WO1995006477A1 - Inhibition of beta amyloid binding to glycosaminoglycans for treatment of alzheimer's disease - Google Patents

Abstract

The present invention relates to compounds that inhibit the binding of glycosaminoglycans and proteoglycans to beta amyloid peptide (A(beta); Aβ), and to compounds that inhibit Aβ activation of complement. In one aspect, the compound is a peptide having an amino acid sequence X-X-N-X, in which X is an amino acid with a cationic side chain and N is a neutral amino acid. In another aspect, the compound is a peptide having an amino acid sequence X-X-N-X-Z, in which X is an amino acid with a cationic side chain, N is a neutral amino acid and Z is a neutral amino acid. In yet another aspect, the compound is peptide having an amino acid sequence X1-N-X2-X3, in which at least two of X1, X2, and X3 are independently an amino acid with an anionic side chain and the third X is an amino acid with an anionic side chain or a neutral amino acid and N is independently a neutral amino acid. In yet a further aspect, the compound is an anionic disaccharide. The invention also relates to a method for treating Alzheimer's disease comprising administering a therapeutically effective amount of a compound as described above to a subject suffering from Alzheimer's disease. Pharmaceutical compositions are also provided.

Description

INHIBITION OF BETA AMYLOID BINDING TO GLYCOSAMINOGLYCANS FOR TREATMENT OF ALZHEIMER'S DISEASE

1. FIELD OF THE INVENTION

The present invention relates to compounds that inhibit the binding of glycosaminoglycans and/or proteoglycans to beta amyloid peptide (A(beta) ; A/3) , and to compounds that inhibit A(beta) activation of complement. The invention further relates to treatment of Alzheimer's disease by administration of the inhibitory compounds.

2. BACKGROUND OF THE INVENTION A pathological hallmark of Alzheimer's disease (AD) is the presence of senile (neuritic) plaques within the cortex, hippocampus and certain subcortical nuclei of the brain (Terry and Katzman, 1983, in Neurology of Aging, Katzman and Terry, eds. F.A. Davis, p. 51; Terry et al., 1981, Ann. Neurol. 10:184- 192) . The most well-studied component of these plaques is /3A4 or beta amyloid peptide (A/3) , a 40-43 residue peptide that is derived from the amyloid precursor protein (APP) (Glenner, 1984, Biochem. Biophys. Res. Comm. 120:885-890; Glenner, 1988, Cell 52:307-308; Mori et al. 1992, J. Biol. Chem. 267:17082-17086). It has been proposed that increased production of A/3 and its eventual deposition into insoluble plaques is a causative event in the etiology of AD. A/3 in insoluble plaques appears to activate the classical pathway complement (C) cascade without immunoglobulin mediation (Rogers et al., 1992, Res.

Immunol. 143 (6) :624-630; Rogers et al., 1992, Proc.

Natl. Acad. Sci. USA 89:10016-10020). 2.1. INTERACTION OF Ag AND PROTEOGLYCANS A_/8, while necessary, may not be sufficient in itself to induce the neuropathology of AD (Frederickson, 1991, Neurobiol. Aging 13:239-253). This concept receives further support from the recent observations that hβ is produced normally by a variety of cell types (Haass et al., 1992, Nature 359:322-327; Shoji et al., 1992, Science 258:126-129). While A/3 is ^a major component of AD plaques, many other macromolecules have been identified within these structures, including dermatan sulfate proteoglycan (Snow et al., 1992, J. Histochem. Cytochem. 40:105- 113) and heparan sulfate proteoglycan (HSPG) (Snow et al., 1988, Am. J. Pathol. 133:456-463). The latter appears to be found throughout the plaque core, and recent studies indicate that HSPG binds both APP (Narindrasorasak et al., 1991, J. Biol. Chem. 266:12878-12883) and A/3 (Snow et al., 1991, Soc. Neurosci. Abst. 17:1106) with high affinity. The significance of proteoglycan localization within senile plaques is not known. It had been previously postulated that proteoglycan binding to amyloid may induce the latter to be deposited as insoluble fibrils, and that such interaction may also make the amyloid resistant to proteolytic degradation (Frederickson, 1991, Neurobiol. Aging 13:239-253).

The aforementioned interactions of HSPG with A_/3 and its precursor appear to be mediated in part by the core protein of the proteoglycan (Narindrasorasak et al., 1991, J. Biol. Chem. 266:12878-12883; Snow et al., 1991, Soc. Neurosci. Abstr. 17:1106). However, the glycosaminoglycan chains of the proteoglycan may also be involved in the association with APP and A/3 since binding of both are diminished in the presence of heparin (Narindrasorasak et al., 1991, J. Biol. Chem. 266:12878-12883; Snow et al. , 1991, Soc. Neurosci. Abstr. 17:1106).

2.2. COMPLEMENT ACTIVATION ASSOCIATED WITH ALZHEIMER'S DISEASE

The classical complement pathway is an immune- defense mechanism that results in activated molecules that can initiate inflammation, phagocytosis and cell lysis. There are 11 distinct protein components (Clq,r,s, and C2-C9) in the classical complement system, and activation of the cascade results in the proteolytic conversion of many of these molecules into forms that subsequently trigger further amplification of the pathway. An ultimate result of complement activation is the formation of "membrane attack complexes" (MACs) composed of C5-C9 that insert into cellular membranes and form large transme brane channels that can lyse or damage foreign cells. A side-effect of complement activation can be "bystander" damage due to MAC attachment to nearby cells that are not specifically targeted for destruction.

The usual biological signal that causes activation of the classical complement pathway is the binding of specific immunoglobulins to antigens on foreign cells. If two immunoglobulin (Ig) G molecules are within close proximity on a cellular membrane, complement component Clq binds the Fc portions of the IgG proteins, with resulting activation of Cl and subsequent induction of the complement cascade. Prior research has shown that the binding of a single IgG to Clq is insufficient to cause activation, and it appears that two or more IgG molecules must bind to the multivalent Clq protein.

In Alzheimer's disease (AD) brain, there is evidence of activated complement components associated with the senile plaques and dystrophic neurites that are the hallmark pathological features of this devastating disease (see, e .g. , Johnson et al., 1992, Neurobiol. Aging 13:641-648). The presence of complement proteins in the AD brain is surprising, since normal brain is essentially devoid of complement. Previous reports of MAC localization to dystrophic neurites would suggest that the neuronal damage seen in AD results at least in part from "bystander" effects of complement activation.

Rogers and co-workers have shown directly in vitro and indirectly in situ that /3-amyloid deposited in amyloid plaques activates complement (Rogers et al., 1992, Proc. Natl. Acad. Sci. USA 89:10016-10020). In particular, Rogers et al. showed that Clq immunoreactivity colocalizes with A/3 containing AD pathological structures and not immunoglobulins in the brain of AD patients but not of nondemented elderly control patients; A/3 activates the classical complement pathway in a standard complement activation assay and an ELISA complement activation assay, and Aβ activates the full classical pathway in vivo (Rogers et al., 1992, Proc. Natl. Acad. Sci. USA 89:10016-10020; Rogers et al., 1992, Res. Immunol. 143 (6) :624-630) . Rogers et al. propose, based on several lines of direct and indirect evidence, that A/8-mediated complement activation may be a pathogenic mechanism in AD (Rogers et al., 1992, Res. Immunol. 143(6) :624-630) .

Citation of a reference herein shall not be construed as an admission that such reference is prior art to the present invention. 3. SUMMARY OF THE INVENTION The present invention relates to a compounds that inhibit the binding of glycosaminoglycans and ₅ proteoglycans to beta amyloid peptide (A(beta) ; A/3) , and to compounds that inhibit A/3 activation of complement. In one aspect, the compound is a peptide having an amino acid sequence X-X-N-X, in which X is a amino acid with a cationic side chain and N is a ₀ neutral amino acid. In another aspect, the compound is a peptide having an amino acid sequence X-X-N-X-Z, in which X is a amino acid with a cationic side chain, and N and Z are each independently a neutral amino acid. In yet another aspect, the compound is peptide ₅ having an amino acid sequence X_j-N-X₂-X₃, in which at least two of X_{l t} X₂, and X₃ are independently an amino acid with an anionic side chain and the third X is an amino acid with an anionic side chain or a neutral amino acid, and N is independently a neutral amino o acid. In yet a further aspect, the compound is an anionic disaccharide.

The present invention is based, in part, on the surprising discovery that relatively small molecules, such as peptides of 4 or 5 amino acids, and ₅ disaccharides, can inhibit the binding of Aβ with glycosaminoglycans and proteoglycans. Although not intending to be bound by any theory, the compounds of the invention are believed to reverse the glycosaminoglycan-mediated resistance of A_/3 peptide to ₀ proteolysis, thus allowing natural degradation of the Aβ peptide and elimination of the Aβ plaques. Deposition of Aβ results in the formation of amyloid plaques in Alzheimer's disease.

The invention is also based, in part, on the ₅ surprising discovery that relatively small molecules, i.e., the compounds of the invention, inhibit complement activation associated with beta-amyloid. Although not intending to be limited by any particular theory, it is believed that the molecules inhibit the interaction of Clq with A/3.

The invention further relates to a pharmaceutical composition comprising one or more of the inhibitory compounds and a pharmaceutically acceptable carrier.

The invention also relates to a method for treating Alzheimer's disease comprising administering a therapeutically effective amount of a compound as described above to a subject suffering from Alzheimer's disease.

4. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1. The amino acid sequence (SEQ ID NO:l) of the beta amyloid peptide (A/3) .

Figure 2. Heparin-induced aggregation of Aβ(1-28) examined as a function of pH. Aβ(1-28) at a concentration of 0.25 mM was incubated with 20 μm heparin in saline solutions at a variety of pH values. The percentage of soluble amyloid remaining in solution after incubation and centrifugation was determined for single samples at each pH value. Figure 3. Heparin affinity chromatography of A/8(l-28) at pH 4.0 (A) and 8.0 (B) . A/3(l-28) was injected onto a heparin affinity column as described in Section 6.1, infra , and non-bound peptide was eluted with either 20 mM sodium phosphate pH 8.0 or 20 mM sodium acetate pH 4.0. Bound peptide was eluted with a linear gradient of 0-0.5 M NaCl. Peptide elution was monitored by determining the absorbance at 280 nm.

Figure 4. The ability of heparin and heparan sulfate to cause aggregation of A|8(l-28) as a function of glycosaminoglycan concentration. A/3(1-28) at a concentration of 0.5 mM was incubated with various concentrations of either heparin (open circles) or heparan sulfate (closed circles) at pH 4.0, and the ₅ amount of soluble peptide remaining after incubation and centrifugation was measured for single samples as described in Section 6.1, infra .

Figure 5. Comparison of the effects of various glycosaminoglycans on the solubility of A/3(1-28) at pH ₀ 4.0 and 8.0. A/5(1-28) at a concentration of 0.25 mM was incubated either in the absence (-GAG) or presence of 20 μM dermatan sulfate (light diagonal hatch) , chondroitin sulfate (cross-hatch) or heparan sulfate (heavy diagonal hatch) at pH values of 4.0 and 8.0. ₅ Following centrifugation, the amount of peptide remaining soluble was determined at both pH's and expressed as a percentage of peptide found in non- centrifuged samples. All samples were assayed in triplicate, with the standard error of the mean shown o f°r each condition.

Figure 6. Glycosaminoglycan-induced aggregation of A_/S(l-40) at pH 3.5 and 8.0. Aβ(1-40) at a concentration of 0.25 mM was incubated either in the absence (-GAG) (heavy diagonal hatch) or presence of ₅ heparan sulfate (HS) (cross hatch) or chondroitin sulfate (CS) (light diagonal hatch) at pH values of 3.5 and 8.0. After centrifugation, the amount of peptide remaining in solution was determined for each sample and expressed as a percentage of peptide found ₀ in non-centrifuged samples (solid bar) . All samples were assayed in triplicate, with the standard error of the mean shown for each condition.

Figure 7. The binding of Aβ(13-17) to a heparin affinity column. A/3(13-17) was injected onto a ₅ heparin affinity column as described in Section 6, infra , and non-bound peptide was eluted (solid line) with either 20 mM sodium phosphate pH 4.0 (A) (open circles) or 20 mM sodium acetate pH 8.0 (B) , (closed circles) . Bound peptide was eluted with a linear gradient of 0-0.5 M NaCl (dotted line). -Peptide elution was monitored by determining the absorbance at 230 nm of collected fractions.

Figure 8. Competition of Aβ(13-17) with Aβ(1-28) for heparin binding. A. Heparin concentration of 5 /-----• B. Heparin concentration of 15 μ . Reaction solutions (20 μl, pH 4.0) contained 0.25 mM A/3(l-28), 5.0 mM A/8(13-17) and heparin, or 0.25 mM A/3(l-28) either alone or with heparin as a control. The reactions were incubated 45 in prior to centrifugation. Each bar represents one of the duplicate assay samples. Samples 1 and 2 contained A/5(1-28) alone; 3 and 4 contained A/3(1-28) with heparin; and 5 and 6 contained Aj3(13-17), Aj8(l-28) and heparin. Figure 9. Competition of Aβ(10-17) with Aβ(1-28) for heparin binding. All reactions (20 μl, pH 4.0) were incubated 45 min prior to centrifugation. Samples 1 and 2 contained A/3(1-28). Sample 3 contained 0.25 mM Aβ(1-28) and 15 μM heparin. Samples 4 and 5 contained A/3, heparin, and 5.0 mM A/3(10-17). Figure 10. Competition of Aβ(13-17) with A/5(l- 28) for chondroitin sulfate proteoglycan (CSPG) binding. Reaction solutions (10 μl, pH 4.0) were incubated 60 min. Samples 1 and 2 contained 0.25 mM A/5(l-28); samples 3 and 4 contained 0.25 mM A/3(l-28) and 8 μg CSPG; and samples 5 and 6 contained Aβ(1-28), CSPG, and 5.0 mM A/5(13-17).

Figure 11. Competition of poly-L-lysine with A/3(1-40) for heparin binding at pH 3.5. Samples 1 and 2 are A/3(l-40); samples 3 and 4 are A/3(l-40) with poly-L-lysine; samples 5 and 6 are Aj8(l-40) with heparin and samples 7 and 8 are A/5(1-40) with heparin and poly-L-lysine. Each bar represents a single sample. The assays were performed in duplicate. The specific concentrations of reagents and reaction conditions are described in the notes to Table 5, infra.

Figure 12. Competition of poly-L-lysine with A/5(1-40) for CSPG binding at pH 3.5. Samples 1 and 2 are Aβ (l-AO) ; samples 3 and 4 are A/5(l-40) and CSPG; and samples 5 and 6 are Aβ(1-40), CSPG, and poly-L- lysine. Each bar represents a single sample. The assays were performed in duplicate. The specific concentrations of reagents and reaction conditions are described in the notes to Table 5, infra .

Figure 13. Effect of CSPG on the proteolysis of A/5(1-40). Sample A contained 19.35 μg of A/3(1-40); sample B contained Aβ(1-40) and papain (2 μg) ; sample C contained A/3(l-40) and CSPG (30 μg) , which were incubated together for 1 hour prior to the addition of papain. The amount of A/3(1-40) was determined by densitometry after gel electrophoresis as described in Section 7.1.2, i-nfra.

Figure 14. The effect of increasing concentration of CSPG on papain proteolysis of A/3(l- 40). Each sample contained 21.5 μg of A_/S(l-40) at pH 4.0. The amounts of CSPG added were 0, 24 μg, 48 μg, 72 μg and 96 μg, followed by a 1 hour incubation. The reaction mixtures were then incubated with 2 μg of papain for 12 hours, electrophoresed, and the amount of A/5(1-40) quantified using a densitometer.

Figure 15. The effect of Aβ(13-16) and A/3(13-17) on CSPG-mediated protection of Aβ(1-40) from papain proteolysis. Figure 16. The effect of Aβ(20-24: Phe₂₀ → Glu₂₀) on CSPG-mediated protection of A/3(1-40) from papain proteolysis. A/3(1-40) (4.3 μg) was treated with papain after a 45 minute incubation with CSPG (30 μg) and a subsequent incubation with the Aβ(20-24: Phe₂₀ → Glu₂₀) peptide Glu-Ala-Glu-Asp-Val (150 μg) , all at pH 6.5. Control samples were incubated without the peptide, without the peptide or CSPG, and without the peptide, CSPG, and papain. Papain digestion ran for 18 hours, followed by gel electrophoresis. A/3(1-40) bands were quantitated using a densitometer.

Figure 17. The effect of heparin disaccharide on CSPG-mediated protection of Aβ(1-40) from papain proteolysis. The reaction conditions were the same as described for Figure 17, with the exception that the heparin disaccharide (9UA-2s-[l->4]-GlcNS-6S) was used at 120 μg in place of the A/3(20-24: Phe₂₀→Glu₂₀) peptide.

Figure 18. A variation of a complement fixation assay, described in Section 8.1.1 infra , was utilized to assay for amyloid-induced complement activation. Complement activity is expressed as a percent of control samples which did not receive Aβ during the initial incubation. Two forms of peptides were used; fresh peptide (Fl-40) was used within 3 days of solubilization and there was no evidence of appreciable aggregate formation, while aged peptide (Al-4) was solubilized and stored for 27 days before use. The aged peptide solutions were Congo Red birefringent, indicating that the peptide exists as fibrils.

Figure 19. A solid-phase binding assay for the evaluation of Clq binding to A/3(1-28). This assay is described in Section 8.1.2, infra. The amount of Clq binding to the dotted Aj3(l-28) was detected by im unoenzyme staining and quantified through densitometric analysis of the stained membranes. Figure 20. The effect of Aβ(13-16) on A/3(1-40)- induced complement activation was investigated in the complement fixation assay described in Section 8.1.1, infra . Heat-aggregated human gamma globulin (AHGG) was included in the experiment since immunoglobulins are the normal activators of complement. Aβ(13-16) was added at 50-fold molar excess relative to A/3(l- 40) . Figure 21. The effect of Aβ(20-24: Phe₂₀-Glu₂₀) on AjS(1-40)-induced complement activation was investigated using the complement fixation assay described in Section 8.1.1 infra . The pentapeptide was added at 12.5 to 50-fold molar excess relative to A/3(1-40).

Figure 22. Clq binding to Aβ(25-35) and A0(l-40) was investigated using the methodology described in Section 8.1.2, infra .

5. DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a compounds that inhibit the binding of glycosaminoglycans and/or proteoglycans to beta amyloid peptide (A(beta) ; A/3;

SEQ ID NO:l), and to compounds that inhibit A_/3 activation of complement, and therapeutic methods and compositions based thereon.

5.1. INHIBITORY COMPOUNDS OF THE INVENTION The inhibitory compounds of the invention are relatively small molecules, such as tetra- to hexapeptides or sulfated disaccharides that have the ability to inhibit the aggregation of A/3 with proteoglycans or glycosaminoglycans. Preferably, the molecules of the invention consist of a sequence of not greater than 6 amino acid residues, and generally not greater than 8 amino acid residues, and, in a particular embodiment, comprise the sequences of peptides described in the subsections below. In a preferred aspect, hydrophobic residues are added to the peptides of the invention, to enhance the ability of the peptides to cross the blood-brain barrier.

Alternatively, the compounds of the invention have the ability to inhibit A/3-mediated activation of complement. Specific tests for the ability to inhibit the aggregation of A/S with glycosaminoglycans; to reverse the glycosaminoglycan-mediated protection of A/3 from proteolysis; and to inhibit Aβ activation of complement are described in the Examples in Sections 6, 7 and 8, respectively. Accordingly, the compounds of the invention can inhibit or reverse the neuropathology associated with Alzheimer's disease. Although not intending to be limited to any particular theory, it is believed that the neuropathology can result from beta amyloid deposition, the inhibitory effects of proteoglycans and glycosaminoglycans in amyloid on nerve growth, and/or inappropriate complement activation that can destroy nerve cells. The present invention provides three classes of compounds that can inhibit aggregation of A/3 and glycosaminoglycans, or inhibit A/3-mediated complement activation, or both, described in detail below.

The peptide inhibitory compounds of the invention are preferably prepared using standard synthetic chemistry from, preferably, naturally occurring amino acids, or obtained commercially.

However, the peptides of the invention can also contain non-natural amino acids or cyclic peptides. Non-classical amino acids include but are not limited to the D-isomers of the common amino acids, α-amino isobutyric acid, 4-aminobutyric acid, hydroxyproline, sarcosine, citrulline, cysteic acid, t-butylglycine, t-butylalanine, phenylglycine, cyclohexylalanine, β- alanine, designer amino acids such as /5-methyl amino acids, Cα-methyl amino acids, Nα-methyl amino acids, and amino acid analogs in general. Furthermore, the amino acid can be the D (dextrorotary) or L (levorotary) amino acid.

The peptide may be prepared by methods that are known in the art. For example, in brief, solid phase peptide synthesis consists of coupling the carboxyl group of the C-terminal amino acid to a resin and successively adding N-alpha protected amino acids. The protecting groups may be any known in the art. Before each new amino acid is added to the growing chain, the protecting group of the previous amino acid added to the chain is removed. The coupling of amino acids to appropriate resins is described by Rivier et al., U.S. Patent No. 4,244,946. Such solid phase syntheses have been described, for example, by

Merrifield, 1964, J. Am. Chem. Soc. 85:2149; Vale et al., 1981, Science 213:1394-1397; Marki et al., 1981, J. Am. Chem. Soc. 103:3178 and in U.S. Patent Nos. 4,305,872 and 4,316,891. In a preferred aspect, an automated peptide synthesizer is employed.

Purification of the synthesized peptides can be carried out by standard methods including chromatography (e .g. , ion exchange, affinity, and sizing column chromatography) , centrifugation, differential solubility, or by any other standard technique for the purification of proteins. In a preferred embodiment, reverse phase HPLC (high performance liquid chromatography) is employed. The disaccharide compounds can be prepared synthetically by standard procedures or obtained commercially. 5.1.1 THE HEPARIN BINDING DOMAIN OF Aβ In one embodiment, the inhibitory compound is a peptide that competes with A/S for binding to a glycosaminoglycan or proteoglycan, and therefore inhibits Aβ binding thereto. Although not intending to be bound by any particular theory, it is believed that aggregation of A/5 with proteoglycans and glycosaminoglycans results in protection of the Aβ from proteases, and thus prevents the clearance of such amyloid deposits, resulting in neuropathology. It is also believed that the glycosaminoglycans or proteoglycans can directly contribute to neuropathology by inhibiting nerve growth and regeneration.

In a specific embodiment, infra, the inhibitory compound of the invention is a peptide of four or five amino acids based on the putative heparin binding domain of Aβ . The sequence of the heparin binding domain of A/5 is histidine₁₃-histidine₁₄-glutamine₁₅- lysine₁₆-leucine_I7. In the examples sections infra , peptides corresponding to A/5(13-17) and A/S(13-16) were able to inhibit the binding of A/S to heparin, chondroitin sulfate proteoglycan, and dextran sulfate, as shown in aggregation assays and by the ability of the peptides to reverse proteoglycan/glycosaminoglycan-mediated protection of A/5 from papain proteolysis.

In one aspect, the compound is a peptide having an amino acid sequence X-X-N-X, in which X is a amino acid with a cationic side chain and N is a neutral amino acid. Preferably, X is selected from the group consisting of histidine, lysine and arginine, and N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, and cysteine, and hydrophobic amino acids such as isoleucine, phenylalanine, and leucine. In a preferred embodiment, the peptide has the amino acid sequence histidine-histidine-glutamine- lysine, which corresponds to Aβ(13-16). In another embodiment, the peptide has the sequence X-X-X-X, with X as defined above; in a specific embodiment, such a peptide has the sequence Lys-Lys-Lys-Lys (SEQ ID NO:2) or His-His-His-His (SEQ ID NO:3). In another aspect, the compound is a peptide having an amino acid sequence X-X-N-X-Z, in which X is a amino acid with a cationic side chain, N and Z are each independently a neutral amino acid. Preferably, X is selected from the group consisting of histidine, lysine and arginine, N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, and hydrophobic amino acids such as isoleucine, phenylalanine, and leucine, and Z is selected from the group consisting of glycine, alanine, valine, leucine, isoleucine, serine, threonine, asparagine, glutamine, tryptophane, tyrosine, phenylalanine, methionine and cysteine. In a preferred embodiment, the peptide has the amino acid sequence histidine-histidine-glutamine-lysine-leucine, which corresponds to A/3(13-17) .

Any basic amino acid, including but not limited to histidine, lysine, asparagine, di-aminobutaric acid, and amino acid analogs having amines, guanidines, or other basic side chains and D-amino acids, can be substituted for any of the basic amino acids corresponding to positions 13, 14 and 16 of the Aβ heparin binding domain.

Any neutral (i.e., uncharged) amino acid can be substituted for the amino acid (N) corresponding to the 15 position of the A/3 peptide. For example, the amino acids glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine, as 5 well as analogs and optical isomers thereof, can be used at that position.

5.1.2 THE 20-24 REGION OF Aθ In another embodiment, the compound of the 0 invention is a peptide that competes with glycosaminoglycans for binding to the heparin binding domain of A/5. In particular, although not intending to be bound by any particular theory, it is believed that the 20-24 region of A/S binds to the heparin 5 binding domain of A/3 in an antiparallel fashion. A pentapeptide having the sequence Phe-Ala-Glu-Asp-Val or an analog thereof and thus corresponding to the amino acids 20-24 region of A/S is specifically preferred since such a peptide will target A/3, whereas o a peptide corresponding to Aβ(13-16) or (13-17) (see Section 5.1.1) will target all glycosaminoglycans or proteoglycans.

In specific example, infra , the peptide corresponding to the 20-24 region of Aβ and having the 5 structure phenylalanine-alanine-glutamic acid-aspartic acid reversed the glycosaminoglycan-mediated protection of A/3 from proteolysis. In another specific example, an analog of the 20-24 region, in which phenylalanine was substituted with glutamic acid 0 (SEQ ID NO:6) or aspartic acid (SEQ ID NO:7) had the same activity.

In one aspect, the compound is a peptide having an amino acid sequence X,-N-X₂-X₃, in which at least two of Xj, X₂, and X₃ are independently an amino acid with 5 an anionic side chain, and the third X is an amino acid with an anionic side chain or a neutral amino acid, and N is independently a neutral amino acid. Preferably, the amino acid with the anionic side chain is selected from the group consisting of aspartic acid and glutamic acid or other anionic amino acid analog known to one skilled in the art, and the non-anionic X and N are independently selected from the group consisting of any neutral amino acid, but preferably glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, and the hydrophobic amino acids isoleucine, phenylalanine, and leucine. In a specific embodiment, the peptide has the amino acid sequence phenylalanine-alanine-glutamic acid-aspartic acid, which corresponds to Aβ(20-23). In another specific embodiment, the peptide has the amino acid sequence aspartic acid-alanine-glutamic acid- aspartic acid (SEQ ID NO:4). In yet another embodiment, the peptide has the amino acid sequence glutamic acid-alanine-glutamic acid-aspartic acid (SEQ ID NO:5) .

A pentapeptide is also provided, having the sequence X,-N-X₂-X₃-B, with X,, N, X_2/ and X₃ as defined above, and B being any hydrophobic amino acid, including but not limited to leucine, valine, isoleucine, and phenylalanine. In this peptide corresponding to the 20-24 region of A/3, only two acidic amino acids (with an anionic side chain) need to be found in the pentapeptide; preferably, however, there are three acidic amino acids, to act as complements to the three basic amino acids in the heparin binding domain of Aj8.

The non-anionic amino acid in the structure can be any neutral amino acid, but preferably glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine. 5.1.3 THE 25-35 REGION OF AiS In another embodiment, the compound of the invention is a peptide that inhibits A3-induced 5 activation of the complement cascade. In a specific embodiment, such a peptide is A/5(13-17) or an analog thereof. In another specific embodiment, such a peptide is Aβ(25-35) or an analog thereof.

₀ 5.1.4 SULFATED DISACCHARIDES

In another embodiment, the compound of the invention is a sulfated disaccharide that competes with glycosaminoglycans for binding to the heparin binding domain of A/5. In particular, although not ₅ intending to be bound by any particular theory, it is believed that the sulfated disaccharide binds to the heparin binding domain of A/S. Such a disaccharide is preferred since it will target A/3 specifically instead of glycosaminoglycans and proteoglycans generally. o Thus, in yet a further aspect, the compound is an anionic disaccharide. Although any sulfated disaccharide is envisioned, in a specific embodiment the disaccharide is derived from heparin. In one embodiment, sulfated disaccharides of the invention ₅ contain either uronic or glucuronic acid. More particularly, the disaccharide is α-4-deoxy-L-threo- hex-4-enopyranosyluronic acid-[l->4] D-glucosamine-N- sulfate-6-sulfate.

₀ 5.2. MODIFICATION OF THE COMPOUNDS OF THE INVENTION In a preferred embodiment, the compounds of the invention are modified so as to permit or enhance their ability to cross the blood brain barrier. Such compounds would be preferred for oral or parenteral ₅ administration other than intraventricularly.

Suitable modifications of the compounds to enhance their ability to cross the blood-brain barrier include, but are not limited to, adding hydrophobic amino acids, coupling the compound to a lipid, coupling to transferrin, coupling to an antibody which recognizes the transferrin receptor, coupling to avidin, etc. Preferably, the chemical linkage effectuating the coupling is labile, e . g. , a disulfide bond. Other modifications of the peptides of the invention, such as chemical modifications known in the art can be carried out, e .g. , acetylation, amidation, phosphorylation, etc. In a specific embodiment, acetylation of the amino terminus and/or amidation of the carboxy-terminus are carried out.

5.3. DEMONSTRATION OF THERAPEUTIC UTILITY The peptides or disaccharides are tested in vitro and then preferably in vivo for the desired therapeutic utility.

Any in vitro assay known in the art can be used to detect inhibition of proteoglycan/glycosaminoglycan bind to A/3, or inhibition of complement activation by a peptide disaccharide of the invention. In a preferred aspect, the assays described in the examples sections infra are employed.

Peptides or disaccharides demonstrated to have the desired activity in vitro can be tested in vivo for the desired inhibitory activity. For example, such compounds can be tested in suitable animal model systems prior to testing in humans, including but not limited to rats, mice, chicken, cows, monkeys, rabbits, etc. Suitable model systems are also used to demonstrate therapeutic utility. For in vivo testing, prior to administration to humans, any animal model system known in the art may be used.

5

5.4. PHARMACEUTICAL COMPOSITIONS

According to the present invention, the peptide or disaccharide inhibitory compounds of the invention can be preferably prepared as a pharmaceutical ₀ composition with a pharmaceutically acceptable carrier for administration to a subject. As used herein, the term "pharmaceutically acceptable" means approved by a regulatory agency of the Federal or a state government or listed in the U.S. Pharmacopeia or other generally ₅ recognized pharmacopeia for use in animals, and more particularly in humans. The term "carrier" refers to a diluent, adjuvant, excipient, or vehicle with which the compound is administered. Such pharmaceutical carriers can be sterile liquids, such as water and o oils, including those of petroleum, animal, vegetable or synthetic origin, such as peanut oil, soybean oil, mineral oil, sesame oil and the like. Water is a preferred carrier when the pharmaceutical composition is administered intravenously. Saline solutions and ₅ aqueous dextrose and glycerol solutions can also be employed as liquid carriers, particularly for injectable solutions. Suitable pharmaceutical excipients include starch, glucose, lactose, sucrose, gelatin, malt, rice, flour, chalk, silica gel, ₀ magnesium carbonate, magnesium stearate, sodium stearate, glycerol monostearate, talc, sodium chloride, dried skim milk, glycerol, propylene, glycol, water, ethanol and the like. These compositions can take the form of solutions, ₅ suspensions, tablets, pills, capsules, powders, sustained-release formulations and the like. Suitable pharmaceutical carriers are described in "Remington's Pharmaceutical Sciences" by E.W. Martin. Such compositions will contain a therapeutically effective amount of the active compound, preferably in purified form, together with a suitable amount of carrier so as to provide the form for proper administration to the patient.

In a specific embodiment, the composition is formulated in accordance with routine procedures as a pharmaceutical composition adapted for intravenous administration to human beings. Typically, compositions for intravenous administration are solutions in sterile isotonic aqueous buffer. Where necessary, the composition may also include a solubilizing agent and a local anesthetic to ease pain at the site of the injection. Generally, the ingredients are supplied either separately or mixed together in unit dosage form, for example, as a dry lyophilized powder or water free concentrate in a hermetically sealed container such as an ampoule or sachette indicating the quantity of active, agent. Where the composition is to be administered by infusion, it can be dispensed with an infusion bottle containing sterile pharmaceutical grade water or saline. Where the composition is administered by injection, an ampoule of sterile water for injection or saline can be provided so that the ingredients may be mixed prior to administration. The peptides or disaccharides of the invention can be formulated as neutral or salt forms. Pharmaceutically acceptable salts include those formed with free amino groups such as those derived from hydrochloric, phosphoric, acetic, oxalic, tartaric acids, etc., and those formed with free carboxy1 groups such as those derived from sodium, potassium, ammonium, calcium, ferric hydroxides, isopropylamine, triethylamine, 2-ethylamino ethanol, histidine, procaine, etc. The invention also provides a pharmaceutical pack or kit comprising one or more containers filled with one or more of the ingredients of the pharmaceutical compositions of the invention. Optionally associated with such container(s) can be a notice in the form prescribed by a governmental agency regulating the manufacture, use or sale of pharmaceuticals or biological products, which notice reflects approval by the agency of manufacture, use or sale for human administration.

5.5. THERAPEUTIC ADMINISTRATION OF THE COMPOUNDS

The compounds of the present invention can be used for the treatment of symptoms of amyloidosis associated with Alzheimer's disease or other diseases including but not limited to AA (inflammation- associated)-amyloid, AL-amyloid (amyloid with deposition of immunoglobulin light chains), Down's syndrome, and prion diseases such as Creutzfeldt-Jakob disease, Gerstmann-Straussler syndrome, kuru, and scrapie. Preferably, the compounds of the invention are used for treatment of Alzheimer's disease (AD) . It is contemplated that peptide or disaccharide compounds can be administered to a subject in need of prophylactic or therapeutic treatment. As used herein, the term "subject" refers to an animal, more preferably a mammal, and most preferably a human. Thus, the therapeutic treatment can commence with diagnosis of AD, or the onset of AD, according to the appropriate criteria. Once a subject has been identified as having a need for therapeutic treatment of AD, a therapeutically effective dose of a compound of the invention can be administered to the subject. What constitutes a therapeutically effective amount in a particular case will depend on a variety of factors within the knowledge of the skilled practitioner. Such factors include the physical condition of the subject being treated, the severity of the condition being treated, the disorder or disease being treated, and so forth.

The peptide or disaccharide compounds, particularly those optimized to cross the blood-brain barrier, can be administered systemically, and more preferably parenterally, i.e., via an intraperitoneal, intravenous, perioral, subcutaneous, intramuscular, intraarterial, etc. route, in order to treat Alzheimer's disease. Methods of introduction include but are not limited to intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, intranasal, epidural, and oral routes. The peptides or disaccharides may be administered by any convenient route, for example by infusion or bolus injection, by absorption through epithelial or mucocutaneous linings (e .g. , oral mucosa, rectal and intestinal mucosa, etc.) and may be administered together with other biologically active agents. Administration can be systemic or local. In a preferred aspect, the compounds are directly administered to the cerebrospinal fluid by intraventricular injection. Pulmonary administration can also be employed.

In a specific embodiment, it may be desirable to administer the peptides or disaccharides of the invention locally to the area in need of treatment; this may be achieved by, for example, and not by way of limitation, local infusion during surgery, topical application, by injection, by means of a catheter, by means of a suppository, or by means of an implant, said implant being of a porous, non-porous, or gelatinous material, including membranes, such as sialastic membranes, or fibers.

In another embodiment, the therapeutic compound can be delivered in a vesicle, in particular a liposome (see Langer, Science 249:1527-1533 (1990); Treat et al., in Liposomes in the Therapy of Infectious Disease and Cancer, Lopez-Berestein and Fidler (eds.), Liss, New York, pp. 353-365 (1989); Lopez-Berestein, ibid. , pp. 317-327; see generally ibid. )

In yet another embodiment, the therapeutic compound can be delivered in a controlled release system. In one embodiment, a pump may be used (see Langer, supra; Sefton, CRC Crit. Ref. Biomed. Eng. 14:201 (1987); Buchwald et al., Surgery 88:507 (1980); Saudek et al., N. Engl. J. Med. 321:574 (1989)). In another embodiment, polymeric materials can be used (see Medical Applications of Controlled Release, Langer and Wise (eds.), CRC Pres., Boca Raton, Florida (1974) ; Controlled Drug Bioavailability, Drug Product Design and Performance, Smolen and Ball (eds.), Wiley, New York (1984); Ranger and Peppas, J. Macromol. Sci. Rev. Macromol. Chem. 23:61 (1983); see also Levy et al., Science 228:190 (1985); During et al., Ann. Neurol. 25:351 (1989); Howard et al., J. Neurosurg. 71:105 (1989)). In yet another embodiment, a controlled release system can be placed in proximity of the therapeutic target, i.e., the brain, thus requiring only a fraction of the systemic dose (see, e . g. , Goodson, in Medical Applications of Controlled Release, supra , vol. 2, pp. 115-138 (1984)). Other controlled release systems are discussed in the review by Langer (Science 249:1527-1533 (1990)). The invention can be better understood by reference to the following examples, which are provided merely by way of exemplification and are not intended to limit the invention.

6. EXAMPLE: pH-DEPENDENT BINDING OF SYNTHETIC β-AMYLOID PEPTIDES TO GLYCOSAMINOGLYCANS

In this example, we have tested whether a

A/S(12-17) peptide, having the sequence

Val-His-His-Gln-Lys-Leu, interacts with glycosaminoglycans in a pH-dependent manner; in particular, whether there is increased association of the peptide with the sulfated polysaccharides below pH

6-7 where the histidine residues at position 13 and 14 would be protonated. This example also reports on the ability of peptides to compete with Aβ peptide or on

A/S(1-28) peptide for binding to heparin and chondroitin sulfate proteoglycan.

6.1. MATERIALS AND METHODS 6.1.1 Aβ AGGREGATION ASSAY The association of Aβ peptides with glycosaminoglycans was evaluated by analyzing the amount of peptide that would remain soluble after incubation with the polysaccharides followed by centrifugation. Amyloid peptides (stored as 2 mM stock solutions in water) were diluted to concentrations of 0.25 or 0.5 mM and mixed with glycosaminoglycans (concentrations ranging from 0.1- 100 μM) for approximately 1 h at room temperature (21- 24° C) in either 20 mM sodium acetate pH 3.5, 4.0 or 5.0 with 150 mM NaCl, or 20 mM sodium phosphate pH 6.0, 7.0 or 8.0 containing 150 mM NaCl. The precise concentrations of amyloid and glycosaminoglycans used for individual experiments are noted in the figure legends and Tables. Typically, the final volume of the mixtures was 10 or 20 μl. The peptides used were A/3(l-28) ([glnⁿ]-/S-amyloid; Bachem, Inc.), A/5(l-40) (synthesized by Biosynthesis, Inc.), A/S(25-35) (Bachem, Inc.), and A/3(13-17) (synthesized as described infra) . The glycosaminoglycans employed were heparin (porcine intestinal mucosa; CalBiochem, Inc.), heparan sulfate (bovine kidney; Seikagaku Corp.), chondroitin sulfate (bovine trachea chondroitin sulfate A; Sigma Chemical Co.), and dermatan sulfate (bovine mucosa chondroitin sulfate B; Sigma Chemical Co.). In addition to the naturally occurring glycosaminoglycans, the polymers dextran sulfate (mol. wt. = 5,000; Sigma Chemical Co.) and dextran (mol. wt. = 10,200; Sigma Chemical Co.) were used in the aggregation assay. Poly-L-lysine, molecular weight 30-70,000, was obtained from ICN. Subsequent to the incubation period, the mixtures were centrifuged in a microfuge at 10,000 x g for 10 min, and aliquots of the resulting supernatants were assayed for absorbance at 230 nm. As controls, samples were prepared without glycosaminoglycan addition and treated as above to determine if amyloid peptides were sedimenting in the absence of polysaccharide. The absorbance values obtained for the controls and samples were compared to the absorbance of peptide in the appropriate buffer that had not undergone centrifugation. The values obtained were expressed as percent soluble amyloid.

6.1.2 HEPARIN AFFINITY CHROMATOGRAPHY

A/S(l-28), A/3(25-35) and A/3(13-17) were injected onto a heparin affinity column (Bio-Rad Econo-Pac cartridge; 5 ml volume) that was equilibrated with either 20 mM sodium acetate, pH 4.0 or 20 mM sodium phosphate, pH 8.0. The peptides were at a concentration of 0.25 mM in water, with 0.2 ml injected. Non-bound peptides were eluted from the column with approximately 15 ml of the equilibration 5 buffer (flow rate of 1 ml/min) , and bound peptides were then eluted with a linear gradient of 0 to 0.5 M NaCl in equilibration buffer over 25 min. The change in NaCl concentration was monitored with an in-line conductivity meter. The elution of Aβ(1-28) was 0 monitored at 280 nm. For A/3(25-35), and A/S(13-17), 2 ml fractions were collected from the column and assayed for absorbance at 230 nm using a spectrophotometer.

5 6.1.3 COMPETITION ASSAYS

Aggregation assays with Aβ(1-28) or A/3(1-40) were performed as described in Section 6.1.1, supra , in the presence of excess amounts of test peptides. The test peptides for inhibition of aggregation were Aβ(13-17), o Aβ (lO-n) , A/8(l-28) (tested for its ability to inhibit aggregation of A/3(l-40), and poly-L-lysine, typically at 5.0 mM (about a 20-fold molar excess) concentration.

Reactions were incubated for 45 minutes to three 5 hours at room temperature, then centrifuged for 10 minutes at 14,000 rpm in a microfuge. Five to 10 μl of each supernatant was carefully removed and added to an equal volume of Tris-Tricine sample buffer (8% SDS, 24% glycerol, 0.1 M Tris base, 0.1 M Tricine, 0.05% 0 Bromophenol Blue). Samples were analyzed on 16.5% Tris-Tricine gel (16.5% acrylamide, 1 M Tris base, 0.1% SDS, 13.3% glycerol, pH 8.45) with a 4% stacking gel (4% acrylamide, 1 M Tris base, 0.1% SDS, pH 8.45). The anode buffer used was 0.2 M Tris base, pH 8.9, and 5 the cathode buffer was 0.1 M Tris base, 0.1 M Tricine, 0.1% SDS. The gels were stained with 0.2% Coomassie brilliant blue in 45% methanol and 10% acetic acid, then destained in 45% methanol and 10% acetic acid. The gels were scanned with densitometer and Aβ(1-28) or A/8(l-40) protein bands were quantified.

6.1.4 SYNTHESIS OF Aβ(13-17) A/8(13-17) was synthesized by stepwise elongation from the COOH-terminus using solid-phase methodology (Merrifield, 1963, J. Am. Chem. Soc. 85:2149-2154). Rink resin (Rink, 1987, Tetrahedron Lett. 28:3787-3790) was used as the solid support and fluorenylmethoxycarbonyl (Carpino and Han, 1970, J. Am. Chem. Soc. 92:5748-5749; Carpino and Han, 1972, J. Org. Chem. 37:3404-3409) moieties for protection of the amino groups. The side chains of lysines were protected with butoxycarbonyl and those of histidines with fluorenylmethoxycarbonyl. All amino acid residues except glutamine were introduced using diisopropylcarbodiimide hydroxybenzotriazole; for glutamine, pentafluorophenyl ester was employed. Fluorenylmethoxycarbonyl groups were removed with piperidine-dimethylformamide and the desired peptide sequences were cleaved from the solid support with trifluoroacetic acid. The resulting peptides were purified by Sephadex G10 chromatography followed by Cjg-reversed phase HPLC.

6.2. RESULTS 6.2.1 pH-DEPENDENT INTERACTION OF HEPARIN WITH A3 To assay for the association of A/5(1-28) with heparin, we mixed aliquots of amyloid and heparin solutions together in aqueous saline solutions buffered at pH values ranging from 4 through 8. Visual inspection of the mixtures revealed that the more acidic solutions contained a precipitate which could be sedimented by centrifugation. Spectrophotometric analysis of the supernatant at 230 nm following centrifugation gave a measure of the amount of remaining soluble peptide, and a plot of this value as a function of pH is shown in Fig. 2. The ability of the heparin to interact with A/8(1-28) and cause precipitation was pH-dependent , with increased aggregation below pH 7. Essentially no precipitable material was seen at pH 8. The pH-dependent precipitation of A/3(1-28) was due to heparin and not to a gross change in the solubility of the peptide itself, since we did not observe sedimentation of A/3(1-28) in the absence of glycosaminoglycan at any of the pH values tested. These observations are consistent with the involvement of one or more histidine residues in heparin binding, and suggest that the histidines at residues 13 and/or 14 of A?(l- 28) are involved in the interaction with the glycosaminoglycan.

To confirm that the aggregation of A/3(1-28) by heparin was a reflection of pH-dependent binding of the glycosaminoglycan to the peptide, the association of Aβ(1-28) with a heparin affinity column was investigated as a function of pH. As noted in Fig. 3, A/3(1-28) bound to the heparin matrix at pH 4.0 and was eluted with a NaCl gradient (elution at approximately 0.22 M NaCl). In contrast, the vast majority of the peptide was not retained on the affinity column at pH 8.0, and that peptide which did bind was eluted with a low concentration of NaCl (0.04 M) . These data support the notion that A/3 interacts with heparin in a pH- dependent fashion. 6.2.2 GLYCOSAMINOGLYCAN SPECIFICITY OF THE Aβ INTERACTION

The pH-dependent interaction of heparin with A/8(1-28) implied that this highly sulfated glycosaminoglycan was associating with the consensus heparin-binding domain located at residues 12-17 of the peptide. The binding of heparin to Aβ(1-28), with resulting aggregation, was concentration-dependent (Fig. 4). With a peptide concentration of 0.5 mM at pH 4.0, appreciable precipitation occurred over a range of 2-50 μM heparin. Although heparin resembles the heparan sulfate glycosaminoglycan chains of HSPG in that both contain predominantly N-acetylglucosamine and uronic acids as constituent monosaccharides, the former has a higher degree of sulfation than the HSPG polysaccharides. To determine whether the observed interaction of heparin with Aβ was a result of this increased level of sulfation, we examined the more physiologically relevant polysaccharide, heparan sulfate, in the aggregation assay at pH 4.0. As demonstrated in Fig. 4, heparan sulfate and heparin showed nearly identical dose-dependent aggregations of

While heparan sulfate appeared to be comparable to heparin in its ability to precipitate A/3(1-28), it was of interest to determine whether glycosaminoglycans containing differing monosaccharide units could also interact with the amyloid peptide. Chondroitin sulfate and dermatan sulfate both contain N-acetylgalactosamine residues in addition to uronic acids in their polysaccharide chains. These two glycosaminoglycans caused a level of aggregation of A/S(1-28) at pH 4.0 that was similar to that seen with heparan sulfate (Fig. 5) . None of the sulfated polysaccharides caused significant precipitation of Aβ(1-28) at pH 8.0, consistent with the involvement of one or more histidine residues in the interaction.

The ability of a variety of glycosaminoglycans to bind to Aβ(1-28) at pH values below neutrality indicated that the association was relatively sugar non-specific, and implied that the binding was mediated through ionic interaction of sulfate groups with the positively charged amino acids of the peptide. To further address the nature of the glycosaminoglycan-A/3 interaction, we examined whether the polymer, dextran sulfate, could also bind to and aggregate Aβ(1-28) . As with the sulfated glycosaminoglycans, dextran sulfate addition to A/3(l- 28) at pH 4.0 resulted in appreciable precipitation of the peptide (Table 1) . Further evidence that this binding was due to the sulfate groups of the polysaccharide is seen in the inability of dextran to cause aggregation of Aβ (Table 1) .

Table 1

Comparison of Dextran Sulfate- and Dextran-Induced Aggregation of A3C1-28)

Polymer Added % Soluble A/3(1-28)

Dextran Sulfate 14.6

Dextran 99.2

Aβ(1-28) at a concentration of 0.25 mM was incubated with either 0.4 mg/ml dextran sulfate (80 μM) or 0.4 mg/ml dextran (40 μM) at pH 4.0. The percentage of peptide remaining in solution following centrifugation was then determined as described in Materials and Methods. 6.2.3 ASSOCIATION OF GLYCOSAMINOGLYCANS WITH OTHER Aβ PEPTIDES

The association of Aβ(1-28) with glycosaminoglycans suggested that such an interaction may be relevant to the localization of proteoglycans within the senile plaques of AD brain. The amyloid peptide of the AD plaques is 40-43 amino acids in length (SEQ ID NO:l) (Masters et al. , 1985, EMBO J.

4:2757-2763; Mori et al., 1992, J. Biol. Chem.

266:12878-12883). We examined whether A/3(l-40) could be precipitated by glycosaminoglycans. Addition of either heparan sulfate or chondroitin sulfate to A/3(l-

40) at pH 8.0 resulted in essentially no aggregation and precipitation of the peptide (Fig. 6). When A/3(l-

40) was incubated at pH 4.0 in the absence of glycosaminoglycan, we noted that approximately 55% of the peptide could be sedimented. The amount of A/3(l-

40) that was precipitable in the absence of glycosaminoglycan increased to more than 85% at pH values of 5.0 and 6.0 (data not shown). This increased insolubility of the peptide between pH 4-7 is consistent with previous reports (Barrow and

Zagorski, 1991, Science 253:179-182; Zagorski and

Barrow, 1992, Biochemistry 31:5621-5631), and thus precluded accurate determination of glycosaminoglycan- induced aggregation of the peptide at pH 4-7. At pH

3.5, Aβ(1-40) remained soluble after centrifugation in the absence of glycosaminoglycan (Fig. 6) . Addition of heparan sulfate or chondroitin sulfate to A/3(1-40) at pH 3.5 caused a significant increase in the amount of peptide that could be sedimented (Fig. 6) , indicating that the sulfated polysaccharides were interacting with the amyloid peptide.

To investigate whether the glycosaminoglycan- binding domain of A/3 resides within residues 12-17, a pentapeptide containing the His-His-Gln-Lvs-Leu sequence corresponding to amino acids 13-17 was analyzed for its ability to bind to a heparin affinity ₅ column at pH 4.0 and 8.0. As with A/3(l-28), this short peptide showed an acidic pH requirement for heparin association (Fig. 7) , with little binding to the column at pH 8.0. Essentially all of the A/3(13- 17) bound to the heparin column at pH 4.0, and the 0 peptide was eluted at a NaCl concentration of approximately 0.45 M. This binding was tighter than that seen with A/3(1-28), presumably reflecting a greater accessibility of the charged residues of the pentapeptide to the heparin matrix. In contrast to 5 the results seen with A/3(13-17), a peptide containing residues 25-35 of the amyloid peptide sequence did not bind to a heparin affinity column at pH 4.0 (data not shown) . This is consistent with the localization of the glycosaminoglycan-binding domain within the first o 25 amino acids of the amyloid peptide.

6.2.4 COMPETITION OF Aβ PEPTIDES FOR BINDING GLYCOSAMINOGLYCANS

Peptides corresponding to the heparin binding domain of Aβ were tested for the ability to inhibit 5 binding of Aβ(1-28) or A/3(1-40) to the glycosamino¬ glycans heparin and chondroitin sulfate proteoglycan. The aggregation assay was used to demonstrate binding • of the Aβ peptide to the glycosaminoglycan.

Aβ(13-17) inhibited binding of A/3(1-28) to 0 hepariri at both 5 and 15 μM (Fig. 8) . The amount of free Aβ(1-28) decreased by precipitating with heparin, and increased in the presence of A/3(13-17), which indicates that A/3(13-17) inhibited the binding of A|8(l-28) with heparin. The results shown in Fig. 8 5 are summarized below in Table 2. Table 2

Competition of Aβ(13-17) With A/3(1-28) For Heparin Binding

Sample - 5uM Heparin Av . 0.D. % Soluble

1-28 0.645 100

1-28 + Heparin 0.336 52

1-28 + Heparin + 13-17 0.503 78

Sample - 15μM Heparin Avg. 0.D. % Soluble

1-28 0.807 100

1-28 + Heparin 0.086 11

1-28 + Heparin + 13-17 0.395 49

Inhibition of Aβ(1-28) binding to heparin was also tested with an Aβ(10-17) peptide (Fig. 9 and Table 3) . The results of this assay indicate that A/3(10-17) does not compete with Aβ(1-28) for heparin binding at pH 4, although Aβ(13-17) clearly does. It is possible that differences in the charge density of the A/3(10-17) octapeptide are responsible for its inability to compete with Aβ(1-28) for heparin binding.

Table 3

Competition of A/3(10-17) With Aβ(1-28) For Heparin Binding

Sample Avg. P.P. % Soluble 1-28 4.028 100 1-28 + Heparin 0.099 2

1-28 + Heparin + 10-17 0.087 2

Competition of Aβ(13-17) with Aβ(1-28) for binding the proteoglycan chondroitin sulfate (CSPG) was tested at pH 4.0 (Fig. 10; Table 4). A/3(l-28) aggregated with CSPG. The aggregation was inhibited by A/3(13-17), thus indicating that A/3(13-17) competes with Aβ(1-28) for CSPG binding.

Table 4

Competition of A/3(13-17) With A/3(1-28) For CSPG Binding

Sample Avg. O.D. % Soluble

1-28 2.585 100

1-28 + CSPG 0.050 2

1-28 + CSPG + 13-17 0.725 28

To investigate whether the competition for binding to the glycosaminoglycans was dependent solely on ionic interactions, poly-L-lysine (MW 30,000- 70,000) was tested for the ability to compete with Aβ(1-40) for binding to heparin and to CSPG. Poly-L- lysine, which had no effect upon or slightly enhanced Aβ(1-40) solubility, did compete with Aβ(1-40) for binding heparin (Fig. 11) and CSPG (Fig. 12) . These results are summarized in Table 5.

Table 5

Competition of Poly-L-Lysine With Aβ(1-40)

For Binding to Heparin and Chondroitin Sulfate Proteoglycan (CSPG)

Sample¹ Competitor Avg. % Soluble O. D. 0

1. 1-40 only — 0 . 906² 100

2 . 1-40 only 1. 055³ 100

3 . 1-40 + Poly-L-Lys⁴ — 1. 015 112

4 . 1-40 Heparin⁵ 0 . 004 0 5 5. 1-40 + Poly-L-Lys Heparin 0 . 424 47

6. 1-40 CSPG⁶ 0. 021 2

7 . 1-40 + Poly-L-Lys CSPG 0 . 430 41

1 Samples (10 μl, pH 3.5) were incubated 45 min. o The fresh A/3(1-40) was used.

² This value was obtained from the control of the heparin binding assays and was used to normalize to 100% for samples 3-5.

³ This value was obtained from the control of the CSPG binding assays and was used to normalize to ⁵ 100% for samples 6 and 7.

⁴ Poly-L-lysine was present at 6 μg per sample.

⁵ The concentration of heparin was 15 μM.

⁶ CSPG was present at 4 μg per sample. 0

6.3. DISCUSSION

We found that a variety of glycosaminoglycan chains bind to and aggregate A/S(1-28). The 5 interaction of these sulfated polysaccharides with

A/3(1-28) increases as the pH value is lowered below 6- 7, and there is essentially no association of the peptide with the glycosaminoglycans at pH 8. These data are consistent with the involvement of one or more positively charged histidine residues in the binding of the peptide to the sulfated sugar moieties, and suggest that the interaction is primarily ionic in nature. This latter interpretation is supported by the observation that the anionic polymer, dextran sulfate, can bind the amyloid peptide whereas the non- charged polysaccharide, dextran, cannot.

In addition to the ability of Aβ(1-28) to bind various sulfated polysaccharides, we have shown that A/3(1-40) also associates with glycosaminoglycans. In contrast, a peptide containing residues 25-35 of the amyloid peptide does not show appreciable association with heparin. We propose that the glycosaminoglycan binding site is found at residues 12-17 of the amyloid peptide (also see Fraser et al., 1992, J. Neurochem. 59:1531-1540; and Kisilevsky, 1989, Neurobiol. Aging 10:499-500). This region of the peptide contains the consensus heparin binding sequence (Cardin and Weintraub, 1989, Arteriosclerosis 9:21-32) Val-His- His-Gln-Lys-Leu, and the ionization state of one or both of the tandem histidines could modulate glycosaminoglycan binding. Supporting this conclusion is our observation that a pentapeptide consisting of the A/3(13-17) sequence binds to a heparin affinity column at pH 4.0, but does not show appreciable binding at pH 8.0.

The aggregated forms of A/3(1-28) and A/3(1-40) that are observed after glycosaminoglycan binding are presumed to be similar to those described by Fraser et al. (1992, J. Neurochem. 59:1531-1540), who demonstrated that S0₄ ^"2 (5-50 mM) or heparin can cause the lateral association of pre-formed fibrils of A/3(ll-28), Aj8(13-28) and A_/8(ll-25). This group noted a decreased peptide aggregation as the pH was raised to 8.0 and no aggregation was seen when A/3(15-28) was examined. These observations are consistent with the requirement of protonated histidines 13 and/or 14 in the S0₄ ⁺² interaction.

Although the present invention is not limited by any particular theory or hypothesis, one possibility is that the core protein of HSPG (and perhaps other proteoglycans) initially associates with A/3, bringing a glycosaminoglycan chain in close apposition to residues 12-17 of the peptide. This could then alter the microenvironment around the histidine residues at positions 13 and 14, causing an elevation of the apparent pK's. Alternatively, Yates et al. (1990, J. Neurochem. 55:1624-1630) have reported decrease of pH in the AD brain, which may be sufficient in magnitude to cause an increased association of amyloid with glycosaminoglycan chains.

Recent evidence suggests that A/3 is formed within lysosomes after internalization of non-cleaved APP (Estus et al., 1992, Science 255:726-728; Haass et al., 1992, Nature 357:500-503). Although the invention is not limited by any particular theory or hypothesis, an intriguing model can be developed based on this information and the observed pH-dependent glycosaminoglycan association with the glycosaminoglycan-binding domain identified in the assays disclosed here. In this speculative scenario, proteoglycan would bind APP at the cell surface through its core protein, and some fraction of the APP with associated proteoglycan would escape secretase cleavage and be internalized into the endosomal/lysosomal pathway. Upon encountering the acidic pH of these internal organelles, a glycosaminoglycan chain of the proteoglycan would bind tightly to the His-His-Gln-Lys domain of APP, thereby shielding the nonamyloidogenic "secretase" cleavage site at lys₁₆. Subsequent lysosomal degradation of APP, which would normally cleave the protein into non¬ amyloidogenic fragments, would be altered in such a way as to allow the formation of Aβ. This model would predict that an increase in proteoglycan content within the AD brain may result in an increased production of AjS.

In summary, glycosaminoglycans bind to Aβ(13-17) at pH values below neutrality. The sulfated polysaccharides appear to bind to a consensus glycosaminoglycan-binding domain at residues 12-17 of the peptide, and one or both of two tandem histidines are likely to confer the pH dependence of this interaction. By inhibiting the interaction of A/3 with glycosaminoglycans, e .g. , by using the 13-17 peptide, amyloidogenesis may be prevented or even reversed.

7. EXAMPLE: AN A/3 TETRAPEPTIDE AND A

PENTAPEPTIDE AND A HEPARIN DISACCHARIDE REVERSE THE GLYCOSAMINOGLYCAN MEDIATED RESISTANCE OF Aβ PEPTIDE TO DEGRADATION The following Example provides direct evidence that AjS peptide resists proteolytic degradation when aggregated with a proteoglycan, i . e . , that proteoglycans protect aged amyloid from degeneration. Furthermore, molecules that competitively inhibit binding of Aβ peptide to glycosaminoglycans were able to reverse this resistance of Aβ to proteolytic degradation. These competitive inhibitors are Aβ(13- 16), A/3(20-24) and a heparin-derived disaccharide, α- 4-deoxy-L-threo-hex-4-enopyrano-syluronic acid- [l->4] D-glucosamine-N-sulfate-6-sulfate. 7.1. MATERIALS AND METHODS 7.1.1 REAGENTS AjS(1-40) was obtained from Bachem, Inc. Aged A/3(1-40) was solubilized 2-3 months prior to use. Fresh Aβ(1-40) was used within 5-15 days after solubilization. Chondroitin sulfate proteoglycan (CSPG) , MW 2000 kDa based on column chromatography, was purified from bovine nasal septum by dissociative guanidine HC1 extraction. Papain, dextran, the heparin-derived disaccharide and other chemicals were obtained from Sigma. Aβ(20-24) and Aβ(20- 24) :Phe₂₀->Asp₂₀) were prepared by solid phase peptide synthesis as described in Section 6.1.4, supra .

7.1.2 PROTEOLYSIS RESISTANCE ASSAY

The ability of Aβ to resist proteolysis was evaluated by testing the amount remaining after incubation with polysaccharide followed by treatment with papain. Reaction mixtures consisting of combinations of AjS(l-40), CSPG, dextran, and papain were prepared in 100 mM Tris, pH 6.0-7.4. The reaction mixture was incubated 45 min. Samples were incubated an additional 45 min with addition of the inhibitory molecule. Samples were then treated with papain for 8 - 18 hours. To the reaction mixture was added 2X Tris-Tricine sample buffer (8% SDS, 24% glycerol, 0.1 M Tris base, 0.1 M Tricine, 0.05% Bromophenol Blue). Aliquots were loaded on 16.5% Tris-Tricine gel. The separating gel consisted of 16.5% acrylamide, 1 M Tris base, 0.1% SDS, 13.3% glycerol, pH 8.45. The stacking gel consisted of 4% acrylamide, 1 M Tris base, 0.1% SDS, pH 8.45. After electrophoresis; gels were stained with 0.2% Coomassie brilliant blue in 45% methanol, 10% acetic acid. Gels were scanned and Aβ bands were quantified using densitometry.

7.2. RESULTS

7.2.1 CSPG-MEDIATED PROTECTION

OF Aβ PEPTIDES FROM PROTEOLYSIS

To determine whether proteoglycans protect AjS from proteolysis, aged A/S(1-40) was incubated with chondroitin sulfate proteoglycan (CSPG) . The reaction mixture was then treated with the proteolytic enzyme papain. Control samples were run in the absence of any additives or without CSPG. The results of this experiment are shown in Fig. 13. The amount of A/3(l- 40) in the sample treated with papain was greatly reduced compared to the control. When incubated with CSPG prior to papain treatment, the amount of A/3(1-40) recovered after proteolysis was greater than the amount recovered in the absence of CSPG. This result indicated that CSPG protects aged A/S(1-40) at pH 6.0 from proteolysis.

To evaluate whether the extent of protection of AjS(1-40) from proteolysis corresponds to the amount of CSPG, samples of A/3(1-40) (21.5 μg) incubated with varying amounts of CSPG were treated with papain (2 μg) for 12 hours. The results are shown in Fig. 14. These results suggest a linear relationship between the amount of CSPG present in the sample and the degree of protection of the A/3(1-40) from proteolysis. Based on these data, the approximate amount of CSPG required to completely protect the Aβ from proteolysis was calculated as 145 μg. 7.2.2 COMPETITIVE INHIBITORS OF A/S BINDING TO CSPG REVERSE CSPG-MEDIATED PROTECTION OF Aβ FROM PROTEOLYSIS To evaluate whether the small molecules, such as peptides that inhibit the binding of A/3 to proteoglycans, can reverse the proteoglycan/glycosaminoglycan-mediated protection of

AjS from proteolysis, potential inhibitors were added after incubation of the A/3 with CSPG, and prior to the addition of papain. The three potential inhibitors of 19 week aged A/3(1-40)- proteoglycan/glycosaminoglycan binding that were tested were the peptide representing the heparin sulfate proteoglycan binding site of A/3 (two peptides, A/3(13-17) and Aj8(13-16), were tested), the peptide Glu-Ala-Glu-Asp-Val, which corresponds to AjS(20-24:Phe₂₀ → Glu₂₀) , and a heparin disaccharide - dUA-2s-[l->4]-GlcNS-6S) . Incubation for 45 min with any of the three inhibitor molecules was able to reverse the CSPG (30 μg) mediated protection of A/3(l- 40) (4.3 μg) from papain (4 μg) digestion for 18 hours at 37°C (Figs. 15, 16 and 17).

For the results depicted in Fig. 15, the materials, procedure, and conclusion were as follows: Materials:

Aθ(l-40) : MW 4329, LOT #ZJ209 from Bachem, Inc. CSPG: High molecular weight CSPG was sonicated for 14 hours to generate low molecular weight material. Peptides: A/S(13-16) and (13-17) were synthesized. Papain: MW 23,000 obtained from Sigma Chemical Company Procedure:

1. Following were made in 100 mM Tris, pH 6.5:

B. A/3(4.3 μg)+Papain(3 μg)

C. A/3(4.3 μg)+CSPG(30 μg)+Papain(3 μg) D. A/3 (4.3 μg) +CSPG (30 μg) +A/S (13 -17) +Papain (3 μg)

E. Aj8 (4. 3 μg) +CSPG (30 μg) +A/3 ( 13 -16) +Papain (3 μg)

2. AjS and CSPG were incubated together for 1 hour, following which 120 μg of AjS(13-16) and (13-17) were added and the incubation repeated for 1 hour at room temperature. 3. Papain (3 μg) was added and the samples were incubated for 18 hours at 37° C.

4. To the reaction mixture was added 2X Tris- Tricine sample buffer (8% SDS, 24% glycerol, 0.1 M Tris base, 0.1 M Tricine, 0.05% Bromophenol blue) and samples were subjected to 16.5% Tris-Tricine SDS- polyacrylamide gel electrophoresis:

Separating gel: 16.5% acrylamide, 1M Tris base,

0.1% SDS, 13.3% glycerol, pH 8.45)

Stacking gel: 4% acrylamide, 1 M Tris base, 0.1% SDS, pH 8.45

Gel staining: Coomassie brilliant blue (0.2% in

45% methanol and 10% acetic acid)

5. Gels were scanned and AjS bands were quantified using a densitometer. Conclusion

Aβ(13-16) and Aβ(13-17) disrupt A/3-CSPG association and cause 70% increase in Aβ proteolysis. These peptides bind proteoglycans.

The A/3(20-24: Phe₂₀→Glu₂₀) pentapeptide was synthesized based on the sequence of amyloid (residues 20-24) that are believed to bond naturally in an anti- parallel jS-strand to the glycosaminoglycan-binding domain of Aβ (residues 13-17) . The naturally occurring 20-24 sequence is Phe-Ala-Glu-Asp-Val, and Glu₂₂ and Asp₂₃ are likely to form stabilizing ion-pairs with His₁₃ and His₁₄ of the adjacent anti-parallel strand. Substituting a negatively-charged glutamic acid for the phenylalanine at residue 20 (Phe₂₀ → Glu₂₀) , to allow the negative carboxyl of the aspartic acid to ion-pair with the positively charged Lys₁₆, was tested to enhance binding of the pentapeptide to the glycosaminoglycan binding domain of Aβ . This should stabilize the interaction of the pentapeptide with the glycosaminoglycan-binding region at residues 13-17 as well as reduce the overall positive charge density of the 13-17 domain. The binding of the pentapeptide to residues 13-17 of existing amyloid fibrils might thus shield the positively charged. amino acid residues (His₁₃, His₁₄, and Lys₁₆) from negatively charged glycosaminoglycan moieties, thereby reducing proteoglycan binding.

As shown in Fig. 16, the peptide A/3 (20-24:Phe₂₀ →Glu₂₀) can reverse the CSPG-mediated protection of Aβ from papain proteolysis.

7.3. DISCUSSION This example demonstrates the evaluation of potential inhibitors of Aβ binding to glycosaminoglycans by testing ability to reverse chondroitin sulfate proteoglycan (CSPG)-mediated protection of amyloid fibrils to proteolytic degradation. Normally, the non-specific protease papain will degrade fibrillar A/3, but prior addition of proteoglycan (e.g. CSPG or heparan sulfate proteoglycan) to the amyloid results in an inhibition of degradation. This is of significance in Alzheimer's disease, since the association of proteoglycans with amyloid within senile plaques may make the resulting structure resistant to breakdown in the brain. Addition of Aj8(13-16) or A/3(13-17), which contains the heparin binding domain of Aβ, reversed the CSPG-mediated protection of aged Aβ(1-40). This result is interesting since AjS(13-17) could inhibit aggregation of A/3(1-28) but not Aβ(1-40) with heparin or CSPG (see Section 6, supra) . Since reversal of the proteolysis protection mediated by glycosaminoglycans may be more meaningful therapeutically, the inability of AjS(13-17) to inhibit aggregation of A/5(1-40) with heparin is less important.

Addition of the pentapeptide A/3(20-24:Phe₂₀-Glu₂₀) to Aβ(1-40)-CSPG mixtures reversed the proteoglycan- mediated protection against proteolysis. Thus, A/S(20- 24 Phe₂₀ → Glύ₂₀) appears to be effective at reducing proteoglycan-amyloid interaction. This compound may be of greater therapeutic value than the A/3(13-17) or Aβ(13-16) compounds, since it should bind _/SAP and not the glycosaminoglycan chains of proteoglycans. This could make the pentapeptide a more specific drug.

In addition to the A/3(20-24) pentapeptide, the ability of a disaccharide derived from heparin to interfere with jSAP-proteoglycan binding has been examined. This disaccharide is commercially available, and has the following structure: α-4-deoxy-L-threo-hex-4-enopyranosyluronic acid- [1 → 4] D-glucosamine-N-sulfate-6-sulfate

This molecule is similar to the A/3(20-24) peptide in that this disaccharide is derived from the glycosaminoglycan chains that bind residues 13-17 of amyloid. Thus, this small sugar (molecular weight = 665) competes with glycosaminoglycan for binding to amyloid fibrils. As shown above, it too reverses the proteoglycan-mediated protection against proteolysis. 8. EXAMPLE: INHIBITION OF A/3-MEDIATED COMPLEMENT ACTIVATION

While the normal activators of complement are immunoglobulins, the evidence supporting the existence of immunoglobulins in the AD brain is equivocal. This would suggest that a substance within the AD brain might bind to Clq and initiate the complement cascade.

The present example confirms the observation by Rogers et al. (1992, Proc. Natl. Acad. Sci. USA 89:10016-10020; 1992, Res. Immunol. 143(6) :624-630) that the /3-amyloid peptide (A/3) is capable of specifically activating the complement cascade. This example also shows that a tetrapeptide corresponding to A/3(13-16) and a pentapeptide derived from A_/S(20-24) are effective inhibitors of AjS-mediated complement activation.

8.1. MATERIALS AND METHODS 8.1.1 COMPLEMENT FIXATION ASSAY A variation of the complement fixation assay described by Palmer and Whaley (1986, in Manual of Clin . Lab . Microbiol . , 3rd Ed., Rose et al. (Eds.), Am. Soc. Microbiol., pp 57-66) was utilized to assay for amyloid initiation of complement activation. Amyloid peptides were incubated overnight at 4°C in 0.2 ml of barbital buffer (pH 7.4) containing 80 μl of diluted human serum (1:100). During this incubation, peptide-mediated activation of complement results in the consumption of complement components. As a control, parallel overnight incubations of human serum were performed in the absence of added peptides. Subsequently, an aliquot (0.2 ml) of sensitized-sheep red blood cells (s-SRBC) and 0.1 ml of buffer were added to each tube and the suspension was incubated for 1 h in a 37°C water bath. Any complement remaining from the initial overnight incubation with peptide causes lysis of the s-SRBCs, releasing hemoglobin into the medium. The suspension was centrifuged for 5 minutes at 300 x g, and the extent of complement-mediated lysis of the s-SRBCs was determined by reading the optical density of the supernatant at 410 nM. Complement activity was expressed as a percent of control samples which did not receive peptide during the initial incubation.

A solid-phase binding assay was utilized for the evaluation of Clq binding to amyloid peptide. This assay, which is similar to that recently used by Rogers et al. (1992, Proc. Natl. Acad. Sci. USA 89:10016-10020), employs Aβ which is immobilized on membranes, followed by incubation with human Clq. Briefly, A/3 is dotted onto pre-soaked PVDF membranes. The membranes are then rinsed in Tris-buffered saline pH 7.4 (TBS), followed by blocking of the membrane with 5% dry milk in TBS. After an additional rinsing in TBS, the membrane is incubated in a solution of Clq (Quidel; 10 μg/ml in TBS + 5% dry milk) for 2 h. Non- bound Clq is removed by rinsing in TBS, and bound Clq is determined by overnight incubation with rabbit anti-human Clq antibody (Quidel; 1:1000 dilution).

After rinsing, the Clq antibody is visualized by a 2 h incubation with biotinylated goat anti-rabbit antibody (Quidel; 1:50 dilution) followed by color development with a Vectastain ABC Elite kit (peroxidase type) . The amount of bound Clq was quantified through densitometric analysis of the stained membranes (PDI model DNA35 densitometer) .

The peptides used in these studies were A/3(1-40) (Bachem, Inc.), A/3(l-28) (Bachem, Inc.), or scrambled AjS(l-40) (custom synthesized by Biosynthesis, Inc.). The scrambled peptide contains the same amino acid composition as Aβ(1-40), with its sequence randomly assigned. Amyloid peptides were examined for activity as a function of their age in solution. In such experiments, peptides are referred to as "fresh" if they have been in aqueous solution for 1-3 days, and as "aged" if they have been in solution >25 days at 4°C. The "aged" Aβ is aggregated and of high molecular weight as judged by the sedimentation of the peptide after centrifugation at 100,000 x g for 30 min. Moreover, the aged amyloid appears to exist as jS-fibrils as evidenced by its Congo Red birefringence. Ultracentrifugation reveals that the freshly prepared A/S(1-40) solutions are not appreciably aggregated. The actual ages of the "aged" peptide solutions used in each experiment are indicated in the results. The A/3(13-16) and A/5(20-24) peptides were prepared by solid phase synthesis using Fmoc chemistry.

8.2. RESULTS

The major component of the AD senile plaque, jS-amyloid peptide (AjS) , is capable of specifically triggering the complement cascade in vitro (Fig. 18) .

This activation is not initiated by control peptides, suggesting that the neuronal and axonal damage seen in

AD is likely to be due in part to the formation of MAC in response to /3-amyloid peptide.

The knowledge that A/5 is capable of initiating the classical complement cascade provides opportunities for specific interventions that would ameliorate complement-mediated cell damage in AD.

Utilizing a solid-phase binding assay in which amyloid peptides are immobilized onto membranes and allowed to incubate with solutions containing Clq, a specific binding site residing within the first 28 residues of the 39-42 amino acid A/3 peptide was found (Fig . 19 ) . Based on the knowledge that the Clq binding site on immunoglobulins appears to be highly charged, and residues 13-16 of A/3 are surface exposed and bind glycosaminoglycans, it was postulated that the motif of AjS may be responsible for Clq binding. To test this concept, a competition experiment in which a synthetic tetrapeptide consisting of residues 13-16 of AjS was incubated at various concentrations with Clq for 1 h prior to the addition of the Clq solution to immobilized AjS(1-28) was performed. While Clq in the absence of the tetrapeptide showed appreciable binding to A/3(1-28), the association between Clq and Aβ(1-28) was completely inhibited by the tetrapeptide (Fig. 19). This suggests that A/3(13-16) binds to sites on Clq and blocks subsequent interaction with AjS(1-28), and thus implies that a Clq binding motif resides at residues 13-16 of the intact amyloid peptide.

The ability of the tetrapeptide comprising residues 13-16 of AjS to effectively block AjS activation of the entire complement cascade is seen in Fig. 20. Utilizing the complement fixation assay, we find that addition of a 50-fold molar excess of Aβ(13- 16) results in approximately 60% inhibition of A/3(1-40)-induced complement activation. Of particular significance is the finding that AjS(13-16) does not inhibit the activation of complement by aggregated human immunoglobulin, indicating that amyloid must bind to a different region of Clq than immunoglobulin. This is an important observation, since immunoglobulin is the normal activator of complement and compounds based on A/S(13-16) should not interfere with immunoglobulin-mediated complement activation throughout the body. In addition ti the A/3(13-17) peptide, a second peptide effective to inhibit A/3-induced complement activation has been identified. Based on modeling of the structure assumed by A/5 in solution, residues 20- 23 of the amyloid peptide are predicted to normally hydrogen-bond or bond ionically with residues 13-16 of amyloid peptide in an anti-parallel jS-sheet:

13 14 15 16 His—His—Gin—Lys

+ + +

Asp—Glu—Ala—Phe

23 22 21 20

Residues 22 and 23 have negatively charged amino acids that neutralize the positive charge density around residues 13 and 14. To develop a peptide with the ability to bind specifically to residues 13-16 of A/3 and reduce the charge density of that motif, the pentapeptide Glu-Ala-Glu-Asp-Val was prepared. This peptide corresponds to AjS(20-24), with the naturally occurring Phe of residue 20 substituted with a negatively charged glutamic acid. This substitution was designed to allow ion-pairing between the glutamic acid and Lys₁₆ of the full-length Aβ:

12 13 14 15 16 Val—His—His—Gin—Lys

+ + + - Val—Asp—Glu—Ala—Glu

24 23 22 21 20

When the pentapeptide was added to the complement fixation assay, it blocked A/S(1-40)-induced complement activation in a dose-dependent fashion (Fig. 21) . This suggests that the altered AjS(20-24) associates with the Clq binding motif at residues 13-16 of A/3, thereby blocking the accessibility of Clq to this domain.

Data from the Clq binding assay suggest that full-length Aβ contains a second Clq binding region within residues 25-35 (Fig. 22) . Since complement activation requires that two of the six globular heads of Clq are occupied at once, it may be that Clq must bind both the 13-16 and 25-35 regions of A/3 for effective initiation of the cascade. The two classes of potential therapeutic compounds mentioned above would block binding to the 13-16 motif, and therefore disrupt one of the two required binding sites. It is possible that compounds that either resemble the binding site within residues 25-35 or interact with this site would effectively inhibit A/3-induced activation of the complement cascade.

8.3. DISCUSSION The results presented here confirm the observation that AjS in amyloid plaques can activate complement. More importantly, these results show that small peptides can inhibit A/3-mediated complement activation. Both A/3(13-16), a highly cationic tetrapeptide that is expected to represent a Clq binding site of AjS, and A/3(20-24), the putative binding site of AjS(13-16) segment with the full-length A/3 peptide, inhibit A/S-mediated complement activation. As demonstrated in the preceding Examples, both molecules are also effective in reversing glycosaminoglycan-mediated protection of Aβ from proteolysis. Thus these peptides, and their derivatives, are attractive therapeutic agents to reverse the formation of amyloid plaques and reduce the activation of complement within the brains of AD patients. 5 Based on these results, disaccharides such as heparin disaccharide are expected to inhibit A/3-mediated complement activation as well.

In conclusion, these results indicate that the following groups of molecules can serve as therapeutic 0 agents to reduce the activation of complement in AD brain, and thus diminish the neurological symptoms and pathologies of the disease:

1) Specific small peptides or peptide mimetics based on Clq binding motif within residues 13-16 or 5 residues 25-35 of jS-amyloid peptide. These agents associate with Clq in such a way as to block the binding of /3-amyloid;

2) Specific small peptides or peptide mimetics that will associate with the Clq binding motif within o residues 13-16 or residues 25-35 of /3-amyloid peptide. These agents should associate with A/3 in such a way as to block the binding of Clq; or

3) Known inhibitors of MAC attachment to membranes, such as vitronectin, clusterin, or 5 protectin or peptides based on sequences from these proteins, could be administered to AD patients. These agents would reduce C5-C9 (MAC) attachment to neuronal membranes, and hence would reduce the "bystander" damage that leads to dystrophic neurites. 0

The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the invention in addition to those described herein will become 5 apparent to those skilled in the art from the foregoing description and the accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

Various publications are cited herein, the disclosures of which are incorporated by reference in their entireties.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: Brunden, K. et al

(ii) TITLE OF INVENTION: Inhibition of Beta Amyloid Binding to Glycosaminoglycans for Treatment of Alzheimer's Disease

(iii) NUMBER OF SEQUENCES: 7

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Pennie & Edmonds

(B) STREET: 1155 Avenue of the Americas

(C) CITY: New York

(D) STATE: New York

(E) COUNTRY: U.S.A.

(F) ZIP: 10036-2711

(V) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: To be assigned

(B) FILING DATE: On even date

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Misrock, S. Leslie

(B) REGISTRATION NUMBER: 18,872

(C) REFERENCE/DOCKET NUMBER: 6739-042

( ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE : 212 790-9090

(B) TELEFAX: 212 869-8864/9741

(C) TELEX: 66141 PENNIE

( 2 ) INFORMATION FOR SEQ ID NO: l :

( i ) SEQUENCE CHARACTERISTICS :

(A) LENGTH: 42 amino acids

(B) TYPE : amino acid

(C) STRANDEDNESS : single

(D ) TOPOLOGY : unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

Asp Ala Glu Phe Arg His Asp Ser Gly Tyr Glu Val His His Gin Lys 1 5 10 15

Leu Val Phe Phe Ala Glu Asp Val Gly Ser Aεn Lys Gly Ala lie lie 20 25 30

Gly Leu Met Val Gly Gly Val Val lie Ala 35 40

(2) INFORMATION FOR SEQ ID NO:2: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 4 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Lys Lys Lys Lys

1

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 4 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3.

His His His His

1

(2) INFORMATION FOR SEQ ID NO: :

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 4 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

Asp Ala Glu Asp

1

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 4 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID Nθ:5:

Glu Ala Glu Asp

1

(2) INFORMATION FOR SEQ ID NO:6: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 5 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:

Glu Ala Glu Asp Val 1 5

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 5 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: unknown

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7i

Asp Ala Glu Asp Val 1 5

Claims

WHAT IS CLAIMED IS:

1. A pharmaceutical composition comprising a peptide, in which the amino acid sequence of said peptide consists of X-X-N-X, in which X is a amino acid with a cationic side chain and N is a neutral amino acid; and a pharmaceutically acceptable carrier.

2. The pharmaceutical composition of claim 1 in which X is selected from the group consisting of histidine, lysine and arginine.

3. The pharmaceutical composition of claim 1 in which N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

4. The pharmaceutical composition of claim 1 in which the amino acid sequence consists of histidine- histidine-glutamine-lysine (a portion of SEQ ID NO:l).

5. A pharmaceutical composition comprising a peptide, in which the amino acid sequence of said peptide consists of X-X-N-X-Z, in which X is a amino acid with a cationic side chain, and N and Z are each independently a neutral amino acid; and a pharmaceutically acceptable carrier.

6. The pharmaceutical composition of claim 5 in which X is selected from the group consisting of histidine, lysine and arginine.

7. The pharmaceutical composition of claim 5 in which N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

8. The pharmaceutical composition of claim 5 in which Z is selected from the group consisting of glycine, alanine, valine, leucine, isoleucine, serine, threonine, asparagine, glutamine, tryptophane, tyrosine, phenylalanine, methionine and cysteine.

9. The pharmaceutical composition of claim 5 in which the amino acid sequence consists of histidine- histidine-glutamine-lysine-leucine (a portion of SEQ ID NO:l) .

10. A pharmaceutical composition comprising a peptide, in which the amino acid sequence of said peptide consists of X_!-N-X₂-X₃, in which at least two of ^χι, X₂/ ^and X₃ ^are independently an amino acid with an anionic side chain and the third X is an amino acid with an anionic side chain or a neutral amino acid and N is independently a neutral amino acid; and a pharmaceutically acceptable carrier.

11. The pharmaceutical composition of claim 10 in which the amino acid with the anionic side chain is selected from the group consisting of aspartic acid and glutamic acid.

12. The pharmaceutical composition of claim 10 in which the non-anionic X and N are independently selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

13. The pharmaceutical composition of claim 10 in which the amino acid sequence consists of phenylalanine-alanine-glutamic acid-aspartic acid (a portion of SEQ ID NO:l).

14. The pharmaceutical composition of claim 10 in which the amino acid sequence consists of aspartic acid-alanine-glutamic acid-aspartic acid (SEQ ID NO:4) .

15. The pharmaceutical composition of claim 10 in which the amino acid sequence consists of glutamic acid-alanine-glutamic acid-aspartic acid (SEQ ID NO:5) .

16. A method for treating Alzheimer's disease in a subject comprising administering a therapeutically effective amount of a peptide, in which the amino acid sequence of said peptide consists of X-X-N-X, in which X is a amino acid with a cationic side chain and N is a neutral amino acid, to a subject having Alzheimer's disease.

17. The method according to claim 16 in which X is selected from the group consisting of histidine, lysine and arginine.

18. The method according to claim 16 in which N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

19. The method according to claim 16 in which the amino acid sequence consists of histidine- histidine-glutamine-lysine (a portion of SEQ ID NO:l).

20. A method for treating Alzheimer's disease in a subject comprising administering a therapeutically effective amount of a peptide, in which the amino acid sequence of said peptide consists of X-X-N-X-Z, in which X is a amino acid with a cationic side chain, and N and Z are each independently a neutral amino acid, to a subject having Alzheimer's disease.

21. The method according to claim 20 in which X is selected from the group consisting of histidine, lysine and arginine.

22. The method according to claim 20 in which N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

23. The method according to claim 20 in which Z is selected from the group consisting of glycine, alanine, valine, leucine, isoleucine, serine, threonine, asparagine, glutamine, tryptophane, tyrosine, phenylalanine, methionine and cystine.

24. The method according to claim 20 in which the amino acid sequence consists of histidine- histidine-glutamine-lysine-leucine (a portion of SEQ ID NO:l) .

25. A method for treating Alzheimer's disease in a subject comprising administering a therapeutically effective amount of a peptide, in which the amino acid sequence of said peptide consists of X_J-N-X_J-X^ in which at least two of X_{l t} X₂, and X₃ are independently an amino acid with an anionic side chain and the third X is an amino acid with an anionic side chain or a neutral amino acid, and N is independently a neutral amino acid, to a subject having Alzheimer's disease.

26. The method according to claim 25 in which the amino acid with the anionic side chain is selected from the group consisting of aspartic acid and glutamic acid.

27. The method according to claim 25 in which the non-anionic X and N are independently selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

28. The method according to claim 25 in which the amino acid sequence consists of phenylalanine- alanine-glutamic acid-aspartic acid (a portion of SEQ ID NO:l) .

29. The method according to claim 25 in which the amino acid sequence consists of aspartic acid- alanine-glutamic acid-aspartic acid (SEQ ID NO:4).

30. The method according to claim 25 in which the amino acid sequence consists of glutamic acid- alanine-glutamic acid-aspartic acid (SEQ ID NO:5).

31. A method for treating Alzheimer's disease in a subject comprising administering a therapeutically effective amount of an anionic disaccharide to a subject having Alzheimer's disease.

₅ 32. The method according to claim 31 in which the anionic disaccharide is a heparin-derived disaccharide.

33. The method according to claim 32 in which ₀ the heparin-derived disaccharide is α-4-deoxy-L-threo- hex-4-enopyranosyluronic acid-[l->4] D-glucosamine-N- sulfate-6-sulfate.

34. A peptide, in which the amino acid sequence ₅ of said peptide consists of X-X-N-X, in which X is a amino acid with a cationic side chain and N is a neutral amino acid.

35. The peptide of claim 34 in which X is o selected from the group consisting of histidine, lysine and arginine.

36. The peptide of claim 34 in which N is selected from the group consisting of glycine, ₅ alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

37. The peptide of claim 34 in which the amino ₀ acid sequence consists of histidine-histidine- glutamine-lysine (a portion of SEQ ID NO:l).

38. A peptide, in which the amino acid sequence of said peptide consists of X-X-N-X-Z, in which X is a ₅ amino acid with a cationic side chain, and N and Z are each independently a neutral amino acid.

39. The peptide of claim 38 in which X is selected from the group consisting of histidine, lysine and arginine.

40. The peptide of claim 38 in which N is selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

41. The peptide of claim 38 in which Z is selected from the group consisting of glycine, alanine, valine, leucine, isoleucine, serine, threonine, asparagine, glutamine, tryptophane, tyrosine, phenylalanine, methionine, and cysteine.

42. The peptide of claim 38 in which the amino acid sequence consists of histidine-histidine- glutamine-lysine-leucine (a portion of SEQ ID NO:l).

43. A peptide, in which the amino acid sequence of said peptide consists of X₁-N-X₂-X₃, in which at least two of Xj, X₂, and X₃ are independently an amino acid with an anionic side chain, and the third X is an amino acid with an anionic side chain or a neutral amino acid, and N is independently a neutral amino acid.

44. The peptide of claim 43 in which the amino acid with the anionic side chain is selected from the group consisting of aspartic acid and glutamic acid.

45. The peptide of claim 43 in which the non- anionic X and N are independently selected from the group consisting of glycine, alanine, valine, serine, threonine, asparagine, glutamine, methionine, cysteine, isoleucine, phenylalanine, and leucine.

₅ 46. The peptide of claim 43 in which the amino acid sequence consists of phenylalanine-alanine- glutamic acid-aspartic acid (a portion of SEQ ID NO:l) .

₀ 47. The peptide of claim 43 in which the amino acid sequence consists of aspartic acid-alanine- glutamic acid-aspartic acid (SEQ ID NO:4).

48. The peptide of claim 43 in which the amino ₅ acid sequence consists of glutamic acid-alanine- glutamic acid-aspartic acid (SEQ ID NO:5).

49. A peptide, in which the amino acid sequence of said peptide consists of the sequence shown in o Figure 1 from amino acid numbers 25-35 (a portion of SEQ ID NO:l) .

50. A molecule comprising the amino acid sequence X-X-N-X, in which X is an amino acid with a ₅ cationic side chain and N is a neutral amino acid, said molecule consisting of not greater than 8 amino acid residues.

51. A molecule comprising the amino acid ₀ sequence Xj-N-X₂-X₃, in which at least two of X_x , X₂, and X₃ are independently an amino acid with an anionic side chain and the third X is an amino acid with an anionic side chain or a neutral amino acid and N is independently a neutral amino acid, said molecule ₅ consisting of not greater than 8 amino acid residues.

52. A molecule comprising the sequence histidine-histidine-glutamine-lysine (a portion of SEQ-ID NO. 1) , said molecule consisting of not greater than 8 amino acid residues.

53. A molecule comprising the sequence glutamic acid-alanine-glutamic acid-aspartic acid (SEQ-ID NO. 5) , said molecule consisting of not greater than 8 amino acid residues.

54. A method for treating Alzheimer's disease in a subject comprising administering a therapeutically effective amount of a molecule comprising an anionic disaccharide to a subject having Alzheimer's disease.