. Sc Pearson, CM. (1973) N. Engl . J. Med. 288, 704-706, and Hammer, R.E., Maika, S.D., Richardson, J.A., Tang, J.-P. S. Taurog, J.D. (1990) Cell 63, 1099- 1112. The autoimmune response involves binding of an autoantigen peptide by HLA-B27. Endogenous and viral peptides that bind to HLA-B27 have an arginine in their P2 position, as described by Jardetzky, T.S., Lane, W.S., Robinson, R.A. , Madden, D.R. & Wiley, D.C. (1991) Nature 353, 326-329. The d-guanidinium group of the arginine is bound within a deep pocket where four polymorphic residues project toward it. See, Madden, D.R., Gorga, J.C, Strominger, J.L. & Wiley, D.C (1991) Nature 353, 321-325, and Madden, D.R., Gorga, J.C, Strominger, J.L. Sc Wiley, D.C. (1992) Cell 70, 1035-1048.

Activation of T cell receptors on CD8⁺ and CD4⁺ T cells requires the recognition of antigenic peptides bound in the groove of MHC molecules on the surface of antigen-presenting cells. Antigenic peptides bind in an extended conformation to class I and class II MHC molecules. X-ray crystallography studies have revealed that for all class I MHC complexes, the peptide N- and C- termini form hydrogen bonds with conserved MHC amino acid residues located at both ends of the binding site, some peptide side chains are buried in deep pockets along the binding site, and the central region of the peptide arch away from the floor of the binding site with side chains generally pointing toward T cell receptors. .Antigenic peptides of 8- to 10-amino acids long have been shown to adopt this mode of binding. A decamer has been shown to extend out of the binding groove at the C-terminus and an octamer has been shown to preferentially leave the N- terminal end of the binding site unoccupied. On the basis of these observations, other modes of binding can be described: extension of a decamer out of the N- terminal end of the binding site and binding of an octamer preferentially at the N-terminal end of the groove leaving the C-terminal end of the groove unoccupied. In contrast to class I MHC molecules, class II MHC molecules bind to longer antigenic peptides. This difference in length can be ascribed to structural differences between the two binding sites and the position of key MHC amino acid residues that interact with the peptide backbone. The few X-ray structures of class II MHC complexes that have been determined so far indicate that the antigenic peptide is stretched out along most of its length and adopts a bound conformation characteristic of a type II polyproline helix. The peptide' s N- and C-termini extend out of the binding groove. Side chains of some amino acid residues bind specifically in pockets made of polymorphic MHC amino acid residues while other side chains, spaced approximately three amino acid residues apart, point away from the binding site. Interaction of T cell receptors with both types of MHC complexes involves amino acid residues on both the peptide and MHC molecules. For class I MHC complexes, amino acid residues from the central portion of the peptide, where side chains are most exposed, play a more crucial role in this recognition event, whereas for class II MHC complexes, amino acid residues having side chains pointing up are found along the entire sequence. In designing peptides of this invention, it is preferable to select as anchor amino acid residues those residues with side chains pointing toward T cell receptors.

References for the structures of class I MHC molecules include: Madden, D.R., Garboczi, D.N. Sc Wiley, D.C. (1993) Cell 75, 693-708; Bjork an, P.J., Saper, M.A., Samraoui, B., Bennett, W.S., Strominger, J.L. & Wiley, D.C. (1987) Nature 329, 506-512; Madden, D.R., Gorga, J.C, Strominger, J.L. Sc Wiley, D.C (1991) Nature 353, 321-325; Silver, M.J., Guo, H.-C, Strominger, J.L. Sc Wiley, D.C (1992) Nature 360, 367-369; Guo, H.-C, Jadertzky, T.S., Garrett, T.P.J., Lane, W.S., Strominger,

J.L. & Wiley, D.C. (1992) Nature 360, 364-366; and

Madden, D.R., Gorga, J.C, Strominger, J.L. & Wiley, D.C.

(1992) Cell 70, 1035-1048. References for the structures of class II MHC molecules include Brown, J.H., Jardetzky, T.S., Gorga, J.C, Stern, L.J., Urban, R.G., Strominger, J.L., and Wiley, D.C. (1993) Nature 364, 33-39; Stern L.J. and Wiley, D.C. (1994) Structure 2, 245-251; and Stern, J.L., Brown, J.H., Jardertzky, T.S., Gorga, J.C, Urban, R.G., Strominger, J.L., and Wiley, D.C. (1994) Nature 368, 215-221.

According to the structural model of the binding site shown in Fig. 2A, His-9, Thr-24, Glu-45, and a bound water form a hydrogen bond network with the guanidinium group of the ligand. The P2 pocket of HLA-B27 forms a planar network of hydrogen bonds to the guanidinium group of arginine at P2 of a bound peptide. Perpendicular to the planar network of hydrogen bonds, the free thiol of Cys-67 is separated from the guanidinium group by about 3.7 A. The sulfhydryl group of Cys-67 is located directly above the plane of the guanidinium group of the ligand.

The arrangement of protein side chains (amino acid residues) suggested that it might be possible to induce the pocket to act as an enzyme active site, by designing a suitable substrate. A ligand was synthesized with a reactive side chain designed to mimic arginine. Although other reactive groups at this position can be used, an aziridine is most preferred. The reactive ligand contains an aziridine and binds covalently by reaction of the aziridine upon binding in the arginine-specific P2 pocket of HLA-B27. A peptide was designed such that a covalent bond with HLA-B27 will form between the ligand and the protein spontaneously and selectively upon binding. An aziridine-containing amino acid mimic of arginine was selected (Fig. 2B) . According to Fig. 2B, the aziridine-containing ligand can bind in place of the guanidinium group with the aziridine amine, forming a hydrogen bonded salt bridge to Glu-45. In this model with one diastereomer of the aziridine depicted, the free thiol of Cys-67 is poised for in-line nucleophilic attack of the aziridine group.

Fig. 2C shows the expected geometry of the covalent complex between Cys-45 and the aziridine- containing peptide that results from the nucleophilic attack of the thiol which opens the aziridine group. The positively charged secondary amine of the aziridine ring is expected to form a hydrogen-bonded charged-pair with Glu-45 in the arginine-specific pocket, positioning the eta-carbon of the aziridine ring for an in-line, ring- opening attack by the thiolate of Cys-67 (Fig. 2C) . Selectivity is expected from the polarizability of the aziridine by Glu-45 and the propinquity of the Cys-67 thiol to the bound aziridine ring.

Using tryptic digestion followed by mass spectrometry and amino acid sequencing, the aziridine- containing ligand is shown to alkylate specifically cysteine 67 of HLA-B27. Neither free cysteine in solution nor an exposed cysteine on a class II MHC molecule can be alkylated, showing that specific recognition between the anchor side-chain pocket of an MHC class I protein and the designed ligand (propinquity) is necessary to induce the selective covalent reaction with the MHC class I molecule.

The aziridine within the ligand was cyclized from the mesylate in solution, before addition to the folding buffer containing recombinantly expressed, purified HLA- B27 heavy chain and β₂- .

In general, conservative residue substitutions in the middle portion of antigenic peptides of T cells are introduced to selectively inhibit activation of T cells without significantly altering binding affinity to the relevant MHC molecules. See, Sloan-Lancaster, J. , Evavold, B.D. & Allen, P.M. (1993) Nature 363, 156-159; De Magistris, M.T., Alexander, J. , Coggeshall, M. , Altman, A., Gaeta, F.C.A., Grey, M.H. & Sette, A. (1992) Cell 68, 625-634; and Alexander, J. , Snoke, K. , Ruppert, J., Sidney, J. , Wall, M., Southwood, S., Oseroff, C, Arrhenius, T., Gaeta, F.C.A., Colon, S.M., Grey, H.M. Sc Sette, A. (1993) J. Immunol. 150, 1-7. Activation and regulation of immune responses (e.g., autoimmune responses, responses to viral infection, and responses to cancerous cells) can be influenced by the compounds of the invention, since the compounds bind to specific protein binding pockets irreversibly. It is possible to activate T cell responses by treatment with a preformed ligand-protein complex or by treatment with the reactive ligand or a precursor thereof.

Without further elaboration, it is believed that one skilled in the art can, based on the description herein, utilize the present invention to its fullest extent. All publications disclosed herein are incorporated by reference. The specific examples set forth below are, therefore, to be construed as merely illustrative, and not limitative of the remainder of the disclosure in any way whatsoever.

Ligand Synthesis

Nonamer peptides containing the aziridine were synthesized by coupling the dipeptide to synthetic heptamers corresponding to the seven C-terminal amino acids of two peptides known to bind to HLA-B27, GRIDKPILK and GRAFVTIGK. See, Jardetzky, T.S., Lane, W.S., Robinson, R.A., Madden, D.R. & Wiley, D.C (1991) Nature 353, 326-329. The peptide sequence of the ligand shown in Fig. 1 and described here is based upon a ribosomal peptide GRIDKPILK. Other peptide sequences and analogues can be substituted in the scheme shown in Fig. 1 to afford other reactive ligands. Diastereomers of the aziridine were unresolved. All reactions followed standard laboratory practices. All intermediates and products were characterized by ^XH NMR, ¹³C NMR, IR and fast atom bombardment mass spectrometry (F.AB MS) .

Boc-amine (Compound 1) . Boc-amine Compound 1 was prepared according to the methods described in Arnold, L.D., Kalantar, T.H. & Vederas, J.C. (1985) J. Am. Chem. Soc. 107, 7105-7109, and Arnold, L.D., Drover, J.C.G. & Vederas, J.C. (1987) J. Am. Chem. Soc. 109, 4649-4659.

Boc-Glycine- (2-amino-hept-6-enoic acid) dipeptide methyl ester (Compound 2) . To a solution of Compound 1 (211 mg, 0.911 mmol) in 5.0 mL of CH₂C1₂ was added 5.0 mL of trifluoroacetic acid. The mixture was swirled, allowed to stand for 20 minutes, diluted with 4 mL of toluene, and concentrated in vacuo to afford the crude deprotected amine. The deprotected amine was dissolved in 5.0 mL of CH₂C1₂. Diisopropyl ethylamine (0.476 mL, 0.273 mmol) was added to the solution containing the deprotected amine, followed by Boc-glycine (175 mg, 1.00 mmol) and benzotriazol-1-yloxy- tris (dimethylamino)phosphonium hexafluorophosphate (442 mg, 1.00 mmol) . The mixture was stirred for 6 hours and then quenched by adding 5 drops isopropyl amine. The mixture was concentrated in vacuo and chromatographed (40% ethyl acetate :hexanes) to give 178 mg (0.542 mmol, 59%) of Compound 2.

Epoxy Dipeptide (Compound 3 ) . To a solution of Compound 2 ( 145 mg , 0 . 441 mmol ) in 7 . 0 mL of CH₂C1₂ was added m-chloroperoxybenzoic acid (91.4 mg, 0.529 mmol). The mixture was stirred for 9 hours and then diluted with 40 mL of ethyl acetate, washed with 10% aqueous NaS₂0₅ (20 mL) , washed with saturated NaHCO-, solution (2 x 20 mL) , washed with brine (10 mL) , dried over MgS0₄, concentrated in vacuo, and chromatographed (60% ethyl acetate :hexanes) to give 91.0 mg (0.264 mmol, 50%) of epoxy dipeptide Compound 3.

Compound 3 was characterized by: IR (Film) : n 2934, 1744, 1719, 1678, 1524, 1456, 1437, 1412, 1391, 1368, 1250, 1211, 1171, 1051, 1030, 939, 864, 835, 781, 615 cm^"1; Η NMR (CDC1₃) : d 1.41 (s) , 1.48 (m) , 1.52 (m) , 1.62 (m) , 1.82 (m) , 2.40 (q) , 2.69 (t) , 2.84 (t) , 3.69 (s) , 3.78 (m) , 4.57 (q) , 5.26 (d) , 6.71; and ¹³C NMR (CDC1₃) : d 24.82, 25.32, 25.38, 28.16, 31.94, 44.10, 46.77, 51.87, 51.96, 52.20, 79.92, 156.02, 169.41, 172.59.

Azido Dipeptide (Compound 4) .

Compound 3 (66.0 mg, 0.191 mmol) in 5.0 L of a 9:1 methanol :water mixture was added NH₄C1 (80 mg, 1.49 mmol) and NaN₃ (300 mg, 4.61 mmol) . The mixture was heated to reflux and stirred for 2 hours. The mixture was then cooled to 25°C, diluted with 40 mL of ethyl acetate, washed with water (2 x 20 mL) , washed with brine (10 mL) , dried over MgS0₄, and concentrated in vacuo to give 71.0 mg (0.191 mmol, 100%) of azido dipeptide Compound 4.

Compound 4 was characterized by: IR (Film) n

3328, 2976, 2936, 2865, 2101, 1742, 1671, 1524, 1439, 1393, 1368, 1279, 1252, 1217, 1167_; 1051, 1030, 941, 864, 632 cm^"1; ^XH NMR (CDC1₃) : d 1.28 (m) 1.38 (s) , 1.61 (m) , 1.79 (m) , 1.82 (d) , 3.17 (m) , 3.24 (m) , 3.66 (s) , 3.74 (m) , 4.52 (m) , 5.49 (t) , 6.90 (d) ; and ¹³C NMR (CDC1₃) : d 24.60, 24.72, 24.82, 28.19, 32.01, 33.80, 44.09, 51.75, 51.89, 52.34, 56.86, 64.84, 70.12, 70.38, 80.18, 156.16, 169.59, 172.72. Boc-amine Dipeptide (Compound 5) . A 5% solution of Pd on carbon in ethyl acetate (6.0 mL) was stirred at room temperature, under atmospheric pressure of hydrogen gas, for one hour. Addition of Compound 4 (82.3 mg, 0.24 mmol) and Boc-anhydride (104.3 mg, 0.48 mmol) required 1.0 mL ethyl acetate, followed by two ethyl acetate washes of 1.0 mL each. After stirring 12 hours, and Celite was added. The solution was filtered and concentrated in vacuo . The residue was purified by flash chromatography, running neat ethyl acetate, to yield 86.6 mg (0.188 mmol, 78%) of Boc-amine dipeptide Compound 5.

Compound 5 was characterized by: IR (Film) : n 3328, 2978, 2934, 2865, 1744, 1686, 1524, 1456, 1439, 1412, 1393, 1368, 1275, 1250, 1169, 1051, 1030, 864, 781, 735 cm^"1; ^JH NMR (CDCl₃) : d 1.39 (s) , 1.41 (s) , 1.48 (t) , 1.64 (m) , 1.81 (m) , 2.92 (m) , 3.22 (br s) 3.60 (d) , 3.69

(s) , 3.77 (d) , 4.55 (m) , 5.13 (br s) , 5.43 (d) , 6.86 (br s) ; and ¹³C NMR (CDC1₃) : d 24.56, 24.73, 24.95, 28.08, 31.71, 33.90, 43.82, 46.35, 51.75, 52.15, 70.38, 70.63, 79.06, 79.84, 156.11, 156.56, 169.76. Mesylate Dipeptide Methyl Ester (Compound 6) .

Compound 5 (96.0 mg, 0.208 mmol) was weighed into a pear shaped flask and azeotroped with xylenes, before being dissolved in dry methylene chloride (3.5 mL, 0.06 M) . After cooling to 0°C, mesyl anhydride (174.2 mg, 0.312 mmol) was added. Approximately 20 minutes later, the reaction was complete and was quenched by the addition of water. Excess ethyl acetate was added. The organic phase was washed with 5% aqueous HC1 , saturated aqueous sodium bicarbonate, and brine. Following drying over MgS0₄, the mesylate dipeptide methyl ester Compound 6 was isolated in vacuo.

NHBoc

Mesylate Dipeptide Carboxylic Acid (Compound 7) .

Compound 6 was dissolved in methanol (0.56 mL) and tetrahydrofuran (2.25 mL) , and cooled to 0°C. .An aqueous solution of IN LiOH (0.56 mL) was added, and the reaction was stirred for half an hour. The reaction was quenched by the addition of excess ethyl acetate and IN aqueous HCl (0.59 mL) . The organic phase was isolated in vacuo and azeotroped with xylenes to afford the mesylate dipeptide carboxylic acid Compound 7.

Mesylate Polypeptide (Compound 8). Compound 7, prepared in the step above, was added in N-methyl pyrrolidone to Wang resin coupled to the sequence 0 H₂N-IDKPILK. 1-Hydroxybenzotriazole (61.2 mg, 0.4 mmol), 2- (lH-benzotriazol-1-yl) -1,1,3, 3-tetramethyluronium hexafluorophosphate (83.4 mg, 0.22 mmol), and Hunig's base (0.061 mL, 0.47 mmol) were added, and the resin was shaken for one hour at room temperature to give the 5 resin-bound mesylate polypeptide. The polypeptide was cleaved from the resin using 95% trifluoroacetic acid in water to afford the mesylate polypeptide Compound 8. Compound 8 was purified by reverse phase HPLC, running a gradient from 85:15 to 50:50 (water with 0.1% 0 trifluoroacetic acid:methanol) . Mass spectrometry confirmed the identity of Compound 8 (MH⁺ 1133) .

Aziridine Polypeptide (Compound 9) . Compound 8 was treated with sodium hydroxide (10 equivalents at room temperature for half an hour) , or lithium hydroxide in methanol, to afford aziridine polypeptide Compound 9. Mass spectrometry confirmed the identity of Compound 9 (MH⁺ 1055) . Additionally, Compound 9 was subjected to Edman degradation and microsequencing, which confirmed its composition to be G (az) IDKPILK, where (az) is the unnatural aziridine amino acid residue.

The polypeptide G(az)AFVTIGK was prepared using a different resin-bound peptide. The aziridine within a peptide was cyclized from the mesylate in solution before addition to the folding buffer containing recombinantly expressed, purified HLA-B27 heavy chain and β₂-m .

8

LiOH MeOH

NH, Ligand-Protein Complexation and Complexes

Specificity of the Aziridine-containing Ligand for HLA-B27. The selectivity of the aziridine-containing peptide for the free thiol on HLA-B27 was demonstrated in three ways .

First, neither the aziridine-peptide nor its mesylate precursor reacted with a 200 molar equivalent excess of the amino acid components of the P2 binding pocket: the free amino acids cysteine, threonine, or glutamic acid. To 950 mL of HLA-B27 folding buffer, the unprotected amino acids cysteine, threonine and glutamic acid were added each to a final concentration of 20 mM. 220 mg of mesylate Compound 8 was made basic in 50 mL methanol by the addition of 2 mL IN NaOH. After 15 minutes, the solution contained approximately equimolar distributions of mesylate precursor Compound 8 and aziridine-containing ligand Compound 9 (verified by reverse phase HPLC) and the solution was added to the folding buffer (both Compound 8 and Compound 9 have a final concentration of 0.1 mM in the folding buffer), which was incubated at 10 °C After half an hour and approximately every 12 hours later, 100 mL of the folding buffer was removed and subjected to reverse phase HPLC, running a gradient of 100:0 to 40:60 (0.1% trifluoroacetic acid in water :acetonitrile) over 30 minutes and monitoring at 214 nm.

Elution profiles of reverse phase HPLC after designated incubation times show constant levels of the aziridine and mesylate containing peptides, as shown in Fig. 3. The reactivity was accessed in a time course experiment over three days. The stabilities of the ligands (Compound 8 and Compound 9) were established in aqueous solution over that time period, although a small fraction of the mesylate (Compound 8) gradually converted to the aziridine (Compound 9) . The aziridine-nonameric- peptide (G (az) IDKPILK) and its mesylate precursor are stable and do not react with free cysteine, threonine, or glutamic acid. The unprotected amino acids cysteine, threonine, and glutamic acid (each present in 200-fold molar excess) were incubated with equal molar amounts of the aziridine-containing ligand and its mesylate precursor in the HLA-B27 folding buffer containing reduced glutathione (50-fold molar excess of free thiol) . Over three days, the sum of the mesylate precursor and aziridine-containing ligand remained approximately the same. No degradation or reaction products of the aziridine or mesylate were observed. Peaks at 15.63 and 16.05 minutes correspond to the aziridine-containing ligand and its mesylate precursor, respectively, other peaks belong to the folding buffer. Peaks were assigned by injection of standards or FAB-MS. The ordinate is proportional to absorbance with arbitrary units.

Second, the aziridine-containing-peptide failed to alkylate the free cysteine residue in the peptide-binding groove of the class II MHC molecule HLA-DR1. See, Stern, L.J. & Wiley, D.C. (1992) Cell 68, 465-477. Following standard peptide loading procedures for class II MHC molecules described in Stern, L.J. Sc Wiley, D.C. (1992) Cell 68, 465-477, empty, folded HLA-DR1 was incubated at 37°C with an eighteen fold molar excess of the aziridine- containing-peptide, G(az) IDKPILK, Compound 9. Aliquots of the mixture were removed daily for 3 days, denatured in 6M guanidine hydrochloride, and tested for the presence of free thiol using Ellman's reagent. The portion of HLA-DR1 containing the only free cysteine, Cys-30, does not bind arginine and lacks the glutamate and threonine of the arginine binding pocket in HLA-B27 (see, Stern, L.J., Brown, J.H. , Jardetzky, T.S., Gorga, J.C, Urban, R.G. , Strominger, J.L. Sc Wiley, D.C. (1994) Nature 368, 215-221), nevertheless it represents a free thiol accessible on the surface of a protein. The free thiol of Cys-30 remained unalkylated and its concentration was constant, throughout the course of the experiment. As a control the addition of one molar equivalent of iodoacetic acid to the denatured HLA-DR1, after 2 hours, was shown to decrease measured amounts of free thiol by greater than 50%, demonstrating that the response to Ellman's reagent was due to the presence of a free thiol . Third, the aziridine ligand alkylates Cys-67 during the folding of recombinant HLA-B27 despite a 5000 molar equivalent excess of cysteine in the glutathione of the refolding buffer. Each complex was folded by incubation at 10°C of HLA-B27 heavy chain (1 mM) and β₂ -m (2 mM) in a N₂ (g) sparged solution of 20 mM TES (pH = 8.0), 0.2 M L-arginine (pH = 8.0), 2 mM Na₂EDTA, 5 mM reduced glutathione, 0.5 mM oxidized glutathione, and 20 mM ligand. The ligand was described in Garboczi, D. N. , Hung, D. T. c Wiley, D. C (1992) Proc . Natl . Acad. Sci . USA 89, 3429-3433. Yields of the complex were sometimes enhanced with rigorously anaerobic folding conditions.

Gel filtration HPLC with a running buffer of 150 mM NaCl, 50 mM Tris (pH = 8.0) was used to purify folded HLA-B27 complexes from aggregates and excess 3₂-m. Folded complexes had a retention time consistent with their molecular weight of 44 kD. Lower yields were consistently obtained with the aziridine ligand Compound 9; typical yields, confirmed by amino acid analysis, were approximately 60% or less than yields obtained with the control peptide (GRIDKPILK) . The lower yields might result from less efficient hydrogen bonding to the aziridine moiety compared to the planar guanidinium group of the control peptide, or decreased reactivity of one diastereomer of the aziridine ligand Compound 9 (kinetic resolution) . As in previous folding studies of recombinant class I MHC molecules (Garboczi, D.N., Hung, D.T. & Wiley, D.C. (1992) Proc . Natl . Acad. Sci . USA 89, 3429- 3433) , the folding of HLA-B27 was dependent upon a peptide ligand, in this case containing aziridine (Compound 9) . HLA-A2 , an MHC molecule lacking specificity for arginine at P2 , failed to fold in the presence of the aziridine-containing peptide, as expected. Characterization of Covalent Ligand-Protein Bond. The aziridine-containing ligands were shown to form covalent bonds with HLA-B27 by a shift in mobility of the heavy chain on SDS-PAGE; reaction with Cys-67 was shown by tryptic digestion, mass spectrometry (MS) , and Edman microsequencing.

In all preparations of the MHC class I complex based on the aziridine-containing ligand Compound 9, there was an additional SDS-PAGE band corresponding to unmodified heavy chain. This band became more prominent with decreasing concentrations of the aziridine- containing ligand Compound 9 in the folding buffer and low yields of MHC class I molecules, suggesting that the band is due to empty MHC class I molecules.

Fig. 4. depicts the results of an SDS-PAGE (15 %) of HLA-B27 complexed with the aziridine-containing peptide (G(az) IDKPILK (lane 1) and the control peptide (GRIDKPILK) (lane 2) . The SDS-PAGE of HLA-B27 complexed with aziridine-peptides revealed a mobility shift of the heavy chains of the complex due to addition of the aziridine-containing ligand (Fig.4, lane 1), when compared to heavy chain from complexes with the same peptides with arginine in the P2 position (Fig. 4, lane 2) . The mobility of β₂-m remained unchanged.

In order to identify the alkylated amino acid, the aziridine-modified complex was digested with trypsin and the fragments separated by reverse phase HPLC. The carboxyamidomethylated tryptic peptides were separated by narrow-bore HPLC using a Vydac C18 2.1 x 150 mm reverse phase column on a Hewlett-Packard 1090 HPLC/1040 diode array detector. Putative difference peaks from the chromatogram were chosen based on differential UV absorbance at 210, 277, and 292 nm. These were further screened by matrix-assisted laser desorption time-of- flight spectrometry (MALDI-MS) on a Finnigan Lasermat 2000 (Hemel, England) , and submitted to Edman microsequencing on an Applied Biosystems 477A (Foster City, CA) . Details of strategies for the selection of peptide fractions and their microsequencing have previously been described by Lane, W.S., Galat, A., Harding, M.W. & Schreiber, S.L. (1991) J. Prot. Chem. 10, 151-160. Microcapillary HPLC/electrospray ionization/tandem mass spectrometry on a Finnigan TSQ7000 triple quadrupole mass spectrometer (San Jose, CA) as described in Hunt, D. F., Henderson, R.A. , Shabanowitz, J., Sakaguchi, K. , Michel, H. , Sevilir, N. , Cox, A.L., Appella, E. Sc Engelhard, V.H. (1992) Science 255, 1261- 1263.

The chromatogram of the proteolyzed fragments was compared with a tryptic digest of the MHC class I protein formed with the arginine-containing parent-peptide

(GRIDKPILK) . Matrix-assisted laser desorption time-of- flight (MALDI) MS was used to screen rapidly putative difference peaks, which were deficient in UV absorbance at 277 and 292 nm (lacking tyrosine and tryptophan) . A fraction was found with a singly charged ion (M^H+) of the expected mass (m/z = 1759) of the tryptic fragment resulting from alkylation of Cys-67 by the aziridine- containing ligand. Edman microsequencing revealed this fraction to contain the peptide ligand in equimolar amounts with the HLA-B27 heavy chain fragment containing Cys-67. Consistent with the specific alkylation of Cys- 67 by the aziridine, no PTH derivatives were observed at either the sequence position of the aziridine-amino acid in the peptide or Cys-67 in the heavy chain. Microcapillary LC electrospray mass spectrometry (LC-

ESIMS) (11, 12) further confirmed the expected mass of the aziridine-modified fragment (m/z=879.8 M^2H++) .

Stability of Covalent MHC-peptide Complexes. The thermodynamic stability of the class I MHC molecules from mouse and human have been shown to depend on the sequence of the bound peptide. See, Bouvier, M. & Wiley, D.C. (1994) Science 265, 398-402, Fahnestock, M.L., Tamir, I., Narhi, L.O. Sc Bjorkman, P.J. (1992) Science 258, 1658- 1662, and Fahnestock, M.L., Johnson, J.L., Renny Feldman, R.M. , Tsomides, T.J., Mayer, J., Narhi, L.O. & Bjorkman, P.J. (1994) Biochemistry 33, 8149-8158. The thermal transition of the HLA-B27 complexes was measured using circular dichroism (CD) .

An Aviv 62DS (Lakewood, NJ) equipped with thermoelectric temperature controller was used to obtain thermal denaturation curves in triplicate, from 25-95 "C, as described in Bouvier, M. & Wiley, D.C. (1994) Science 265, 398-402. The addition of 50 mM NaCl to the typical 10 mM MOPS (pH = 8.0) buffer was found to prevent aggregation of HLA-B27. The additional salt had no effect on the measured melting temperature (T_m) of the HLA-A2 complex formed with a hepatitis B-derived peptide.

The covalent complexes of HLA-B27 with two aziridine-containing-peptides, G(az) IDKPILK (Compound 9) and G (az) AFVTIGK, both denature at approximately the same temperature, 62 °C G(az) AFVTIGK was prepared by coupling a different peptide to Compound 8 The non-covalent complexes of HLA-B27 with the cognate peptides (GRIDKPILK and GRAFVTIGK) melted at 72°C and 60°C, respectively. The data indicate that covalently bound peptides with an aziridine replacing arginine stabilize HLA molecules approximately like non-covalently bound peptides, in one case stabilizing slightly more and in one less. Although the stability to thermal denaturation is comparable for non-covalent and covalent complexes, the covalently bound peptides should resist dissociating at 37°C for much longer than the non-covalently bound peptides . Covalent MHC-peptide complexes like those reported here may find use generating anergy if administered alone (see, Kozono, H., White, J. , Clements, J. , Marrack, P. Sc Kappler, J. (1994) Nature 369, 151-154) or activation if administered with co-ligands for inducing costimulatory signals as suggested by Schwartz, R.H. (1992) Cell 71, 1065-1068. A non-peptide ligand capable of crossing cellular membranes to block selectively HLA-B27, based on the same aziridine specificity can also be prepared.

OTHER EMBODIMENTS

From the above description, one skilled in the art can easily ascertain the essential characteristics of the present invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions.

For example, the strategy can be used to form covalent complexes of peptides or other ligands with HLA- B27 does not add any new amino acids to the surface of the MHC molecule (that might be immunogenic) , but instead forms the covalent attachment deep in a pocket buried by the bound peptide. In addition, the aziridine strategy can be generalized to other proteins (i.e., recombinant class I MHC molecules) by mutating residues in the binding region (i.e., the P2 pocket residues) to those in HLA-B27. The reverse experiment, engineering an HLA-A2 P2 specificity pocket into HLA-B27 by mutating residues 9, 24, 45, 66, 67, and 70 has been reported by Colbert, R. A., Rowland-Jones, S. L. , McMichael, A. J. Sc Frelinger, J. A. (1993) Proc . Natl . Acad. Sci . USA 90, 6879-6883, resulting in the desired effect of altering the specificity for the P2 position of bound peptide but retaining the same ability to interact with T cells as the unmodified protein. The reverse can be similarly accomplished, HLA-B27 P2 pocket residues can be incorporated into HLA-A2. The success of the Colbert et al . experiment further suggests that any small adjustments in the local shape of the MHC molecule that might accompany covalent bond formation will not affect T-cell recognition, since the covalent bond formation

(Figure 1) is a much smaller modification than six amino acid substitutions.

Also contemplated within the scope of this invention are peptide analogs obtained by back modification of the above-described MHC-blocking peptides, e.g., replacement of at least one of the peptide bonds with -CH₂-NHX, CH₂-S-, or the like. Similarly, alkyl groups can be readily substituted by alkenyl, alkynyl or alkyl ether groups. Thus, other embodiments are also within the claims.

What is claimed is:

Claims

CLAIMS 1. A compound of the formula

wherein each of R_x and R₂, independently, is hydroxy, halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, alkyl carboxylate, a Michael acceptor, amino, alkylamino, ammonium, or alkyl ammonium, or together R_x and R₂ are O, NH, NH₂X or NZ, where Z is alkyl or aryl ; X is C_x to C₆ alkyl, substituted aryl, or deleted; R₃ is H, alkyl, or aryl;

R₄ is alkyl, aryl, -NH-W, or -C(=0)-Y, where W is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group; and

R₅ is H, alkyl, aryl, -NH-W', or -C(=0)-Y', where W' is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y' is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group, or an amino acid residue bonded at its amino group.

2. The compound of claim 1, wherein:

R-_L is halogen, alkyl sulfonate, carboxylate, or aryl sulfonate and R₂ is hydroxy, amino, alkylamino, ammonium, or alkyl ammonium, or together R_j and R₂ are 0, NH₂\ or NH; and

X is a C₂ to C₄ alkyl.

3. The compound of claim 2, wherein: R₃ is H;

R₄ is -NH-W, where W is an amino acid residue bonded at its carbonyl group or a peptide moiety bonded at its terminal carbonyl group; and

R₅ is a -C(=0)-Y', where Y' is a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group.

4. The compound of claim 3, wherein:

W is an amino acid residue bonded at its carbonyl group ; and

Y' is a peptide moiety bonded at its terminal amino group .

5. The compound of claim 4, wherein:

R_x is alkyl sulfonate and R₂ is an ammonium, or together R and R₂ are 0, or NH; X is C₄ alkyl; and the amino acid residue is a glycine residue.

6. The compound of claim 4, wherein a portion of the peptide moiety is identical to a portion of a selected peptide.

7. The compound of claim 4, wherein a portion of the peptide moiety is identical to a portion of an antigenic peptide.

8. The compound of claim 7, wherein the peptide moiety contains between 2 and 10 amino acid residues.

9. A method of forming a ligand-protein complex comprising the steps of: providing a protein having an arginine-specific binding region, the binding region including a nucleophilic residue; and adding a ligand to bind with the protein at the arginine-specific binding region, the ligand having a moiety which reacts with the nucleophilic residue, whereby the moiety of the ligand and the nucleophilic residue of the protein form a covalent bond between the ligand and the protein.

10. The method of claim 9, wherein: the arginine-specific binding region includes a thiol , a thiol and a carboxyl , or two carboxyl groups ; and the moiety is halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, carboxylate, a Michael acceptor, aziridine, or epoxide.

11. The method of claim 10, wherein the arginine- specific binding region includes a thiol and a carboxyl.

12. The method of claim 11, wherein the moiety is aziridine, or epoxide.

13. The method of claim 9, wherein the step of providing the protein includes the step of altering a binding region of the protein to introduce the arginine- specific binding region.

14. The method of claim 13, wherein: the arginine-specific binding region includes a thiol , a thiol and a carboxyl , or two carboxyl groups ; and the moiety is halogen, haloalkyl, haloalkyl carbonyl, alkyl sulfonate, aryl sulfonate, alkyl carboxylate, a Michael acceptor, aziridine, or epoxide.

15. The method of claim 14, wherein the arginine- specific binding region includes a thiol and a carboxyl.

16. The method of claim 15, wherein the moiety is aziridine, or epoxide.

17. A covalently linked ligand-protein complex comprising a protein having an arginine-specific binding region and a ligand covalently bonded to the binding region of the protein, said ligand having the formula

wherein one of R_x and R₂ is hydroxy, alkyl, alkyl carbonyl, amino, alkyl amino, ammonium, or alkyl ammonium, and the other of R_x and R₂ is the covalent bond to the protein;

X is C_± to C₆ alkyl, substituted aryl, or deleted;

R₃ is H, alkyl, or aryl; R₄ is alkyl, aryl, -NH-W, or -C(=0)-Y, where W is H, alkyl, aryl, an amino acid residue bonded at its carbonyl group, or a peptide moiety bonded at its terminal carbonyl group, and Y is alkyl, alkoxy, amino, substituted amino, hydroxyl, a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group; and

18. The complex of claim 17, wherein: the arginine-specific binding region includes a thiol, a thiol and carboxyl, or two carboxyl groups;

R_λ is the covalent bond between the ligand and the protein; R₂ is hydroxy, alkyl, alkyl carbonyl, amino, alkyl amino, ammonium, or alkyl ammonium; R₃ is H; and X is a C₂ to C₄ alkyl.

19. The complex of claim 17, wherein: the arginine-specific binding region includes a thiol, a thiol and carboxyl, or two carboxyl groups;

R_x is hydroxy, alkyl, alkyl carbonyl, amino, alkyl amino, ammonium, or alkyl ammonium;

R₂ is the covalent bond between the ligand and the protein;

R₃ is H; and

X is a C₂ to C₄ alkyl .

20. The complex of claim 19, wherein:

R₄ is -NH-W, where W is an amino acid residue bonded at its carbonyl group or a peptide moiety bonded at its terminal carbonyl group; and R₅ is a -C(=0)-Y', where Y' is a peptide moiety bonded at its terminal amino group or an amino acid residue bonded at its amino group.

21. The complex of claim 20, wherein:

W is an amino acid residue bonded at its carbonyl grou ; and

Y' is a peptide moiety bonded at its terminal amino group.

22. The complex of claim 21, wherein: R_x is hydroxy, amino, or ammonium; X is C₄ alkyl; and the amino acid residue is a glycine residue.

23. The complex of claim 22, wherein the arginine-specific binding region includes a thiol and carboxyl .

24. The complex of claim 21, wherein a portion of the peptide moiety is identical to a portion of a selected peptide.

25. The complex of claim 21, wherein a portion of the peptide moiety is identical to a portion of an antigenic peptide.

26. The complex of claim 25, wherein the peptide moiety contains between 2 and 10 amino acid residues.