WO1992009625A1

WO1992009625A1 - Conformationally constrained peptides i

Info

Publication number: WO1992009625A1
Application number: PCT/US1991/008873
Authority: WO
Inventors: Pradip Kumar Bhatnagar; Renee Louise Des Jarlais; James Scott Dixon; Wayne Arthur Hendrickson; Kenneth D. Kopple; Peter Kwong; Catherine Elizabeth Peishoff; Seong-Eon Ryu; Alemseged Truneh; Raymond W. Sweet
Original assignee: Smithkline Beecham Corporation; The Trustees Of Columbia University In The City Of New York
Priority date: 1990-11-29
Filing date: 1991-11-27
Publication date: 1992-06-11
Also published as: AU9118891A

Abstract

The present invention discloses compounds that inhibit HIV infection by mimicking stereochemical regions of the CD4 receptor protein.

Description

Tit le

CONFORMATIONALLY CONSTRAINED PEPTIDES I

Field of Invention

The present invention generally relates to inhibition of HIV infection by mimicking structures of the CD4 receptor protein.

Background of the Invention The hallmark of the Acquired Immunodeficiency

Disease, AIDS, is a progressive decline in the number of CD4+ cells leading to the demise of immune function and consequent susceptibility to opportunistic infections, the primary cause of death. Human immunodeficiency virus type 1 (HIV-1) is the primary causative agent of AIDS. This highly variable virus shows selective tropism for CD4+ cells which is determined by

recognition of the HIV envelope glycoprotein gp120 by the CD4 cell surface receptor protein. The manner in which HIV infection leads to the slow but progressive decline in CD4 cells has not been established. However, it has been shown that an agent which antagonizes HIV in vitro, AZT, also provides a therapeutic benefit in vivo. Thus antagonism of HIV is the primary therapeutic strategy for AIDS.

The process of viral infection is initiated by the attachment of HIV to cells through a high affinity interaction between gp120 and the CD4 receptor protein, located on the cell surface. This site of binding to CD4 has been localized to the first extracellular domain DI (previously denoted as VI). A region on gp120 which interacts with CD4 has also been identified but the overall structure of the protein is poorly

characterized.

A second process, virus-mediated cell fusion, is also initiated by the interaction of gρ120 with the CD4 receptor protein. Cells infected with the HIV virus can express envelope proteins, ultimately detected on the infected-cell's surface. Thus gp120 on the surface of infected cells can bind to CD4 or uninfected cells leading to the fusion and consequent formation of multinuclear giant cells (i.e., syncytium formation). This process is envisioned as a cell-cell equivalent of the binding and fusion events between HIV and an

uninfected cell. However, the pathology of this virusmediated cell fusion is not well understood.

Specific residues in regions of CD4 are implicated in interacting with the HIV envelope

glycoproteins based upon mutational analysis, antibody inhibition and epitope studies. However, the design of compounds that will selectively block HIV infection poses considerable difficulties without extensive knowledge of the molecular structures or mechanisms involved. One molecular structure has been determined, and it is thus an object of this invention to rationally design selective HIV inhibitors.

Summary of the Invention

The present invention relates to a method for inhibiting HIV infection by administering to a patient a pharmaceutically effective amount of a conformationally constrained compound of the formula: X-A-B-C-D-Y

wherein

A-B-C-D is a ß turn tetrapeptide or a ß-turn tetrapeptide mimic;

X and Y are groups which provide a means to restrict the stereochemical structure of A-B-C-D such that A-B-C-D forms or mimics a ß-turn; and

A-B-C-D binds to at least one HIV envelope protein; thereby inhibiting HIV infection.

In related aspects, when X and Y of the present invention both comprise a naturally occurring amino acid sequence that is present in the D1 domain of the CD4 receptor protein, then the combination of X and Y taken together number less than 31 amino acids.

This invention also relates to a

conformationally restricted compound of the formula:

X-A-B-C-D-Y

wherein

A is selected from the group consisting of Gly, Ala, Ser, Thr, Asp, Asn, Glu, Gln, His, Lys and Arg;

B is any amino acid;

C is any amino acid;

D is Leu, Phe or Phe-minic;

the combination X + Y provides a means to restrict the stereochemical structure of A-B-C-D such that

A-B-C-D forms a beta turn;

with the proviso that when X and Y both comprise a naturally occurring amino acid sequence present in the D1 domain of CD4, then X and Y together number less than 31 amino acids;

or a pharmaceutically acceptable salt thereof.

X and Y can be chemical structures that form covalent bonds or X and Y can be amino acids or

derivatives thereof. When X and Y are amino acids, they may be joined by a disulfide bridge (s) or joined by a peptide bond(s) to restrict the stereochemical structure of A-B-C-D.

In related aspects, the present invention is a recombinant protein in which the CDR-2 or CDR-3 and ß or both regions of an immunoglobulin are replaced with the compounds of the present invention.

In further related aspects, this invention relates to a composition for inhibiting HIV infection which comprises an effective amount of the compound of the present invention and a pharmaceutically acceptable carrier.

Detailed Description of the Invention

The present invention relates to compounds which inhibit the interaction between HIV envelope glycoproteins and the human cell-surface protein, CD4. In particular, they exemplify a class of molecules which inhibit binding andfior subsequent fusion by mimicking structures of the CD4 protein which interacts with HIV envelope glycoproteins (e.g., gp120, gp41, gp160).

These "structures" are compounds which have a restricted stereochemical conformation.

The high affinity binding site for the HIV envelope protein has been localized to the amino

terminal domain of CD4, referred to as D1 (previously denoted as VI). This domain was identified as the region critical for binding through expression of a truncated protein containing residues 1-106 (see, Arthos et al., Cell, 57:466-81 (1989)). This protein was shown to have an affinity for gp120 comparable to that of soluble CD4 (i.e., native receptor).

It is believed that at least two distinct regions on D1 are involved in HIV infection of CD4⁺ cells. Based on mutational data and epitope mapping the region of amino acid residues 41-59 has been implicated as the high affinity binding site for gpl20 (see for example, Arthos et al., Cell, 57:469-81 (1989)).

However, it was not clear which residues in this region are involved in binding since at many positions of a "down-effect" mutation (i.e., decrease in affinity for gpl20), was countered by substitutions having little effect. Blocking of this site with specific mAbs

(monoclonal antibodies) causes inhibition of gp120 binding as well as HIV infection and HIV-induced

syncytium formation.

A second region, which lies outside the high affinity binding site, is thought to be required for events that occur subsequent to the initial binding step. Inhibition of this site (approximately regions 77-85) causes a limited interference of gp120 binding to CD4. Monoclonal antibodies (mAbs) that bind to this site block HIV infection of CD4 cells and the fusion between HIV infected and uninfected cells. It has also been shown that heavily benzylated peptides

corresponding to this region also inhibit HIV

infectivity and syncytium formation (see, for example, Lifson et al., Science, 241:712-16 (1988)).

Furthermore, a single amino acid substitution resulted in loss of syncytium formation without affecting gp120 binding to CD4 (Cammerini et al., Cell, 60:747-754

(1990)). The exact role of this region in the HIV-CD4 interaction is not clear. However, it does appear that this region interacts with HIV differently than the high affinity binding site.

The present invention, then, is capable of inhibiting HIV infection through: antagonism of the viral envelope glycoprotein, gpl20, from binding to the CD4 receptor; or antagonism of the HIV virus - cell mediated fusion process; or antagonism of cell to cell fusion initiated by the interaction of viral envelope protein on infected cells with the CD4 receptor on uninfected cells; or a combination of any of the above.

A more detailed structure of the amino-terminal domains of CD4 has since been determined. A high resolution crystal structure of D1D2 (previously denoted as V1V2) indicates that it is overall of the ß-type.

That is, the backbone of the polypeptide chain is in a ß conformation, where the polypeptide ß strands are arranged side by side in an anti-parallel direction and form interstrand hydrogen bonds. D1D2 comprises two associated domains. D1 (previously denoted as V1), which approximately comprises residues 1 to 98, is arranged into nine ß strands. The second domain D2 (residues 99 to 173), comprises seven ß strands.

The beta strands of D1 are connected by loops at the end of the beta strands (i.e., beta turn loops). That is, a beta or beta turn loop consists of two antiparallel ß-strands and a beta turn. Some of these "loops" comprise hairpin structures having a beta turn of approximately 4 amino acids denoted as i, i+1, i+2 and i+3. Based upon structural similarity, three of the loops are analogous to the complementarity determining regions (CDR) of the immunoblobulin variable-light domains (Ig-V_L). One beta turn loop (defined by ß strands C¹ and C¹¹ or approximately residues 37-46), corresponds to CDR2 of an Ig variable region. Another beta turn loop (defined by ß strands F and G or

approximately residues 79-96), corresponds to CDR3 of an

Ig variable region.

In D1, a hairpin loop comprising a ß-turn at residues Gin⁴⁰ to Phe⁴³ protrudes from the D1D2

molecule. Furthermore, the hydrophobic side chain of Phe⁴³ appears to project away from D1D2. A similar structure is found for residues Val⁸⁶ to Gin⁸⁹, that is, a hairpin loop that protrudes from D1D2.

One embodiment of the present invention are peptides or peptides mimics that are comformationally constrained to stabilize a two-stranded antiparallel beta sheet with a beta turn. The structure of the beta turn is herein referred to as A-B-C-D which corresponds to the i, i+1, i+2 and i+3 positions as previously denoted. Therefore, A-B-C-D is a ß turn or ß turn mimic in the present invention. The structure of the beta turn of a constrained peptide may be verified via structural analysis, e.g. NMR or X-ray crystallography. Such peptides or peptide mimics can have a range in length from 4 to 100, preferably 4 to 75, and more preferably 6 to 30.

When A, B, C and D are all amino acids, the ß turn is herein referred to as a ß turn tetrapeptide. When the peptide backbone of at least one amino acid is replaced by. a non-peptide frame capable of carrying the amino acid side-chain group (s) in a conformation that mimics that of a ß turn, the structure is referred to as a ß turn tetrapeptide mimic. Such ß turn mimics have at least one non-peptidyl bond, e.g., an acyl hydrazide linkage. Another example of ß turn mimics is reported by Olson et al. (J Am Chem Soc, 112:323-333 (1990)) and is incorporated by reference herein.

As used herein, the term "conformationally constrained" or "conformationally restricted" refers to limitations imposed on the number of possible

stereochemical structures that a peptide or peptide mimic may assume. In the present invention, a beta turn or beta turn mimics are formed. Such restrictions are imposed on the conformation (or stereochemical

structure) by the presence of chemical structures or groups, herein referred to as "X" and "Y", (e.g., amino acid or other chemical moieties) surrounding the beta turn. Where specific chemical structures or sequences are presented, it is intended to encompass various modifications (e.g., acetylation, alkylation, etc.) which do not destroy the inhibitory functions of the present invention. For example, conformational stabilization can be achieved by covalently linking residues on either side of the desired beta turn at specific positions with groups that stabilize both the antiparallel beta structure and the desired position in the sequence of the turn.

It is preferable, but not absolute, that the compounds of the present invention are peptides. Such restrictions can be accomplished by covalently linking residues at specific positions to conformationally constrain or restrict the compounds of the presen t invention. For example, this includes, but is not limited to, formation of disulfide bonds, homodetic peptide cyclization, and attachment to larger protein molecules.

One preferred embodiment comprises residues 40 (residue "i" of the beta turn or "A") through 43

(residue "i+3" or "D") of the D1 domain of CD4.

Preferably, residue 43 (i.e., "D" or the i+3 position) is leucine or phenylalanine or a phenylalanine mimic, e.g., tyrosine, tryptophan, 3-thienylalanine,

naphthylalanine or phenylalanine wherein the phenyl group is substituted by halogen, C_1-4 alkyl or C_1-4 alkoxy.

"A" is preferably Gly, Ala, Ser, Thr, Asp, Asn,

Glu, Gln, His, Lys or Arg. Another preferred embodiment are conformationally constrained peptides which comprise residues 86 (i.e., i or "A") through 89 (i+3 or "D". In this embodiment "B" is preferably Gly, Ala, Ser, Thr, Asp, Asn, Glu, Gin, His, Lys or Arg).

For purposes of illustration, the beta hairpin may be shown schematically as follows:

N-term.

C-term.

N-term.

C-term.

Wherein the beta turn comprises residues 40 through 43, or residues 86-89.

The methods for stabilizing the antiparallel beta hairpin conformation may include, but are not limited to, formation of disulfide bonds, homodetic peptide cyclization, a combination of disulfide bonds and peptide cyclization, or attachment to larger protein molecules.

For example, the incorporation of cysteine residue pairs are positioned such that, when they are linked (by oxidation of their side chains) to form a disulfide bond, an antiparallel structure is stabilized. Preferred embodiments include Cys pairs at positions 39,44; at 38,45; at 37,46; at 36,47; at 35,48; at 85,90; at 84,91; at 83,92; at 82,93; at 81,94; at 80,95; and at 79,96. Exemplars of conformational CD4 mimics are depicted below: N-term.

C-term.

Wherein * indicates a disulfide bond between the sulfhydryl groups of cysteine.

N-term.

C-term. N-term.

C-term.

N-term.

C-term.

N-term.

C-term.

N-term.

C-term.

N-term.

C-term.

N-term.

C-term. Another example of conformational restriction is by homodetic peptide cyclization using beta turn favoring amino acid pairs to form a cyclic peptide. Preferably, the cyclic peptide comprises two beta turns connecting extended antiparallel beta segments. Such a backbone cyclization should preferably form rings comprising 4n+2 residues, e.g., 6, 10, 14, etc.

residues. Exemplars are depicted below:

N-term.

C-term. N-term.

C-term.

N-term.

⁴

C-term.

N-term.

C-term.

N-term .

C-term . N-term.

C-term.

The sequence Yaa-Xaa stabilizes this turn and may include such pairs as Pro-D-Pro, D-Pro-Pro (where Pro may be replaced by a proline analog such as beta, beta dimethyl gamma-thiaproline), or pairs with only one proline, e.g., D-Zaa-L-Pro or Zaa-D-Pro, where Zaa is any other amino acid (or modified amino acid) capable of forming a peptide bond to Pro.

As used herein, the term "peptide bond" refers to an amide linkage between a carboxy1 group of one amino acid and the a-amino group of another amino acid.

In another embodiment, the sequence Gly ⁴¹-Ser⁴² or Glu⁸⁷-Asp⁸⁸ may similarly be replaced at the same time to further stabilize the desired structure. For example:

N-term.

C-term.

N-term.

C-term.

where Xaa' and Yaa' are the same or different turn stabilizing pairs as Xaa and Yaa.

Peptides may also be conformationally restricted by a combination of homodetic peptide

cyclization and disulfide bonds. For example, the

Gly⁴¹-Ser⁴² or Glu⁸⁷-Asp⁸⁸ sequence is replaced by a turn stabilizing sequence (see above) and a Cys-Cys disulfide bond is incorporated to comprise a homodetic cyclic peptide which has a stabilizing disulfide link.

For example:

N-term.

C-term .

N-term .

C-term.

Conformationally restrained peptides can also be stabilized by attachment to larger protein molecules, such as an immunoglobulin. For example, all or a portion of the CDR-2 region from the CD4 receptor protein (approximately residues 37 to 46) can be

substituted for the structurally analogous loop of an immunoglobulin. Preferably the immunoglobulin is a human IgG subclass. More preferably it comprises a variable light chain, e.g., the variable light chain (V,) of the Bence-Jones REI protein (residues 48-54)

[ See Sattentau et al . , J Exp Med , 17 0 : 1319-34 ( 1989 ) or Epp et al., Biochem, 14:4943 (1975)], resulting in the sequence below wherein residues from CD4 are underlined: (N-term) Asp lie Gin Met Thr Gin Ser Pro Ser Ser Leu Ser Ala Ser Val Gly Asp Arg Val Thr lie Thr Cys Gin Ala Ser Gin Asp lie lie Lys Tyr Leu Asn Trp Tyr Gin Gin Thr Pro Gly Lys Ala Pro Lys Leu Leu Leu Gly Asn Gln Gly Ser Phe Leu Thr Lys Gln Ala Gly Val Pro Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Tyr Thr Phe Thr Ile Ser Ser Leu Gin Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln Gln Tyr Gln Ser Leu Pro Tyr Thr Phe Gly Gln Gly Thr Lys Leu Gln Ile Thr (C-term)

In another embodiment, the CDR-3 region of the CD4 receptor protein (residues 81 to 93) can be

substituted for the structurally analogous loop in the Bence-Jones REI protein (residues 85-102), resulting in the sequence below wherein residues from CD4 are underlined: (N-Term) Asp Ile Gln Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly Asp Arg Val Thr Ile Thr Cys Gln Ala Ser Gln Asp Ile Ile Lys Tyr Leu Asn Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro Lys Leu Leu Ile Tyr Glu Ala Ser Asn Leu Gln Ala Gly Val Pro Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Tyr Thr Phe Thr Ile Thr Ile Ser Ser Leu Gin Pro Glu Asp Ile Ala Thr Tyr Ile Cys Glu Val Glu Asp Gin Lys Glu Glu Val Lys Leu Gln Ile Thr (C-term)

In still another embodiment, the compounds of the present invention can be constructed by substituting the CDR-2 and CDR-3 residues of CD4 for the structurally analogous loops in the Bence-Jones REI protein (residues 48-54 and 85-102, respectively), resulting in the sequence below wherein the residues from CD4 are

underlined:

(N-term) Asp He Gln Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly Asp Arg Val Thu Ile Thr Cys Gln Ala Ser Gln Asp Ile Ile Lys Tyr Leu Asn Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro Lys Leu Leu Leu Gly Asn Gln Gly Ser Phe Leu Thr Lys Gin Ala Gly Val Pro Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr asp Tyr Thr Phe Thr Ile Ser Ser Leu Gln Pro Glu Asp He Ala Thr Tyr Ile Cys Glu Val Glu Asp Gln Lys Glu Glu Val Lys Leu Gln Ile Thr (C-term)

Peptides of the present invention may be synthesized by any suitable method. Methods of wellknown peptide synthesis are generally set forth by Ali et al., J Med Chem, 29:984 (1986) and J Med Chem,

30: 2291 (1987) and are incorporated by reference herein. Preferably, the peptides are prepared by the solid phase technique of Merrifield (J Am Chem Soc, 85:2149 (1964)). However, a combination of solid phase and solution synthesis may be used, as in a convergent synthesis in which di-, tri-, tetra-, or penta-peptide fragments may be prepared by solid phase synthesis and either coupled or further modified by solution synthesis.

During synthesis, the side chain functional groups (e.g., -NH₂, -COOH, -OH, -SH) are protected during the coupling reactions. Normally, the a-amino group is temporarily protected as t-Butoxycarbonyl (BOC) but other acid or base labile protecting groups can be used, e.g., fluorenylmethoxycarbonyl (FMOC). The amino side chain group of lysine is protected as

benzyloxycarbonyl or p-chlorobenzyloxycarbonyl (Z or ClZ). Para-methylbenzyl (p-MBz) or acetomidomethyl protection is used for cysteines. Hydroxy groups are protected as benzyl ethers and carboxyl groups are protected as benzyl (Bz) or cyclohexyl esters.

The peptides of the present invention can be synthesized either from the C-terminus or the N-terminus. Preferably it is from the C-terminus. Prior to coupling the alpha-carboxyl group (of a suitable protected amino acid) is activated. One skilled in the art can activate the protected group in a number of ways. For example, one may use N,N'

dicyclohexylcarbodiimide (DCC), p-nitrophenyl esters (pNp), hydroxybenzotriazole ester (HOBt), N-hydroxy succinimidyl ester (Osu) mixed anhydride or symmetrical anhydride.

Solution synthesis of peptides is accomplished using conventional methods to form amide bonds.

Typically, a protected Boc-amino acid which has a free carboxyl group is coupled to a protected amino acid which has a free amino group using a suitable

carbodiimide coupling agent, such as N, N¹ dicyclohexyl carbodiimide (DCC), optionally in the presence of a catylyst such as 1-hydroxybenzotriazole (HOBT) and dimethylamino pyridine (DMAP).

In solution phase synthesis, the coupling reactions are preferablye carried out at low temperature (e.g., -20°C) in such solvents as dichloromethane (DCM), dimethyl formamide (DMF), N-methyl pyrrolidone (NMP), tetrahydrofuran (THF) acetonitrile (ACN) or dioxane.

If solid phase methods are used, the peptide is built up sequentially starting from the carboxy terminus and working towards the amino terminus of the peptide. Solid phase synthesis begins by covalently attaching the C terminus of a protected amino acid to a suitable resin, such as methyl benzhydrylamine (mBHA).

In the solid phase synthesis, the first amino acid residue is normally attached to an insoluble polymer. For example, two commonly used polymers are polystyrene (1% cross-linked with divinyl benzene) and 1% cross-linked polyamides. These polymers are

functionalized to contain a reactive group, e.g., -OH, -NH-, and -CH₂Cl to link the first amino acid of the targeted peptide (i.e., carboxy terminus). The choice of the linkage between the first amino acid and the polymer is dictated by the carboxy terminus of the peptide. For example, peptides having a carboxyl group at the C-terminus would be linked by an ester linkage and for peptides with a carboxamide ending would have an amide linkage.

Once the first protected amino acid has been coupled to the desired resin, the protected a amino group is hydrolyzed by mild acid treatment, and the free carboxyl of the next (protected) amino acid is coupled to this amino group. This process is carried out sequentially, without isolation of the intermediate, until the peptide of interest has been formed. The completed peptide may then be deblocked and ß or cleaved from the resin in any order.

Preferred solvents for the coupling reactions include, but are not limited to, dichloromethane (DCM), dimethyl formamide (DMF) and N-methyl pyrrolidone (NMP). After the desired sequence is synthesized, the peptide is deprotected and cleaved from the resin using standard techniques (e.g., using hydrofluoric acid (HF)).

The preferred method for cleaving a peptide from the support resin is to treat the resin supported peptide with anhydrous HF in the presence of a suitable cation scavenger, such as anisole or dimethoxy benzene.

To obtain the conformationally restricted peptides of the present invention, the synthetic peptide may be cyclized using methods well known in the art.

For example, there are numerous protocols for forming intramolecular bonds between cysteine residues. In performing these reactions, water, methanol, acetic acid, DMF or a suitable mixture of these solvents can be used.

Formation of the disulfide bond may be accomplished by several known methods. If the sulfurcontaining amino acids of the linear peptide are

protected differently, in such a manner as to allow formation of a mono mercaptan, cyclization may be effected by base catalyzed nucleophilic displacement of the protecting group of the sulfur-containing amino acid. Groups which are especially useful as

displaceable protecting groups are thioalkyl or thioaryl groups. Examplary of this method is the protection of one sulfur-containing amino acid by the thioethyl group, and protection of the second by a substituted benzyl group. Deprotection of such a peptide by HF removes the benzyl group from one amino acid, while leaving the second protected as an ethyl disulfide. Stirring this mercaptofidisulfide in dilute solution at a pH of about 7 to 8 effects displacement of the thioethyl group and cyclization of the linear peptide. When cysteine is protected with an ACM (acetamido methyl) group the preferred method is to oxidize with iodine.

If the corresponding linear peptide is completely deprotected and produced as a dimercaptan, any oxidizing agent known in the art to be capable of converting a dimarcaptan to a disulfide may be used. Examplary of such agents are an alkali metal

ferricyanide, (e.g., potassium or sodium ferricyanide), oxygen gas, diiodomethan or iodine. The reaction is conducted in a suitable inert solvent, such as aqueous methanol or water, at temperatures from about 0 to 40ºC, under high dilution. The pH is uaually maintained at about 7 to 8. Cyclization may be performed upon the peptide while it is still attached to the support resin or while other functional groups are still protected, but it is preferably performed on the deprotected free peptide.

The conformationally restricted peptides can also be prepared by forming a peptide bond between non-adjacent amino acid residues. One procedure for forming peptide bonds is provided by Schiller et al., Int J

Peptide Protein Res, 25:171 (1985).

Acid addition salts of the peptides are prepared in a standard manner in a suitable solvent from the parent compound and an excess of an acid, such as hydrochloric, hybrobromic, sulfuric, phosphoric, acetic, maleic, succinic, or methanesulfonic. The acetate salt form is especially useful. Certain of the compounds form inner salts or zwitterions which may be acceptable. Cationic salts are prepared by treating the parent compound with an excess of an alkaline reagent, such as a hydroxide, carbonate or alkoxide containing the appropriate cation. Cations such as Na⁺, K⁺, Ca²⁺ and

NH₄ are examples of cations present in pharmaceutically acceptable salts.

The peptides can then be purified by a number of techniques. Preferred embodiments include reverse phase HPLC, counter current distribution (CCD) and crystalization. More preferably, HPLC is used. The purified products can then be analyzed for purity using HPLC, amino acid analysis and fast atom bombardment mass spectrometry (FAB-MS).

The compounds of the present invention can be assayed for their ability to inhibit HIV infection via binding assays and functional (fusion or infectivity) assays. One binding assay entails a competition ELISA, measuring the binding inhibition of sT4 to immobilized recombinant gp120 in the presence and absence of the compounds of the invention. A similar competition RIA entails measuring the binding inhibition of labelled sT4 to (immobilized) recombinant gp120. Another method entails the inhibition of HIV gp120 binding to CD4+ cells. Bound gp120 is detected by gamma counting when using ¹²⁵I-labelled gp120 or by glow cytometry when using FITC (fluorescein isothiocyanate)-labelled gp120. Alternatively, anti-gp120 antibodies may also be used to detect binding. These assays are performed essentially as described in Arthos et al. (Call, 57:469-81 (1989)) and Sattentau et al. (J Exp Med, 170: 1319-34 (1989)), and are incorporated by reference herein. It is noted that the compounds of the present invention bind to HIV envelope proteins with the same or greater affinity than the native CD4 or soluble CD4 proteins.

One functional assay (i.e., syncytium assay) comprises the inhibition of cell fusion between

chronically infected cells and uninfected CD4 cells. In essence, HIV infected H9 cells (R. Gallo, National Institute of Health, Bethesda, MD, USA) are cocultivated overnight with uninfected cells at a ratio of 1:2, in the presence of an inhibitory substanceßcompound. Such assays are performed essentially as described by

Sleckman et al., (Nature, 328:351-3 (1987)) and are hereby incorporated by reference.

A non-viral syncytium assay can also be used. This assay measures the inhibition of fusion between cells expressing HIV env protein and CD4⁺ cells as disclosed in U.S application serial number, 07ß587,011, filed Septenber 24, 1990 (Clark et al., "Human Lymphoid Cells Expressing HIV Envelope Protein gp160") and incorporated by reference herein.

Another functional assay is the virus infectivity assay. This assay comprises infection of T- lymphocytes or macrocytefimacrophages with HIV. At six or more days post-infection measurement of particle- associated reverse transcriptase activity and ß or p24 antigen levels can be determined (See, for example, Clapham et al., Nature, 337: 368-370 (1990) or McDougal et al., J Immun Meth, 76: 171-183 (1985)).

Pharmaceutical compositions of the peptides prepared as hereinbefore described and other peptide or polypeptide derivatives may be formulated as solutions of lyophilized powders for parenteral administration. Powders may be reconstituted by addition of a suitable diluent or other pharmaceutically acceptable carrier prior to use. The liquid formulation is generally a buffered, isotonic, aqueous solution. Examples of suitable diluents are normal isotonic saline solution, standard 5% dextrose in water or buffered sodium or ammonium acetate solution. Such formulation is

especially suitable for parenteral administration, but may also be used for oral administration or contained in a metered dose inhaler or nebulizer. It may be

desirable to add excipients such as

polyvinylpyrrolidone, gelatin, hydroxy cellulose, acacia, polyethylene glycol, mannitol, sodium chloride or sodium citrate.

Alternatively, these peptides may be encapsulated, tableted or prepared in a emulsion or syrup for oral administration. Pharmaceutically

acceptable solid or liquid carriers may be added to enhance or stabilize the composition, or to facilitate preparation of the composition. Solid carriers include starch, lactose, calcium sulfate dihydrate, terra alba, magnesium stearate or stearic acid, talc, pectin, acacia, agar or gelatin. Liquid carriers include syrup, peanut oil, olive oil, saline and water. The carrier may also include a sustained release material such as glyceryl monostearate or glyceryl distearate, alone or with a wax. The amount of solid carrier varies but, preferably, will be between about 20mg to about lg per dosage unit. The pharmaceutical preparations are made following the conventional techniques of pharmacy involving milling, mixing, granulating, and compressing, when necessary, for tablet forms; or milling, mixing and filling for hard gelatin capsule forms. When a liquid carrier is used, the preparation will be in the form of a syrup, elixir, emulsion or an aqueous or non-aqueous suspension. Such a liquid formulation may be

administered directly or filled into a soft gelatin capsule.

For rectal administration, the peptides of this invention may also be combined with excipients such as cocoa butter, glycerin, gelatin or polyethylene glycols and molded into a suppository.

The dosage ranges for administration of the compounds of the present invention are those to produce the desired effect whereby symptoms of HIV or HIV infection are ameliorated. As used herein, a

pharmaceutically effective amount refers to the amount administered so as to maintain an amount which

suppresses or inhibits secondary infection by syncytia formation or by circulating virus throughout the period during which HIV infection is evidenced such as by presence of anti-HIV antibodies, presence of culturable virus and presence of p24 antigen in patient sera. The presence of anti-HIV antibodies can be determined through use of standard ELISA or western assays for example, anti-gp120, anti-gp41, anti-tat, anti-p55, anti-p17 antibodies, etc. The dosage will generally vary with age, extent of the infection, and

counterindications, if any, for example, immune

tolerance. The dosage can vary from 0.001 mgßkgßday to 50 mgßkgßday, but preferably 0.01 to 1.0 mgßkgßday.

The Examples which follow are illustrative, and not to be construed as limiting of the present

invention.

EXAMPLES

In the examples which follow all temperatures are in degrees centigrade. Amino acid analysis was performed upon a Dionex autoion 100. Mass spectra were performed upon a VG Zab mass spectrometer using fast atom bombardment. HPLC was performed upon a Beckman 344 gradient chromotography system with a CRIB recording integrator in either an isocratic or continuous gradient mode. Where indicated, the purity of the peptide is based upon integration of the HPLC chromatogram.

In accordance with conventional representation, the amino terminus is on the left and the carboxy terminus is on the right. Unless specified otherwise, all amino acids are assumed to be of the L-configuration.

Example 1. Preparation of:

A manual shaker vessel was charged with 1.45g of methylBHA resin (substitution 0.7 mMßg). The resin was washed with 40% trifluoroacetic acid (TFA) in dichloromethane (DCM). The resulting trifluoroacetate salts were neutralized with 10% diisopropylethyl amine (DIEA) in DCM. After neutralization, BOC-Cys (P-MBz)-COOH was coupled to the resin using

dicyclohexylcarbodiimide (DCC) and hydroxybenzotriazole (HOBt) in 50% DMFßDCM (10 ml). Three equivalents of amino acid, DCC and HOBt were used. The. coupling, monitored using Kaiser's test, is normally complete in two hours. After coupling, the resin was extensively washed with DCM and NMP, the cycle of deprotection, neutralizaiton and coupling step was repeated with the remaining amino acids in the target peptide sequence. Each coupling was monitored for completion using the Kaiser test. After the coupling of the last amino acid, the resin was subjected to a deprotection and

neutralization cycle and the resulting free amino group was acetylated using 10% acetic anhydride in pyridine. The peptide was cleaved from the resin using 10 ml HF and 1 ml anisole at -15°C for two hours. After removal the resin-peptide mixture was extensively washed with ether. The peptide was extracted in glacial acetic acid. The extract was diluted with water so that, the final concentration of the peptide was about 500 mgßl. The pH was adjusted to 7.5 and the solution was stirred at room temperature. Formation of the disulfide bond was monitored using Ellman's reagent. Upon completion of oxidation, the solvent was removed on a rotovap. The crude peptide was purified on a preparative C-18 (2×25 cm) HPLC Vydac column using 0.1% TFAßH₂O and a 0.1% TFAßCH₃CN gradient. It was analyzed for amino acid composition and molecular weight.

HPLC purity: >95%

Amino acid analysis:

Asp 1.01 (1), Lys 1.01 (1), GlußGln 4.03 (4) and Val 1.00 (1)

M+H 1121.5 Example 2 . Preparation of :

An ABI model 430 peptide synthesizer was charged with 700 mg of methyl BHA resin (substitution

0.7 mMßg). A standard protocol (provided by Applied Bio

System Inc. Foster City, CA.) was used to couple the amino acids in the target peptide sequence. After coupling the last amino acid, the resin was subjected to a deprotection and neutralization cycle and resulting free amino group was acetylated using 10% acetic

anhydride in pyridine. The peptide was cleaved from the resin using 10 ml HF and 1 ml anisole at -15ºC for two hours. HF was removed; the peptide-resin mixture was extensively washed with ether and the peptide was extracted in glacial acetic acid. The acetic acid was diluted with water so that the final concentration of the peptide was about 500 mgßl. The pH was adjusted to 7.5 and solution was stirred at room temperature.

Formation of the disulfide bond was monitored using

Ellman's reagent. Once the disulfide formation was complete the solvent was removed on a rotovap. The crude peptide was purified on a preparative C-18 (2×25 cm) HPLC Vydac column using 0.1% TFAßH₂O and a 0.1%

TFAßCH-CN gradient. It was analyzed for amino acid composition and molecular weight.

HPLC purity: >95%

Amino acid analysis :

Asn 1.00 (1), Gin 0.97 (1), Gly 2.02 (2), Leu

1.03 (1), Phe 1.01 (1), Ser 0.79 (1), and Thr

0.88 (1).

M+H 1071.3 Example 3. Preparation of:

An ABI model 430 peptide synthesizer was charged with 700 mg of methyl BHA resin (substitution 0.5 mMßg). A standard protocol provided by (Applied Bio Systems Inc. Foster City, CA.) was used to couple the amino acids in the target peptide sequence. In this synthesis, -SH group of cysteine was protected as ACM. After coupling the last amino acid, the resin was subjected to a deprotection and neutralization cycle and the resulting free amino group was acetylated using 10% acetic anhydride in pyridine. The peptide was cleaved from the resin using 10 ml HF and 1 ml anisole at -15ºC for two hours. After removal of HF, the resin was extensively washed with ether and the peptide extracted in 80% acetic acid. The final concentration of the peptide was about 500 mgßl. Iodine solution (10% in 80% Acetic acid) was titrated into the peptide solution so that the solution stayed a very faint pale yellow.

Formation of the disulfide was monitored using Ellman's reagent. Once the disulfide formation was complete the solvent was removed on a rotovap. The crude peptide was purified on a preparative C-l8 (2×25 cm) HPLC Vydac column using 0.1%TFAßH₂O and a 0.1%TFAßCH₃CN gradient. The peptide was analyzed for amino acid composition and molecular weight.

HPLC purity: >95%

Amino acid analysis:

Asn 1 . 00 (1) , Gly 1 . 02 (1 ) , Gin 0 . 97 (1 ) , Leu

1 . 07 (1 ) , Phe 1 . 12 ( 1 ) and Ser 0 . 77 ( 1 )

M+H 910 .3

Example 4. Additional Peptides

In addition to the peptides disclosed in

Examples 1-3, the following peptides were synthesized using the methods described above:

C'-C" loop (CDR-2 like) mimics. and

F-G loop (CDR-3 like) mimics.

and

and

The above examples and description fully disclose the present invention, including preferred embodiments thereof. This invention, however, is not limited to the precise embodiments described herein, but encompasses all modifications within the scope of the claims which follow.

Claims

What is claimed is:

1. A method to inhibit HIV infection which comprises administering to a patient a pharmaceutically effective amount of a conformationally constrained compound of the formula:

X-A-B-C-D-Y

wherein

A-B-C-D is a ß turn tetrapeptide or a ß-turn tetrapeptide mimic;

A-B-C-D binds to at least one HIV envelope protein; thereby inhibiting HIV infection, with the proviso that when X and Y both comprise a naturally occurring amino acid sequence present in the D1 domain of CD4, then X and Y together number less than 31 amino acids.

2. The method of claim 1 wherein:

A is selected from the group consisting of Gly, Ala, Ser, Thr, Asp, Asn, Glu, Gin, His, Lys and Arg;

B is any amino acid;

C is any amino acid; and

D is Leu, Phe or a Phe-mimic.

3. The method of claim 2 wherein X and Y number less than 13 amino acids.

4. The method of claim 2 wherein X and Y number less than 5 amino acids.

5. The method of claim 2 wherein X and Y number less than 3 amino acids.

6. The method of claim 2 wherein D is Phe.

7. The method of claim 2 wherein A is Gln.

8. The method of claim 2 wherein B is Gly.

9. The method of claim 2 wherein C is Ser.

10. The method of claim 2 wherein A-B-C-D is Gln-Gly-Ser-Phe.

11. A conformationally restricted compound of the formula:

X-A-B-C-D-Y

wherein

A is selected from the group consisting of Gly,

Ala, Ser, Thr, Asp, Asn, Glu, Gln, His, Lys and Arg;

B is any amino acid;

C is any amino acid;

D is Leu, Phe or Phe-minic;

A-B-C-D forms a beta turn;

with the proviso that when X and Y both comprise a naturally occurring amino acid sequence present in the Dl domain of CD4, then X and Y together number less than 31 amino acids;

or a pharmaceutically acceptable salt thereof.

12. The compound of claim 11 wherein D is Phe.

13. The compound of claim 11 wherein A is Gly, Ala, Ser, Asn or Gin.

14. The compound of claim 11 wherein A is Gin.

15. The compound of claim 11 wherein B is Gly, Ala, Val, Ser or Thr .

16. The compound of claim 11 wherein B is Gly.

17. The compound of claim 11 wherein C is Gly,

Ala, Val, Ser or Thr.

18. The compound of claim 11 wherein C is Ser.

19. The compound of claim 11 wherein A-B-C-D is

Gln-Gly-Ser-Phe.

20. The compound of claim 11 wherein X is one or more amino acids and Y is one or more amino acids.

21. The compound of claim 20 wherein X and Y are joined by a disulfide bridge.

22. The compound of claim 21 wherein:

X is Cys-(X₁)n;

Y is Cys-(Y₁)n;

(X₁) is any amino acid except Pro;

(Y₁) is any amino acid except Pro;

and n = 0 to 5.

23. The compound of claim 21 wherein:

X is selected from the group consisting of

Xaa_n-Cys-, Xaa_n-Cys-Asn-, Xaan-Cys-Gly-Asn- and Xaa_n-Cys-Leu-Gly-Asn-;

Y is selected from the group consisting of

Yaan-Cys-, Yaan-Cys-Leu-, Yaan-Cys-Thr-Leu and Yaa_n-Cys-Lys-Thr-Leu-;

Xaa is any amino acid except Pro;

Yaa is any amino acid except Pro; and n = 0 to 11.

24. The compound of claim 20 wherein X and Y are joined by a peptide bond.

25. The compound of claim 24 wherein:

X is selected from the group consisting of

Xaa_n-, Xaa_n-Gly-Asn- and Xaa_n-Ile-Leu-Gly-Asn-;

Y is selected from the group consisting of

Yaa_n-, Yaa_n-Thr-Leu- and Yaa_n-Gly-Lys-Thr-Leu;

Xaa is any amino acid capable of forming a peptide bond with Yaa;

Yaa is any amino acid capable of forming a peptide bond with Yaa;

and n = 1 to 11.

26. The compound of claim 25 wherein the pair

Xaa-Yaa comprises at least one Pro.

27. The compound of claim 25 wherein the pair Xaa-Yaa is Pro-D-Pro or D-Pro-Pro.

28. The compound of claim 24 wherein X and Y further form a disulfide bridge.

29. The compound of claim 28 wherein: X is Xaa -Ile-Leu-Cys-Asn-;

Y is Yaa -Gly-Lys-Cys-Leu-;

Xaa is any amino acid capable of forming a peptide bond with Yaa;

Yaa is any amino acid capable of forming a peptide bond with Xaa;

and n = 1 to 11.

30. The compound of claim 29 wherein the pair Xaa-Yaa comprises at least one Pro.

31. The compound of claim 29 wherein the pair Xaa-Yaa is Pro-D-Pro or D-Pro-Pro.

32. A recombinant protein in which the CDR-2 region of an immunoglobulin is replaced with the

compound of claim 11.

33. The protein of claim 32 wherein the immunoglobulin is of a human IgG subclass.

34. The protein of claim 33 wherein the immunoglobulin is a variable light chain (V_k) .

35. A composition for inhibiting HIV infection which comprises an effective amount of the compound of claim 11 and a pharmaceutically acceptable carrier.

36. The method of claim 1 wherein:

A is any amino acid;

B is selected from the group consisting of Gly, Ala, Ser, Thr, Asp, Asn, Glu, Gln, His, Lys and Arg;

C is any amino acid;

D is any amino acid;

with the proviso that when X and Y both comprise a naturally occurring amino acid sequence present in the D1 domain of CD4, then X and Y together number less than 31 amino acids.

37. The method of claim 36 wherein X and Y number less than 13 amino acids.

38. The method of claim 36 wherein X and Y number less than 5 amino acids.

39. The method of claim 36 wherein X and Y number less than 3 amino acids.

40. The method of claim 36 wherein A is Val.

41. The method of claim 36 wherein B is Glu.

42. The method of claim 36 wherein C is Asp.

43. The method of claim 36 wherein D is Gln.

44. The method of claim 36 wherein A-B-C-D is Val-Glu-Asp-Gln.

45. A conformationally restricted compound of the formula:

X-A-B-C-D-Y

wherein

A is any amino acid;

B is selected from the group consisting of Gly,

Ala, Ser, Thr, Asp, Asn, Glu, Gin, His, Lys and Arg;

C is any amino acid;

D is any amino acid;

A-B-C-D forms a beta turn;

or a pharmaceutically acceptable salt thereof.

46. The compound of claim 45 wherein A is Val.

47. The compound of claim 45 wherein B is Glu.

48. The compound of claim 45 wherein C is Asp.

49. The compound of claim 45 wherein D is Gin.

50. The compound of claim 45 wherein A-B-C-D is

Val-Glu-Asp-Gln.

51. The compound of claim 45 wherein X is one or more amino acid; and Y is one or more amino acids.

52. The compound of claim 51 wherein X and Y are joined by a disulfide bridge.

53. The compound of claim 52 wherein:

X is Cys-(X₁)n;

Y is Cys-(Y₁)n;

(X₁) is any amino acid except Pro;

(Y₁) is any amino acid except Pro;

and n = 0 to 5.

54. The compound of claim 52 wherein:

X is selected from the group consisting of

Xaa_n-Cys-, Xaa_n-Cys-Glu-, Xaa_n-Cys-Cys-Glu- and Xaa_n-Cys-Ile-Cys-Glu-;

Y is selected from the group consisting of

Yaa_n-Cys-, Yaa_n-Cys-Lys-, Yaa_n-Cys-Glu-Lys- and Yaa_n- Cys-Glu-Glu-Lys;

Xaa is any amino acid except Pro;

Yaa is any amino acid except Pro;

and n = 0 to 11

55. The compound of claim 51 wherein X and Y are joined by a peptide bond.

56. The compound of claim 55 wherein: X is selected from the group consisting of Xaa_n-, Xaa_n-Cys-Glu- and Xaa_n-Tyr-Ile-Cys-Glu-;

Y is selected from the group consisting of

Yaa_n-, Yaa_n-Glu-Lys- and Yaa_n-Val-Glu-Glu-Lys-;

Xaa is any amino acid capable of forming a peptide bond with Yaa;

Yaa is any amino acid capable of forming a peptide bond with Yaa;

and n = 1 to 11.

57. The compound of claim 56 wherein the pair

Xaa-Yaa comprises at least one Pro.

58. The compound of claim 56 wherein the pair Xaa-Yaa is Pro-D-Pro or D-Pro-Pro.

59. The compound of claim 55 wherein X and Y further form a disulfide bridge.

60. The compound of claim 59 wherein: X is Xaa_n-Tyr-Ile-Cys-Glu-;

Y is Yaa_n-Val-Glu-Cys-Lys-;

Xaa is any amino acid capable of forming a peptide bond with Yaa;

Yaa is any amino acid capable of forming a peptide bond with Xaa;

and n = 1 to 11.

61. The compound of claim 60 wherein the pair Xaa-Yaa comprises at least one Pro.

62. The compound of claim 60 wherein the pair Xaa-Yaa is Pro-D-Pro or D-Pro-Pro.

63. A recombinant protein in which the CDR-2 region of an immunoglobulin is replaced with the compound of claim 45.

64. The protein of claim 63 wherein the immunoglobulin is of a human IgG subclass.

65. The protein of claim 64 wherein the immunoglobulin is a variable light chain (V_k).

66. A composition for inhibiting HIV infection which comprises an effective amount of the compound of claim 45 and a pharmaceutically acceptable carrier.