WO1993024518A1

WO1993024518A1 - Design and synthesis of a cd4 beta-turn mimetic for inhibiting the binding of hiv gp120

Info

Publication number: WO1993024518A1
Application number: PCT/US1993/005084
Authority: WO
Inventors: Michael Kahn
Original assignee: University of Illinois at Urbana Champaign; University of Illinois System
Current assignee: University of Illinois at Urbana Champaign; University of Illinois System
Priority date: 1992-05-28
Filing date: 1993-05-28
Publication date: 1993-12-09
Anticipated expiration: 1994-11-28
Also published as: AU4396693A

Abstract

The invention provides materials and methods for synthesizing novel beta-turn mimetics, as well as the novel beta-turn mimetics themselves, and peptides containing the same. The invention specifically provides beta-turn mimetics that are inhibitors of the binding of HIV gp120 to amino acid residues 40-45 of human CD4, and mimetics of the CDR3 like region involved in syncytial formation.

Description

DESIGN AND SYNTHESIS OF A CD4 BETA-TURN MIMETIC FOR INHIBITING THE BINDING OF HIV GP120

Government Support

Portions of this invention were supported by National Science Foundation Grant CHE-8657046 and National Institute of Health Grants GM38260. The government has certain rights in this invention.

BACKGROUND OF THE INVENTION

1. Field of the Invention

The invention relates to the development of therapeutic agents for inhibiting HIV-1.

2. Summary of the Related Art The T cell surface glycoprotein CD4 acts as the cellular receptor for human immunodeficiency virus type 1 through recognition of the virus envelope glycoprotein gpl20. The development of agents that can inhibit CD4-gpl20 interaction is a critical goal in the field of therapeutic approaches to AIDS treatment. Fisher et al., Nature 221: 76-78 (1988) teaches that recombinant soluble

CD4 is an inhibitor of both virus replication and syncytium formation.

Harris et al., Eur. J. Biochem. 1££: 291-300 (1990), characterizes the 368 amino acid protein secreted from Chinese Hamster Ovary cells as recombinant

CD4. Layne et al., Nature 246: 277-279 (1990). discloses a mechanism for soluble CD4 inhibition of HIV infection.

Since the weight of agent that must be administered varies inversely for any given affinity with the molecular size of the agent, it is especially desirable to develop inhibitory agents smaller than intact recombinant CD4 protein. Berger et al., Proc. Natl. Acad. Sci. USA £5: 2357-2361 (1988), teaches that a truncated 180 amino acid fragment, representing approximately the N- terminal half of the extracellular region of CD4. is capable of binding gpl20.

Jameson et al., Science 24Q: 1335-1339 (1988). teaches that a synthetic analog of CD4 amino acid residues 25 to 58 inhibits HIV-1-induced cell fusion in a concentration dependent manner, and proposes that CD4 amino acids 37-53 comprise a binding site for the AIDS virus.

SUBSTITUTE SHEET Kalyanaraman et al., J. Immunol. 145.: 4072-4078 (1990), teaches that benzyl or acetyl derivative peptides corresponding to CD4 amino acids 81-92 inhibited HIV infection in vitro, whereas no peptide corresponding to any other CD4 region, including amino acids 23-56, inhibited HIV infection. Shapira-Nahor et al., Cell. Immunol. 12£: 101-117 (1990), discloses that peptides corresponding to CD4 amino acids 74-95 or 81-95 inhibit HIV infection in vitro, whereas no other peptide tested produce in vitro inhibition.

Rausch et al., Ann. Ny. Acad. Sci. £1£: 125-148 (1990), teaches that only peptides corresponding to CD4 amino acids 81-92 or 81-101 are effective antiviral agents.

Wang et al., Nature 24S: 411-418 (1990) and Ryu et al., Nature 24£: 419- 426 (1990) teach that amino acids within the CDR2-like region of CD4 are involved in gpl20 binding, that the region occupies a very prominent surface exposed site in the CD4 structure, and that the hairpin loop at Gln^ -Phe4 is highly accessible.

Although synthetic peptides hold some promise for the treatment of AIDS, they have many disadvantages as well. For example, these peptides are subject to degradation by intrinsic peptidases, reducing their in vivo half-life and thus perhaps affecting their effective therapeutic dose. Moreover, such peptides may have problems with bioavailability and antigenicity. Finberg et al. Science 249: 287- 291 (1990) teaches that poor bioavailability, rapid degradation and antigenicity affect the utility of proteinaceous pharmaceuticals. Jameson et al., Science 240: 1335-1339 (1990); and Brodsky et al., J. Immun. 144: 3078-3086 (1990) disclose a failure of peptides derived from the CD4 loop containing Gin 40-Thr 45, in attempts to inhibit HIV binding. A potential means of overcoming these problems is the use of peptide mimetics.

The invention further relates to peptide mimetics, which are chemical structures which serve as appropriate substitutes for peptides in interactions with receptors and enzymes. More particularly, the invention relates to the use of peptide mimetics to inhibit viral protein-receptor interactions necessary for viral infection.

Generally, peptide mimetics can be defined as structures which serve as appropriate substitutes for peptides in interactions with receptors and enzymes. The development of rational approaches for discovering peptide mimetics is a major goal of medicinal chemistry. Such development has been attempted both by empirical screening approaches and by specific synthetic design.

SUBSTITUTE SHEET Specific design of peptide mimetics has utilized both peptide backbone modifications and chemical mimics of peptide secondary structure. Spatola, Chemistry and Biochemistry of Amino Acids. Peptides and Proteins. Vol. VII (Weinstein, Ed.) Marcel Dekker, New York (1983), p. 267, exhaustively reviews isosteric amide bond mimics which have been introduced into biologically active peptides. The beta-turn has been implicated as an important site for molecular recognition in many biologically active peptides. Consequently, peptides containing conformationally constrained mimetics of beta-turns are particularly desirable. Such peptides have been produced using either external or internal beta- turn mimetics.

External beta-turn mimetics were the first to be produced. Friedinger et al., Science 21Q: 656-658 (1980), discloses a conformationally constrained nonpeptide beta-turn mimetic monocyclic lactam that can readily be substituted into peptide sequences via its amino and carboxy termini, and that when substituted for Gly^-Leu^ in luteinizing hormone releasing hormone (LHRH), produces a potent agonist of LHRH activity.

Monocyclic lactams have generally been useful as external beta-turn mimetics for studying receptor-peptide interactions. However, the mimetic skeleton in these molecules is external to the beta-turn, which gives rise to numerous limitations. Chief among these is bulkiness, which requires the use of dipeptide mimetics, rather than mimetics of all four residues in an actual beta-turn. Substantial flexibility retained in these beta-turn mimetics makes it unsafe to assume that expected conformations are present, absent considerable conformational analysis. For example, Vallee et al., Int. J. Pept. Prot. Res. 22: 181-190 (1989), discloses that a monocyclic lactam beta-turn mimetic did not contain an expected type II' beta-turn in its crystal structure. Another limitation of the monocyclic lactam beta-turn mimetics arises from the difficulty of producing molecules that effectively mimic the side chains of the natural peptide. These difficulties arise from steric hindrance by the mimetic skeleton, which results in a more effective mimic of the peptide backbone than of the side chains. Considering the great importance of side chains in receptor binding, these difficulties strongly limit the versatility of monocyclic lactams.

Although the use of bicyclic lactams reduces problems of flexibility somewhat, conformational analysis of peptides containing these mimetics may still be desirable. Moreover, the steric hindrance in these molecules may be even worse than that in the monocyclic lactams. Finally, both monocyclic and bicyclic

SUBSTITUTE SHEET lactams mimic only type II and type II' beta-turns, whereas type I and type III beta- turns are more prevalent in proteins and presumably in peptides.

The limitations presented by external beta-turn mimetics may be minimized by using mimetics in which the mimetic skeleton approximately replaces the space that was occupied by the peptide backbone in the natural beta-turn. Such molecules are known as internal beta-turn mimetics. Internal beta-turn mimetics may not generally reproduce the geometry of the peptide backbone of the particular beta- turn as accurately as external beta-turn mimetics. However, the internal position of the constraint allows replacement of larger sections of peptide, thus making tetrapeptide mimetics possible. The lack of bulk also diminishes the likelihood of steric hindrance of binding by the mimetic skeleton.

Internal beta-turn mimetics having biological activity are known in the art. For example, Krstenansky et al., Biochem. Biophys. Commun. 102: 1368-1374 (1982), discloses a leucine enkephalin analog in which an internal beta-turn mimetic replaced the residues Gly^-Gly^-Phe^-Leu^, and which acted as an analgesic with one-third the potency of morphine. Other internal beta-turn mimetics have been described.

Kahn et al., Tetrahedron Lett. 22: 4841-4844 (1986), discloses an internal beta-turn mimetic, based upon an indolizidinone skeleton, and designed to mimic the lysine and arginine side-chain disposition of the immunosuppressing tripeptide Lys-Pro-Arg.

Kahn et al., Heterocycles 25: 29-31 (1987), discloses an internal beta-turn mimetic, based upon an indolizidinone skeleton, and designed to correctly position the aspartyl and arginyl side chains of beta-turn in the proposed bioactive region of erabutoxin.

Kahn et al., Tetrahedron Lett. 2£: 1623-1626 (1987), discloses a type I beta-turn mimetic which can be incorporated into a peptide via its amino and carboxy termini, and which is designed to mimic an idealized type I beta-turn. See also Kahn et al., J. Am. Chem. Soc. JJD: 1638-1639 (1988); Kahn et al., J. Mol. Recogn. 1: 75-79 (1988).

Similarly, Kemp et al., Tetrahedron Lett. 22: 5057-5060 (1988), discloses a type II beta-turn mimetic which can be incorporated into a peptide via its amino and carboxy termini.

Arrhenius et al., Proc. Am. Peptide Symp., Rivier and Marshall, Eds. , Escom, Leiden (1990), discloses substitution of an amide-amide backbone hydrogen bond with a covalent hydrogen bond mimic to produce an alpha-helic mimetic.

SUBSTITUTE SHEET There have been numerous successes in obtaining mimetics which can force or stabilize peptide secondary structure. However, inhibition of the binding of a viral receptor by a designed peptide mimetic is not known in the art. In view of the devastating effect of HIV on infected individuals, there is a great need for therapeutic agents capable of inhibiting HIV infection. Peptide mimetics capable of inhibiting the binding of gpl20 to the CD4 molecule would provide a potentially life saving therapeutic.

For recent reviews of the related art, see Hruby et al., Biochem. J. 268: 249-262 (1990); Ball et al., J. Mol. Recogn. 2: 55-64 (1990); Morgan et al., Ann. Rep. Med. Chem. 24: 243-252 (1989); and Fauchere, Adv. Drug Res. 15: 29-69 (1986).

SUBSTITUTE SHEET BRIEF SUMMARY OF THE INVENTION The invention provides peptidomimetics that are capable of inhibiting CD4 binding to gpl20 and inhibiting syncytium formation. CD4 is a glycoprotein found on the surface of T lymphocytes; during HIV infection in humans, CD4 is the receptor for the gp 120 envelope glycoprotein of HIV. Extensive mutagenesis and peptide mapping experiments have mapped a critical binding region of CD4 to a single stretch of amino acids. Recent X-ray analysis has shown that within this stretch, residues 40-45 exhibit a beta-turn conformation. In addition, residues 83- 92 apparently are involved in syncytium formation. The invention provides a novel approach to synthesize conformationally restricted peptidomimetics of chain reversals in peptides and proteins. Thus the invention provides a method for the design and synthesis of a mimetic of residues Gin^O-Thr^ which form a m between the C and C" beta-strands of CD4. The method further provides a mimetic that inhibits binding of soluble gpl20 to cells expressing CD4 at moicromolar concentrations and reduces syncytium formation, and will provide a foundation for the development of low molecular weight gpl20 binding inhibitors as therapeutic agents.

SUBSTITUTE SHEET BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 Basic structure of mimetic ring system for residues 40-45 of

CD4. FIG. 2 Full CD4 loop region mimetic structure a (peptidomimetic 1).

FIG. 3 Synthetic scheme described in Example 1.

FIG. 4 Ten membered and twelve mimetic rings corresponding to the Val86_Gin89 region of CD4 that is implicated in syncytium formation. FIG. 5 Retro-Synthetic scheme for the mimetics shown in Figure 4.

FIG. 6 Synthetic pathway for component 4 of the retro-synthetic scheme shown in Figure 5. FIG. 7 Synthetic pathway for component 6 of the retro-synthetic scheme shown in Figure 5. FIG. 8 Synthetic pathway for component 7of the retro-synthetic scheme shown in Figure 5. FIG. 9 Inhibition of syncytium formation by mimetic 1 (asterisks), soluble CD4 (squares), or CD4 hexapeptide of residues 40-45 (crosses). FIG. 10 Data showing inhibition of gpl20 binding by soluble CD4 or by peptidomimetics of the invention. FIG. 11 A peptidomimetic having both CDR2-like and CDR3-like domains.

SU3 r---' ~ UTE SHEJ DETAILED DESCRIPTION OF THE SPECIFIC EMBODIMENTS The invention provides a peptide mimetic (or peptidomimetic) capable of inhibiting CD4 binding to gpl20. Infection of human mononuclear cells by human immunodeficiency virus (HIV) requires the binding of the viral envelope glycoprotein gpl20 to CD4, a cell surface glycoprotein that is found principally on the helper class of T-lymphocytes. Thus, a substance that could mimic CD4 in its ability to bind gpl20 could potentially provide a valuable therapeutic. Amino acids within the CDR2-like region of CD4 have been shown to be involved in gpl20 binding. This region occupies a very prominent surface exposed site in the CD4 structure. In particular, the hairpin loop at Gin40.p e43 j_s highly accessible. The basic structure of the mimetic ring system having inhibitory effect on gpl20 binding is shown in Figure 1. In a preferred embodiment, the constituents of the structure shown in Figure 1 are: R\ is CH2OH, CH3, CH2CH2CONH2 or a side chain of a synthetic or natural alpha amino acid and preferably CH3, R2 is a side chain of a natural or synthetic alpha amino acid or H; R3 is CH2OH a side chain of a natural or synthetic alpha amino acid, or a lower alkyl chain having 1 to 6 carbon atoms; R4 is C^CgHg, a natural or synthetic amino acid side chain, 1-naphthyl- alanine, or 2-naphthyl-alanine; R5 is one or more natural or synthetic amino acids, hydrogen, or an acetyl or benzoyl group; and R is an ether, amine, or one or more natural or synthetic amino acids, or a hydroxyl group. The full CD4 loop region mimetic structure (Mimetic 1) is shown in Figure 2, where Bn represents benzyl. Mimetics according to the invention inhibit gpl20 binding in a concentration dependent manner, with an IC50 in the micromolar range or better, as shown in Figure 10. The invention further provides peptide mimetics of the CDR3-like region, which is implicated in syncytium formation. A ten membered and twelve membered mimetic of this region is shown in Figure 4. In one preferred embodiment, the invention provides peptide mimetics of a beta-turn at the tip of the CDR3-like domain, comprising the residues Val^^, Glu^, As ^S and Gln89. Mimetics of this region may act as viricidal agents by causing premature exposure of the gp41 fiisogenic domain to cellular proteases, thus interfering with the CD4 induced release of gpl20 from virus and infected cells and interfering with syncytial formation. In the preferred embodiment of the mimetic shown in Figure 4, the mimetic structure exactly or closely duplicates the structure of the natural CDR3-like region 86-89. Mimetics according to this embodiment can promote gpl20 shedding prematurely, thereby making it unproductive. The

SUBSTITUTE SHEET constituents of the mimetics according to this embodiment of which Figure 4 is an example are: X is NH or

where R is H or CH3. R\ is an alpha amino acid side chain preferably valine; R2 is a glutamate or aspartate side chain (CH2CO2H or CH2CH2CO2H); R3 is an aspartate or glutamate side chain; R4 is a side chain of glutamine (CH2CH2CONH2), asparagine (CH2CONH2), aspartate, or glutamate; R- is acetyl, benzoyl, H, or a natural or synthetic amino acid, and R is an ether, OH, an amine, or a natural or synthetic amino acid. In a second aspect, the invention provides peptide mimetics having even greater ability to inhibit binding between CD4 and gpl20. For example, peptide mimetics described in the first aspect of the invention contain structures that mimic residues Gin 40-Thr 45 of the CDR2-like region of CD4. Such peptide mimetics have about 400 nM Kj) for gpl20, abrogate binding of gpl20 to CD4+ cells at low micromolar levels, and reduce syncytium formation. However, these properties can be further improved by altering the side chains of one or more of the structures mimicking residues Gln^O-Thr^^. The peptide mimetics according to the first aspect of the invention have side chains corresponding to those found in the natural peptide, a,., Gln-Gly-Ser-Phe-Leu-Thr. Peptide mimetics according to the second aspect of the invention have one or more modified side chain. The modifications result in superior inhibitory properties.

In a third aspect, the invention provides peptidomimetics incorporating both CDR2-like and CDR3-like domains. In such peptidomimetics, the two domains are linked in a fashion that closely mimics their spatial relationship in CD4. Although separated by 46 residues in primary sequence, the distance in the native molecule from the amino terminus of Asn 39 to the amino terminus of Glu 85 is only about 9.2 angstroms. The two loops of CD4 are positioned in the native molecule and

SUBSTITUTE SHEET are positioned by beta sheet. This region can be replaced by naphthalene biscarboxylic acid, which separates the amino termini of residues 39 and 85 by 9.4 angstroms. A preferred embodiment of peptidomimetics according to this aspect of the invention is shown in Figure 11. In this embodiment, R is most preferably either hydrogen or a methyl group. Of course, the substitution of various constituents of the structure can readily be accomplished, as for the individual CDR2-like and CDR3-like mimetics previously discussed, since mimetics according to this aspect of the invention are prepared by coupling the individual CDR2-like and CDR3-like mimetics. Such coupling is achieved by reacting a naphthalene biscarboxylic half acid benzylester with a protected CDR3- like mimetic, followed by hydrogenolytic deprotection and coupling to the CDR2- like loop and acidic removal of the t-butyl esters.

In a fourth aspect, the invention provides a novel method to synthesize conformationally restricted peptidomimetics of chain reversals in peptides and proteins. The method of the invention provides predictable variation of side chain orientation, backbone conformations and distances. The method of the invention allows the design and synthesis of a peptidomimetic of residues Gln40-Thι45 in the C'C" CDR2-like loop of CD4, which incorporates a conformationally restricted type II beta-turn mimetic. At micromolar concentration, this peptidomimetic abrogates binding of soluble gpl20 to cells expressing CD4 on their surface. An example of the method is shown in Figure 3. Synthesis of peptide mimetics corresponding to the Val^-Gln∑Φ region is illustrated in Figure 5 and described in more detail in Example 3.

In a fifth aspect the invention provides pharmaceutical formulations suitable for use in the treatment of AIDS. Such pharmaceutical formulations comprise the gpl20-binding inhibitory peptide mimetics of the invention in a physiologically acceptable carrier or diluent.

The following examples are provided to further illustrate the invention and are not limiting in nature.

EXAMPLE 1 Synthesis of a CDR-2-like Inhibitory Peptide Mimetic As outlined in Figure 3, D-aspartic acid dimethyl ester hydrochloride (a) (2.00 grams, 10.1 millimoles), t-butyldimethylsilyl chloride (1.68 grams, 11.1 millimoles) and catalytic dimethyl-aminopyridine (62 milligrams, 0.51 millimoles) were dissolved in 50 milliliters of methylene chloride. To this mixture was added triethylamine (3.24 milliliters, 23.3 millimoles) at room

SUBSTITUTE SHEET temperature slowly and the mixture was allowed to stir for about 12 hours at room temperature. The mixture was washed with aqueous ammonium chloride, saturated sodium bicarbonate and brine, dried over sodium sulfate and concentrated in vacuo. The residue was dissolved in 50 milliliters of diethyl ether. The solution was cooled to 0°C and 2.0 molar t-butylmagnesium chloride in diethyl ether (5.24 milliliters, 10.5 millimoles) was added dropwise. The mixture was allowed to warm to room temperature for about 12 hours with stirring and was cooled to 0°C again. Saturated ammonium chloride was added and the mixture was stirred for 30 minutes. Water was added to the mixture and the organic layer was separated. The aqueous layer was extracted with diethyl ether (2 x 30 milliliters). The combined organic extracts were washed with brine, dried over magnesium sulfate and concentrated in vacuo. The residue was dissolved in 60 milliliters of methanol. To this solution at room temperature, sodium borohydride (1.14 grams, 30.1 millimoles) was added to a flask equipped with a reflux condenser. The mixture began to reflux during the addition and ceased after 20 minutes. After 45 minutes in total the mixture was cooled to 0°C and aqueous ammonium chloride was added. The mixture was extracted with methylene chloride (3 x 50 milliliters). The combined organic extracts were dried over sodium sulfate and the volatiles were removed in vacuo. The residue was dissolved in 30 milliliters of methylene chloride. To this solution was added t-butyl-dimethylsilyl chloride (1.00 grams, 6.64 millimoles) and 4-dimethylaminopyridine (37 milligrams, 0.30 millimoles). Triethylamine (1.10 milliliters, 7.87 millimoles) was added slowly and the mixture was allowed to stir for about 12 hours at room temperature. The mixture was washed with aqueous ammonium chloride and brine, dried over sodium sulfate and concentrated in vacuo.

Flash chromatography of the residue on silica gel with hexane/ethyl acetate (9/1 on a volume basis) afforded 1.01 grams (30%) of the azetidinone having structure (b) as colorless liquid. *H NMR (400 MHZ, CDCI3): delta 3.74 (dd, J_a=3.96 Hz, J_b= 10.30 Hz, IH) 3.63 (dd, J_a=5.12 Hz, J_bl = 10.30 Hz, IH), 3.59 (m, IH), 3.04 (dd, J_b=5.28 Hz, J_b= 15.22 Hz, IH), 2.76 (dd, J_a=2.49 Hz,m J_b = 15.22 Hz, IH), 0.94 (s, 9H), 0.88 (s, 9H), 0.22 (s, 3H), 0.21 (s, 3H), 0.05 (S, 6H); ¹³C NMR (100 MHz CDCI3): delta 172.7, 65.3, 50.2, 41.2, 26.2, 25.8, -5.4, -5.5, -5.7.

A solution of lithium diisopropyl amide (2.5 millimoles in 25 milliliters of THF) was prepared and cooled to -78 °C. A solution of azetidinone (b) (323 milligrams, 1 millimole) in 10 milliliters of tetrahydrofuran (THF) was added dropwise and allowed to stir for 30 minutes at -78 °C. To this was added

SUBSTITUTE SHEET 400 microliters (4 millimoles) of butenyl bromide. Stirring was continued for 18 hours and the reaction allowed to come to room temperature. The reaction mixture was poured into saturated NH4CI solution and extracted 3 times with 50 milliliters portions of diethyl ether, dried over sodium sulfate and the solvent removed in vacuo. The residue was chromatographed on 15 grams of silica gel to provide 294 milligrams (78%) of the azetidinone having structure (c) (TBDMS = tertbutyldimethylsilyl) .

A flask was charged with a magnetic stirrer, 4 milliliters CCI4/CH3CN/H2O (1:1:2), azetidinone (160 milligrams, 0.44 millimoles) and NaIO4 (469 milligrams, 2.2 millimoles, 5 equivalents). To this biphasic solution, a catalytic amount of RuCl3«3H2θ was added, the mixture was stirred for about 12 hours at room temperature and taken up in ethyl acetate (25 milliliters) and H2O (10 milliliters). The organic layer was separated and the aqueous layer was saturated with sodium chloride (solid) and extracted with ethyl acetate (2 x 20 milliliters). The combined organic extracts were dried over Na2SO4 and concentrated to provide the azetidinone having structure (d) as an oil in 55-65% yield.

The azetidinone acid (d) (238 mg, 0.59 mmol) was dissolved in 30 ml THF and cooled to 0°C. To this solution was added NMM (147 μl, 2.25 equiv.) and iBuOCOCl (81 μl, 1.05 equiv.). The solution was stirred for 15 minutes at room temperature and then added to a solution of O-benzylserine benzylester (e) in 10 ml THF (with 1 equiv. NMM) at 0°C. The reaction was allowed to warm to room temperature and stirred for 12 hours. The reaction was then diluted with 20 ml EtOAc, washed with NaHCO3, brine and H2O and dried over Na2SO4. The volatiles were removed in vacuo to provide 176 mg (45% yield) after chromatography on Siθ2 2:1 Hex:EtOAc. The product was dissolved in methanol, a catalytic amount of 10% Pd/C was added and the reaction was placed under 1 atm H2 gas. After 1 hour the reaction was filtered through celite and volatiles were removed in vacuo to provide a quantitative yield of the acid (f). To a solution of the azetidinone (f) (116 mg, 0.20 mmol in 2 ml of THF at

0°C) was added NMM (22_/x 1, 1 equiv.) and iBuOCOCl (26μ 1, 1 equiv.). The reaction was stirred for 15 minutes at room temperature. To this was added a solution of the hydrazinophenylalanine derivative (g) (132 mg, 0.40 mmol in 2 ml of CH2CI2) (where Z represents a protective group) and the reaction was stirred for 16 hours at room temperature. Column chromatography on silica gel with 50: 1 CH2CI: MeOH as eluent afforded a 37% yield of the precyclization intermediate. Hydrogenolytic deprotection and closure was effected by dissolution in 5 ml

SUBSTITUTE SHEET MeOH, addition of a catalytic amount of 5% Pd/C and placing of this mixture under 1 atm H2 for 1 hour. Filtration through celite and removal of the volatiles in vacuo provided a nearly quantitative yield of the 10-membered ring methyl ester. The ester was dissolved in 2 ml of 4:1 MeOH:H2θ. To this was added 10 mg (1 equiv.) K2CO3 and the reaction was stirred at room temperature for 16 hours. Removal of the solvent in vacuo provided a quantitative yield of the carboxylic acid as its potassium salt (h).

The carboxylic acid potassium salt (h) (38 mg, 0.05 mmol) was dissolved in 400 pi 1 1:1 THF:H₂0. To this was added EDC (11 mg, 1.1 equiv.), HOBT (7.5 mg, 1.1 equiv.) and the protected dipeptide (i) (45 mg, 0.1 mmol) and the reaction was stirred at room temperature for 24 hours. Removal of the volatiles in vacuo and silica gel chromatography (50:1 CH2Cl2:MeOH) afforded 62% yield of the protected analog. A solution of this compound in 2 ml MeOH with 1 ml MeOH saturated with HC1 and 10 mg 10% Pd/C was placed under 1 atm H2 and stirred at room temperature for 16 hours. Filtration through celite and removal of the volatiles in vacuo afforded 22 mg gpl20 binding inhibitor (60% yield) (k).

EXAMPLE 2 Assessment of Inhibition of gp!20 binding

For measuring binding, fluoresceinated gpl20 was incubated with the mimetics or with soluble CD4 at 22 °C in binding buffer (Ca^{2 +} , Mg2+ free HBSS, 0.5% BSA, 0.05% sodium azide, pH 7.4). Approximately 300,000 cells (from a 10x10^ cell/ml stock) were added to tubes at 4°C in binding buffer, with a final volume of 100 microliters. Samples were incubated at 4°C for 40 min. washed in binding buffer and analyzed in FACS immediately. Data was acquired, gating on live cell population (always greater than 90%), and was consistent whether mimetics, gpl20 or other agents were added or not. Results are shown in Figure 10. Inhibition by Mimetic 1 was concentration dependent, with an IC50 of 0.8 micromolar.

EXAMPLE 3 Synthesis of a Peptide Mimetic Corresponding to the CDR3-like Region of CD4

The retro-synthetic pathway for peptide mimetics corresponding to the CDR3-like region of CD4 is shown in Figure 5. Specific reaction conditions are as

SUBSTITUTE SHEET described in Example 1. The azetidinone component, shown as (3) in Figure 5, is synthesized by Arndt-Eistart homologation of L-Valine and subsequent ring closure, as described in Buehschuler et al., Helv. Chim. Acta £Q: 2747-2755 (1977). Component (4), also shown in Figure 5, is synthesized by Evans' oxazolidone methodology. This synthetic pathway is shown in Figure 6. Synthesis of component 6 (in Figure 5) is accomplished by direct amination of glutamine with the oxaziridine shown in Figure 7 and described in Vidal et al., J. Chem. Soc, Chem. Commun. 435-436 (1991). Preparation of component 7 (in Figure 5) is accomplished via reductive amination as shown in Figure 8 and described in Gribble and Nutaitis, Org. Prep. & Proc. Int. 12: 317-384 (1985). The completed mimetics are tested for activity by assessing their ability to affect syncytial formation, as described in Example 4. Inactivity due to the absence of residues beyond the Glu85-Gln89 turn region can be remedied by addition of amino acid residues to the N terminus, C terminus or both.

EXAMPLE 4 Inhibition of Svncvtium Formation Sup Tl cells (see Weiner et al., Pathobiology 4: 1-20 (1991)) were used as target cells for infection. Dilutions (1:2) of soluble CD4, CD4 mimetic, or CD4 peptide were made in 96 well plates in RPMI 1640 media containing 10% fetal calf serum. H9/IIIB infected cells were then plated at a density of approximately 10^ cells per well. Sup Tl target cells were then added (5 x 10^ per well) and syncytium formation was qualitatively and quantitatively determined after a 3 day incubation period. The results using soluble CD4, the mimetic 1 shown in Figure 2, or the CD4 hexapeptide comprising residues 40-45 are shown in Figure 9. The number of syncytia per well counted on visual inspection was plotted against the concentration of CD4, mimetic, or peptide added. The mimetic 1 provided superior inhibition of syncytium formation.

EXAMPLE 5

Peptide Mimetiςs Having Modified Side Chains

Increased effectiveness of inhibition of binding between gpl20 and CD4 and of syncytium formation can be achieved by introducing modified amino acid side chains into the mimetic. For example, substitution of the Ser42 side chain of mimetic 1 by a Leu side chain increased the inhibitory effect of the mimetic. Other

SUBSTITUTE SHEET side chain modifications that can increase the inhibitory effect of mimetic 1 are substitution of the Phe⁴³ side chain by Trp or by 1 or 2 napthylalanine, substitution of the Gly⁴* side chain by Phe,Leu, or Ala, or changing Leu⁴⁴ to HomoArg to result in the following structure:

Certain modifications of side chains in the mimetic of the CDR3-like region of CD4, comprising residues 86-89 can result in an increased diminution of syncytium formation.

SUBSTITUTE SHEET

Claims

WHAT IS CLAIMED IS:

1. An inhibitor of HIV-mediated syncytium formation comprising a cyclic structure that is a beta-turn mimetic of amino acid residues 40-45 of CD4.

2. An inhibitor of HTV-mediated syncytium formation comprising a cyclic structure that is a beta-turn mimetic of amino acid residues 83-92 of CD4.

3. An inhibitor according to claim 2 having the structure shown in Figure 4; wherein X is NH, or

where R is H or CH3; R\ is an amino acid; R2 is a side chain of glutamate or aspartate; R3 is a side chain of aspartate or glutamate; R4 is a side chain of glutamine, asparagine, aspartate, or glutamate; R- is acetyl, benzoyl, H, or a natural or synthetic amino acid; and R5 is OH, and amine, or a natural or synthetic amino acid.

4. An inhibitor according to claim 3; wherein R is valine.

5. An inhibitor of the binding of CD4 to gp 120 having the structure shown in Figure 1; wherein R is CH3, CH2OH, CH2CH2CONH or a side chain of a natural or synthetic amino acid; R2 is a side chain of a natural or synthetic alpha amino acid or H; R3 is CH2OH, a side chain of a natural or synthetic alpha amino acid, or a lower alkyl chain having 1 to 6 carbon atoms; R4 is CH2C6H6, a side chain of a natural or synthetic amino acid, 1-naphthylalanine, or 2-naphthyl-alanine; R5 is one or more namral or synthetic amino acids, hydrogen, an acetyl group, or a benzoyl group; and R is an ether, an amine, OH, or one or more natural or synthetic amino acids.

6. An inhibitor according to Claim 5 wherein R\ is CH3.

SUBSTITUTE SHEET

7. A peptidomimetic having both a CDR2-like domain and a CDR3-like domain.

8. A peptidomimetic according to claim 7, wherein the CDR2-like and CDR3-like domains are connected by naphthalene biscarboxylic acid.

9. A peptidomimetic according to claim 8, having the following strucmre:

SUBSTITUTE SHEET