Cysteine-containing peptides having antioxidant properties

Abstract

Cysteine containing amphipathic alpha helices of the exchangeable apolipoproteins, as exemplified by apolipoprotein (apo) A-IMilano (R173C) and apoA-IParis, (R151C) were found to exhibit potent antioxidant activety on phospholipid surfaces. The addition of a free thiol, at the hydrophobic/hydrophilic interface of an am phipathic alpha helix of synthetic peptides that mimic HDL-related proteins, imparts a unique anteoxidant activity to these peptides which inhibits lipid peroxidation and protects phospholipids from water-soluble free radical initiators. These peptides can be used as therapeutic agents to combat cardiovascular disease, ischemia, bone disease and other inflammatory related diseases.

Description

BACKGROUND OF THE INVENTION

[0003] 1. Field of the Invention

[0004] This invention generally relates to human lipid metabolism, particularly to HDL-related proteins, their mutations, and peptides designed based on these mutations which have antioxidant properties beneficial in the regulation of cardiovascular disease (CVD), bone diseases and other inflammatory related diseases.

[0005] 2. Description of the Related Art

[0006] Cardiovascular disease (CVD) is the number one cause of death in Western societies and its prevalence is increasing worldwide. One of the strongest predictors of risk is the plasma concentration of high density lipoprotein (HDL) which exhibits an inverse relationship to the risk (Gordon, T., et al., Am. J. Med. 62:707-714, 1997; Wilson, P. W. F. , Am. J. Cardiol. 66:7A-10A, 1990). Despite the strong epidemiological data relating increased plasma HDL to protection against CVD, a number of rare inheritable traits have been described which result in low plasma HDL concentrations but no increase in CVD. These inheritable traits are, in part, attributed to mutations in apolipoproteinA-I, the major protein component of HDL (Assmann, G., et al, Circulation 87:[suppl III]:III-28-III-34, 1993).

[0007] ApolipoproteinA-IMilano and apoA-IParis are examples of natural variants of apoA-I that manifest HDL deficiencies but there is no apparent CVD in affected subjects. See Weisgraber, K. H., et al., J. Clin. Invest. 66:901-907, 1980; Franceschini, G., et al., i J. Clin. Invest. 66:892-900, 1980; Bruckert, E., et al., Atherosclerosis, 128:121-128, 1997. Indeed, a recent clinical study showed that carriers of apoA-IMilano exhibited normal intimal thickness of carotid arteries compared to age- and sex-matched controls; whereas, hypoalphalipoproteinemic individuals showed intimal thickening as judged by B-mode ultrasound (Sitori, C. R., et al., Circulation 103:1949-1954, 2001). Studies utilizing mice and rabbits support clinical studies by demonstrating that injection of recombinant apoA-IMilano protects against atherosclerosis (Shah, P. K., et al., Circulation 97:780-785, 1998; Shah, P. K., et al., Circulation 103:3047-3050, 2001; Ameli, S., et al., Circulation 90:1935-1941, 1994). However, the mechanism(s) by which apoA-IMilano and apoA-IParis exert anti-atherogenic effects are not completely understood.

[0008] All known human carriers of apoA-IMilano and apoA-IParis are heterozygous for R173C and R151C mutations in apoA-I primary sequence, respectively (Weisgraber, K. H., et al., J. Clin. Invest. 66:901-907, 1980; Bruckert, E., et al., Atherosclerosis, 128:121-128, 1997). The introduction of a cysteine residue in a normally cysteine-free apolipoprotein allows for the formation of homodimers and heterodimers with apoA-II. Dimerization of the cysteine variants inhibits HDL maturation via mechanisms related, in part, to impaired activation of lecithin: cholesterol acyltransferase, the enzyme that catalyzes cholesterol esterification on HDL (Franceschini, G., et al., J. Biol Chem. 265:12224-12231, 1990; Calabresi, L., et al., Biochem. Biophys. Res. Comm. 232:345-349, 1997; Daum, U., et al., J. Mol. Med. 77:614-622, 1999). ApoA-IMilano apoA-IParis are rapidly cleared from the plasma compartment in humans thus contributing to the HDL deficiency in vivo (Roma P, et al., J. Clin. Invest. 91:1445-1452, 1993; Perez-Mendez, O., et al., Atherosclerosis 148:317-326, 2000). However, the fractional catabolic rate of apoA-IParis appears to be different from that of apoA-IMilano suggesting that the two cysteine variants may differ in their metabolic behavior. Human carriers of apoA-IMilano and apoA-IParis also exhibit mild hypertriglyceridemia in addition to the HDL deficiency (Bruckert, E., et al., Atherosclerosis, 128:121-128, 1997; Franceschini G., et al., Atherosclerosis 7:426-435, 1987).

[0009] The C-terminal lipid-binding domain of ApoA-IWT consists of a series of helical repeats separated by proline residues. The amphipathic alpha helix (a.a. 167-184) containing R173C is flanked by two amphipathic alpha helices of relatively greater lipid binding affinity. The lipid binding affinity of the helical repeats alternate, but the two end helices of apoA-I exhibit the highest lipid-binding affinity (Palgunachari, M. N., et al., Arterioscler. Thromb. Vasc. Biol. 16:328-338, 1996). The relatively low lipid-binding affinity associated with helix 7, where R173C is located, may allow a high degree of movement of this particular helix on phospholipid surfaces thus maximizing the frequency of collision between the free thiol at position 173 with reactive lipid peroxides. Increased flexibility of helix 7, which is located in the central region of the C-terminal lipid-binding domain, may be optimized in the presence of deoxycholate used in the preparation of the phospholipid micelles.

[0010] The paradox of abnormal lipoprotein metabolism and protection from CVD has led to the suggestion that the cysteine substitution for arginine in the lipid-binding domain of apoA-I may impart a gain-of-function protecting against atherosclerosis. As thiol groups in proteins are strong nucleophiles often participating in electron transfer reactions, we hypothesized that the monomeric forms of apoA-IMilano and apoA-IParis which contain a free thiol, may possess an antioxidant activity distinct from that of apoA-IWT.

[0011] Individuals with these substitutions are known to have low levels of the “good” cholesterol HDL, but yet do not suffer from significantly increased levels of CVD. Oda et al. disclose cysteine substitutions in Apolipoprotein A-I in Biochemistry 40 (2001) 1710-1718 other substitutions are disclosed at Atherosclerosis 128 (1997) 121-128 ; Atherosclerosis 135 (1997) 181-185. Antioxidant action of HDL is discussed at Atherosclerosis 135 (1997) 193-204.

[0012] These cysteine for arginine substitutions in the Apo A-I variant is of special interest in treatment of cardiovascular disease. The dimer of Apolipoprotein apoA-IMilano and the process of producing and purifying the dimer composition have been disclosed by Sirtori et al, in U.S. Pat. No. 5,876,968, which is hereby incorporated by reference. The process described by Sirtori et al. relies on converting any monomer present to a substantially pure form of the dimer form of apoA-IMilano of at least 90% purity.

[0013] Segrest et al., in U.S. Pat. No. 4,643,988, discloses amphipathic peptides that are useful for treatment and prevention of athersclerosis. The Segrest peptides, generally referred to as 18A and 18pA, are based on an idealistic model of an amphipathic alpha helix that possesses a primary amino acid sequence distinct from that of apoA-I. However, the peptides form Class A amphipathic alpha helices with positively charged amino acids at the interface of polar/nonpolar region and negatively charged residues located in the middle of the polar face of the helix.

[0014] Segrest et al. describe the use and properties of 18A and 18pA; the latter representing a series of two 18A peptides linked by a proline residue. The sequence of 18A is as follows: DWLKAFYDKVAEKLKGAF. Various conservative substitutions (for example positively charged lysine residues in place of positively charged arginine residues) that do not change the overall design of the class A amphipathic alpha helix are also claimed. Additional substitutions of D- for L- amino acid isoforms are described as well as replacement of naturally occurring amino acids for synthetic derivatives (i.e. substitutions of alanine for alpha-naphthylalanine). While an amphipathic peptide is disclosed, the peptide does not possess a cysteine residue and thus lacks the antioxidant activity shown to be possessed by apoA-IMilano.

[0015] Garber et al., disclose on U.S. Pat. No. 6,156,727, anti-atherosclerotic peptides and a transgenic mouse model of atherosclerosis. Garber et al. utilize the same peptides as described above by Segrest et al., however Garber et al. created transgenic mice that express the peptides 18A and 37pA, the latter sometimes is referred to as 18A-Pro-18A. Again, these peptide does not possess a cysteine residue and thus lack the antioxidant activity shown to be possessed by apoA-IMilano

[0016] Lees et al., disclose in U.S. Pat. No. 5,955,055, synthetic peptides for artierial imaging at vascular imaging sites, that mimic apolipoprotein B (apoB), apoliprotein A-I or elastin proteins and is hereby incorporated by reference in its entirety. The Lees peptides are derived (mostly) from apoB and elastin/collagen and are not similar to the peptides we now disclose. The following sequence is used as is based on apoB: YRALVDTLKFVTQAEGAL. The sequence derived from apoA-I described by Lees et al. is: YVLDEFREKLNEELEALKQ. There is no exact sequence match to apoA-I, probably because of conservative substitutions, and the peptide is not at all similar to any of the peptides we now disclose.

[0017] Moreover, none of the peptides in the above mentioned patents are based on Apolipoprotein E3 (apoE3) and Apolipoprotein A-V (apoAV). This is because the mechanisms responsible for the antioxidant properties of apoE3 have not been fully defined until now. ApoAV is a new apolipoprotein that has recently been described and very little is known about its function. Thus, peptides based on apoAV provide new avenues for development of therapeutic agents. It is also clear from our studies that the antioxidant properties of apoA-IMilano and its peptide mimetics are specifically directed toward phospholipid surfaces which none of these above-mentioned patented peptides are shown to be directed toward.

BRIEF SUMMARY OF THE INVENTION

[0018] The present invention describes a new series of diagnostic and therapeutic peptides that possess a novel antioxidant activity, such as has been associated with the monomeric forms of apoA-IMilano and apoA-IParis proteins. A critical feature of the present peptides is the placement of a cysteine residue at the polar/nonpolar interface of an amphipathic alpha helix just as in the apoA-IMilano and apoA-IParis cysteine variants. The presence of a cysteine residue at the polar/nonploar interface of the synthetic peptides confers a potent antioxidant activity that is directed toward lipid surfaces effectively blocking oxidation of phospholipid. The water accessibility of the free cysteine residue enables potential interaction with water-soluble antioxidants such as reduced glutathione thereby enhancing the overall capacity of the peptides to inhibit phospholipid oxidation. This indicates that the peptides may be used in combination with other safe and effective therapies to promote beneficial interactions for long-term protection against inflammatory related events. Structural analyses revealed identical placement of a cysteine residue at the polar/nonpolar interface of an amphipathic alpha helix within apoE3 thus defining the mechanism for the antioxidant activity of apoE-III. A similar “motif” in apoAV is also used to create new peptides.

[0019] In the present peptides have also been developed in which the position of the cysteine residue is changed around the face of the amphipathic alpha helix. Such changes in placement of the cysteine residue is predicted to specifically alter the functionality of the peptides in a systematic fashion. For example, the cysteine residue is placed in the middle of the nonploar face of the amphipathic alpha helix to inhibit specific interaction with water-soluble antioxidants such as reduced glutathione. This enables the development of important biological tools to determine whether such interactions are important in protecting against disease thus allowing the identification of new drug targets and providing a basis for rationale drug design. This has led to the creation of a generic amphipathic alpha helix for the development of tailor-made pharmacteuticals of defined functionality including specific antioxidant activity attributed to strategic cysteine placement, LCAT activation properties endowed via arginine clustering at the polar/nonpolar interface, and cholesterol efflux properties obtained by either phenylalanine placement or by combining unique helical segments.

[0020] The present invention comprises peptides possessing anti-oxidant activity and which may be derived from naturally occurring HDL -associated proteins or may be designed de novo according to the principles outlined herein. The peptides of the present invention may be characterized as follows, where the conventional single letter amino acid code letters are used:

SDELRQCLAARLEALKEN	167-R173C-184	SEQ ID NO:1
SDELRQRLAARLEALKEN	Control wild type 167-184	SEQ ID NO:2
SDELCQRLAARLEALKEN	167-R171C-184	SEQ ID NO:3
SDELRCRLAARLEALKEN	167-Q172C-184	SEQ ID NO:4
SDELRQRCAARLEALKEN	167-L174C-184	SEQ ID NO:5
SDELRQRLCARLEALKEN	167-R175C-184	SEQ ID NO:6
SDELRQRLACRLEALKEN	167-A176C-184	SEQ ID NO:7
SDELRQRLAARLEACKEN	167-L181C-184	SEQ ID NO:8
GEEMRDCARAHVDALRTH	145-R151C-162	SEQ ID NO:9
GEEMRDRARAHVDALRTH	Control wild type 145-162	SEQ ID NO:10
GEEMCDRARAHVDALRTH	145-R149C-162	SEQ ID NO:11
GEEMRCRARAHVDALRTH	145-D150C-162	SEQ ID NO:12
GEEMRDRCRAHVDALRTH	145-A152C-162	SEQ ID NO:13
GEEMRDRACAHVDALRTH	145-R153C-162	SEQ ID NO:14
GEEMRDRARACVDALRTH	145-H155C-162	SEQ ID NO:15
PVLESFCVSFLSALEEYT	220-K226C-237	SEQ ID NO:16
PVLESFKVSFLSALEEYT	Control wild type 220-237	SEQ ID NO:17
PVLCSFKVSFLSALEEYT	220-E223C-237	SEQ ID NO:18
PVLECFKVSFLSALEEYT	220-S224C-237	SEQ ID NO:19
PVLESCKVSFLSALEEYT	220-F225C-237	SEQ ID NO:20
PVLESFKCSFLSALEEYT	220-V227C-237	SEQ ID NO:21
PVLESFKVCFLSALEEYT	220-S228C-237	SEQ ID NO:22
PVLESFKVSCLSALEEYT	220-F229C-237	SEQ ID NO:23
PVLESFKVSFCSALEEYT	220-L230C-237	SEQ ID NO:24
PVLESFKVSFLCALEEYT	220-S231C-237	SEQ ID NO:25
PVLESFKVSFLSCLEEYT	220-A232C-237	SEQ ID NO:26
PVLESFKVSFLSALCEYT	220-E234C-237	SEQ ID NO:27
PVLESFKVSFLSALECYT	220-E235C-237	SEQ ID NO:28
PVLESFKVSFLSALEECT	220-Y236C-237	SEQ ID NO:29
PALEDLRQGLL PVLESFCVSFLSALEEYT KKLN	SEQ ID NO:30
PALEDLRQGLL PVLESFKVSFLSALEEYT KKLN	SEQ ID NO:31
LKLCDNWDSVTSTFSKLR	44-L47C-61	SEQ ID NO:32
LKLLDNWDSVTSTFSKLR	Control wild type 44-61	SEQ ID NO:33
LCLLDNWDSVTSTFSKLR	44-K45C-61	SEQ ID NO:34
LKCLDNWDSVTSTFSKLR	44-L46C-61	SEQ ID NO:35
LKLLCNWDSVTSTFSKLR	44-D48C-61	SEQ ID NO:36
LKLLDCWDSVTSTFSKLR	44-N49C-61	SEQ ID NO:37
LKLLDNWDSVTSTFSCLR	44-K59C-61	SEQ ID NO:38
PALEDLRQGLLP LKLCDNWDSVTSTFSKLR	209/44-L47C-61	SEQ ID NO:39
PALEDLRQGLLP LKLLDNWDSVTSTFSKLR	Control 209/44-61	SEQ ID NO:40
PALEDLCQGLLP LKLLDNWDSVTSTFSKLR	209-R215C-220/44-61	SEQ ID NO:41
PALEDLRQGLLP LCLLDNWDSVTSTFSKLR	209/44-K45C-61	SEQ ID NO:42
PALEDLRQGLLP LKCLDNWDSVTSTFSKLR	209/44-L46C-61	SEQ ID NO:43
PALEDLRQGLLP LKLLCNWDSVTSTFSKLR	209/44-D48C-61	SEQ ID NO:44
PALEDLRQGLLP LKLLDCWDSVTSTFSKLR	209/44-N49C-61	SEQ ID NO:45
PALEDLRQGLLP LKLLDNWDSVTSTFSCLR	209/44--K59C-61	SEQ ID NO:46
GADMEDVCGRLVQYRGEV	105-R112C-122	SEQ ID NO:47
GADMEDVRGRLVQYRGEV	Control wild type 105-122	SEQ ID NO:48
GADMEDCRGRLVQYRGEV	105-V111C-122	SEQ ID NO:49
GADMEDVRCRLVQYRGEV	105-G113C-122	SEQ ID NO:50
GADMEDVRGCLVQYRGEV	105-R114C-122	SEQ ID NO:51
ARLSRCVQVLSRKLTLKA	219-G224C-236	SEQ ID NO:52
ARLSRGVQVLSRKLTLKA	Control wild type 219-236	SEQ ID NO:53
ARLCRGVQVLSRKLTLKA	219-S222C-236	SEQ ID NO:54
ARLSCGVQVLSRKLTLKA	219-R223C-236	SEQ ID NO:55
ARLSRGCQVLSRKLTLKA	219-V225C-236	SEQ ID NO:56
ARLSRGVQVLSRKCTLKA	219-L232C-236	SEQ ID NO:57
ARLSRCVQVLSRKLTLKAKALHARIQQNLDQLREEL	219-G224C-254	SEQ ID NO:58
ARLSRGVQVLSRKLTLKAKALHARIQQNLDQLREEL	Control 219-254	SEQ ID NO:59
ARLCRGVQVLSRKLTLKAKALHARIQQNLDQLREEL	219-S222C-254	SEQ ID NO:60
ARLSCGVQVLSRKLTLKAKALHARIQQNLDQLREEL	219-R223C-254	SEQ ID NO:61
ARLSRGCQVLSRKLTLKAKALHARIQQNLDQLREEL	219-V225C-254	SEQ ID NO:62
ARLSRGVQVLSRKCTLKAKALHARIQQNLDQLREEL	219-L232C-254	SEQ ID NO:63
ATLKDSLCQDLNNMNKFLEKLR	51-E58C-72	SEQ ID NO:64
ATLKDSLEQDLNNMNKFLEKLR	Control wild type 51-72	SEQ ID NO:65
ATLCDSLEQDLNNMNKFLEKLR	51-K54C-72	SEQ ID NO:66
ATLKDCLEQDLNNMNKFLEKLR	51-S56C-72	SEQ ID NO:67
ATLKDSCEQDLNNMNKFLEKLR	51-L57C-72	SEQ ID NO:68
ATLKDSLECDLNNMNKFLEKLR	51-Q59C-72	SEQ ID NO:69
ATLKDSLEQCLNNMNKFLEKLR	51-D60C-72	SEQ ID NO:70
ETGDLWVGCHP	SEQ ID NO:71
ETGDLWVGCHPNGMKIFFYDSEN	SEQ ID NO:72
LKSLDFNTLVDNISVDP ETGDLWVGCHPNGMKIFFYDSEN	SEQ ID NO:73
DWLCAFYDKVAEKLKEAF	18A-K4C	SEQ ID NO:74
DWLKAFYDKVAEKLKEAF	18A control	SEQ ID NO:75
DCLKAFYDKVAEKLKEAF	18A-W2C	SEQ ID NO:76
DWCKAFYDKVAEKLKEAF	18A-L3C	SEQ ID NO:77
DWLKCFYDKVAEKLKEAF	18A-A5C	SEQ ID NO:78
DWLKACYDKVAEKLKEAF	18A-F6C	SEQ ID NO:79
DWLKAFYDKCAEKLKEAF	18A-V10C	SEQ ID NO:80
DWLKAFYDKVCEKLKEAF	18A-A11C	SEQ ID NO:81
LEKLNSCLRDRLSALTDTPLEELRDSLRSRLDALRST	SEQ ID NO:82
LEKLNSCLRDRLSALTDT	SEQ ID NO:83
LEELRDSLRSRLDALRST	SEQ ID NO:84

[0021] Preferred peptides are selected from helix 1 (amino acids 44-65), helix 6 (amino acids 145-162) and helix 10 (amino acids 209-238) of apoAI, helix 7 (amino acids 167-184) of apoAI, the helix spanning amino acids 105-122 of apoE3, and amino acids 219-236 of apo AV.

[0022] Furthermore, the present invention comprises peptide homologues of the sequences listed above, designed according to the detailed description provided below. The sequences listed above may be modified up to 80% homology without losing the functionality described herein. Furthermore, the sequences of the present invention may be provided with specific cysteine residues engineered into them. These cysteine residues may be substitutes for the residues that are underlined in the sequences listed above. That is, for example, SEQ ID NO: 2, SDELRQRLAARLEALKEN Control wild type 167-184 has, according to the present invention at least one cysteine residue in place of one of the underlined residues.

[0023] Furthermore, the present invention comprises methods for making an anti-oxidant peptide based on the design principles outlined in detail in the Detailed Description below. These methods include the steps of identifying an amphipathic helix by known methods for predicting secondary structure and hydrophobicity. See Chou, P. Y., & Fasman, G. D., Adv. Enzymol. Relat. Areas Mol. Biol. 47, 45-148, 1978. A Human HDL-associated protein of known amino acid sequence may be used for this purpose. The identification of a helix as amphipathic is carried out using conventional hydrophobicity analyses and helical wheel projections. The alpha helices of the present invention will have between 10 and 100 amino acids, often between 8 and 30 amino acids. Being amphipathic, they will have a hydrophobic side and a hydrophilic side when viewed axially through the helix. As part of the design and synthesis of the present peptides, one may modify at least one residue on the hydrophilic side from the naturally ocurring (wild type) amino acid to a cysteine residue to create a modified helix peptide; and then selecting a modified helix peptide that has at least twice the anti-oxidant activity as the unmodified peptide.

[0024] The Human HDL-associated protein may be selected from the group consisting of apoAI, apoE3, apo AV and paroxonase. Alternatively, synthetic or non-natural amino acids may be used in the present peptides.

[0025] The anti-oxidant activity is measured by the ability of the modified helix peptide to inhibit lipid peroxidation by soybean lipoxygenase. The ability of the modified helix peptide to inhibit lipid peroxidation by xanthine oxidase and to (not) inhibit xanthine/xanthine oxidase mediated reduction of cytochrome C may also be used to characterize the present peptides. These assays are described in detail in the Detailed Description below.

[0026] In general, the present peptides as recited above will provide approximately 50% or more protection against maximum accumulation of lipid peroxides at a concentration of no more than 500 micrograms per mL. They will inhibit oxidation of a lipid or phospholipid alone or with the addition of a water soluble anti-oxidant. The water soluble oxidant may be any known biologically effective anti-oxidant, such as GSH, vitamin C, vitamin E and N-acetyl cysteine (NAC).

[0027] The present peptides may be prepared according to known pharmaceutical technology. They may be administered singly or in combination, and may further be administered in combination with other cardiovascular drugs. They may be conventionally prepared with excipients and stabilizers in sterilized, lyophilized powdered form for injection, or prepared with stabilizers and peptidase inhibitors of oral and gastrointestinal metabolism for oral administration.

BRIEF DESCRIPTION OF THE DRAWINGS

[0028] FIG. 1. Helical Wheel Projections of therapeutic peptide mimetics as related to Apolipoprotein apoA-IParis and apoA-IMilano class A amphipathic alpha helices. Each helix is shown looking through the barrel of the helix. The dotted line represents the hydrophobic/hydrophilic interface which separates the nonpolar from the polar face of the amphipathic helix. The cysteine is labeled with an asterisk to note its proximity to the interface.

[0029] FIG. 2. Helical Wheel Projection of therapeutic peptide mimetic related to Apolipoprotein E3. The helix is shown looking through the barrel of the helix. The dotted line represents the hydrophobic/hydrophilic interface which separates the nonpolar from the polar face of the amphipathic helix. Cysteine112 is labeled with an asterisk to note its proximity to the interface.

[0030] FIG. 3. Helical Wheel Projection of synthetic peptide related to the antioxidant domain of Apolipoprotein A-V. The helix is shown looking through the barrel of the helix. The dotted line represents the hydrophobic/hydrophilic interface which separates the nonpolar from the polar face of the amphipathic helix. Cysteine224 is labeled with an asterisk to note its proximity to the interface.

[0031] FIG. 4. Schematic showing protocol for determining the antioxidant activity of synthetic peptide mimetics which sets criteria for peptides useful in the invention. The peptides are first added to a lipoxygenase assay to observe the rate of lipid peroxidation of micelles (100). The micelle substrate (100) is composed of 1-palmitoyl-linoleoylphosphatidylcholine and dispersed in deoxycholate and Borate (pH=9.0)/saline-EDTA. Second, the peptides are added to an assay having xanthine/xanthine oxidase to observe the rate of lipid peroxidation of micelles (100). Thirdly, the peptides are added to an assay to determine whether the peptides directly quench reactive oxygen species and prevent cytochrome reduction.

[0032] FIG. 5. Graphs showing antioxidant activity of peptide mimetics of the apoA-IMilano peptide (SEQ ID NO: 1 in FIG. 5A) and the apoA-IParis peptide (SEQ ID NO: 9 in FIG. 5B). Phospholipid (PL) micelles were exposed to reactive oxygen species (ROS) generated via xanthine/xanthine oxidase (X/Xo, 20 U/ml) in the absence (squares) or presence of increasing concentration of the synthetic 18-mers: diamonds, circles, triangles and hatched squares denote 100, 200, 300, and 400 μg/mL concentrations. Results show that the peptides exhibit antioxidant activity in dose-dependent manner. Results show that the peptides exhibit antioxidant activity in dose-dependent manner where approximately 50% protection against lipid peroxidation is observed using 200 μg/ml.

[0033] FIG. 6. Graphs showing antioxidant activity of apoA-IMilano peptide 167-R173C-184 inhibits oxidation induced via ROS. In FIG. 6A, PL micelles were exposed to xanthine/xanthine oxidase (X/Xo, 20 U/mI) in the presence of increasing concentrations of a cysteine-free control peptide (167-184, SEQ ID NO: 2) show no difference in rate of oxidation of lipids as compared to no peptides. FIG. 6B shows results using the thiol-containing apoA-IMilano peptide (167-R173C-184, SEQ ID NO: 1) which show increased antioxidation in a dose-dependent manner. Symbols and doses are the same as in FIG. 5 for Panels A and B. FIG. 6C shows reduction of cytochrome C assay. The synthetic peptides failed to protect cytochrome C indicating that the thiol-containing peptide based on apoA-IMilano is unable to directly quench ROS in the aqueous phase. SOD=superoxide dismutase control.

[0034] FIG. 7. Biological activities of synthetic apoA-IMilano peptide 167-R173C-184 (SEQ ID NO: 1) and synthetic apoA-IParis peptide 145-R151C-162 (SEQ ID NO: 9). FIG. 7A: Interaction of peptide 167-R173C-184 with GSH. The combination of GSH plus 167-R173C-184 (triangles) inhibits initial rates of lipoxygenase-mediated lipid peroxidation compared to GSH alone (diamonds) and the apoA-IMilano peptide alone (circles). FIG. 7B: Peptide apoA-IParis 145-R151C-162 (SEQ ID NO: 9) can stimulate LCAT activation while peptide apoA-IMilano 167-R173C-184 (SEQ ID NO: 1) failed. Results are expressed as a percentage of activation obtained with apoA-Iwt. FIG. 7C: Peptides unable to stimulate cholesterol efflux from J774 macrophages.

[0035] FIG. 8. Graphs show antioxidant activity of a cysteine containing peptide related to Helix 10 of apoA-I (220-K226C-237, SEQ ID NO: 16). FIG. 8A shows that the thiol containing peptide based on helix 10 of apoA-I inhibits lipoxygenase mediated oxidation of phospholipid in a dose dependent manner similar to the inhibition obtained in FIG. 8B using the 167-R173C-184 peptide. Symbols and doses are the same as in FIG. 5.

[0036] FIG. 9. Graphs showing antioxidant activity of a synthetic peptide based on helix 3 of apolipoprotein E-3 (105-R112C-122, SEQ ID NO: 47). In FIG. 9A, PL micelles exposed to xanthine/xanthine oxidase (X/Xo, 20 U/ml), in the presence of increasing concentrations of a cysteine-free peptide related to the apolipoprotein E4 (apoE4) isoform (105-122, SEQ ID NO: 48), show no difference in rate of oxidation of lipids from the absence of peptides. FIG. 9B shows results using the thiol-containing peptide (105-Ri12C-122, SEQ ID NO: 47) that increasing the peptide concentration inhibits oxidation of phospholipid in a dose-dependent manner where 50% protection is observed at 200 μg/ml. Symbols and doses are the same as in FIG. 5. FIG. 9C shows reduction of cytochrome C (no phospholopids) with X/Xo (squares); triangles X/Xo plus the apoE4 peptide (SEQ ID NO: 48) (400 μg/ml); circles, X/Xo plus the apoE3 peptide (SEQ ID NO: 47) (400 μg/ml). Note the synthetic peptides failed to protect cytochroine C indicating that the thiol-containing peptide (SEQ ID NO: 47) was unable to directly quench ROS in the aqueous phase. The asterisks denote the control SOD (superoxide dismutase).

[0037] FIG. 10. Representation of the general placement of a thiol-bearing residue in peptide mimetics. FIGS. 10A and 10B show the ideal placement of the cysteine residue in an amphipathic alpha helix peptide is at the interface. FIGS. 10C and 10D show alternate positions of the cysteine to be in the hydrophobic or the hydrophilic face.

[0038] FIG. 11. Representation of the general placement of multiple thiol-bearing residues in synthetic peptides. Cysteine residues can be placed on the same side of the interface as in FIGS. 11A, 11B, and 11D, or on opposite sides of the amphipathic interface as in FIGS. 11C, 11E and 11F.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

Definitions

[0039] The term “peptide” herein is used to describe an amino acid sequence between 2 and 100 amino acids in length, the amino acids being joined by peptide linkages. The amino acids may be naturally and non-naturally occurring.

[0040] The term “antioxidant” herein refers to any compound, composition, peptide or protein that inhibits oxidation of phospholipid. A “potent” antioxidant will inhibit oxidation at an effective concentration (EC50) that produces 50% reduction in phospholipid oxidation.

[0041] The term “amphipathic” herein refers to a domain which has both a hydrophobic and a hydrophilic surface that are identified, e.g., as described in Kaiser and Kezdy (Ann. Rev. Biophys. Biophys. Chem. 16: 561, 1987; Science 223:249, 1984. The term “amphipathic” further means that peptides must exhibit “sidedness” and be amphipathic along the axis through the helix, wherein the majority of the residues on the nonpolar, hydrophobic side of the helix are nonpolar residues, preferably leucine, but may include alanine, valine, isoleucine, proline, phenylalanine, tryptophan and methionine. A majority of the residues on the lipophilic side is preferably made up of hydrophilic residues glycine, serine, threonine, cysteine, tyrosine, asparagines, glutamine, aspartate, glutamate, lysine, arginine and histadine.

[0042] The term “homology” or “homologous” means an amino acid similarity measured by the program, BLAST (Altschul et al (1997), “Gapped BLAST and PSI-BLAST: a new generation of protein database search programs”, Nucleic Acids Res. 25:3389-3402), as found at http://www.ncbi.nlm.nih.gov/blast/Blast.cgi and expressed as—(% identity n/n). In measuring homology between a peptide and a protein of greater size, homology is measured only in the corresponding region; that is, the protein is regarded as only having the same general length as the peptide, allowing for gaps and insertions.

[0043] The terms “derived from” or “based on” mean, regarding a peptide amino acid sequence, having a relationship to a native sequence of an HDL-associated protein.

[0044] The term “substantially identical” is herein used to mean having an amino acid sequence which differs only by conservative amino acid substitutions or by non-conservative amino acid substitutions, deletions, or insertions located at positions which do not destroy the biological activity of the peptide.

[0045] The term “HDL-associated protein” or “HDL-related protein” means a protein and/or apolipoprotein that is naturally associated with High Density Lipoproteins (HDL), derived from either plasma or interstitial fluids that can be isolated within the HDL density interval (i.e. d=1.063-1.25 g/ml fraction) of co-isolates with apoA-I upon immunoaffinity procedures. Moreover, the said apolipoproteins may also be present in lipid-free form and participate in HDL metabolic pathways including the ABCA1 cholesterol efflux pathway which gives rise to HDL particles.

[0046] The term “Class A amphipathic alpha helix” refers to an alpha helix in which one surface of the peptides is composed primarily of hydrophobic amino acids and the other surface hydrophilic amino acids. Class A alpha helices possess positively charged amino acids on the polar surface next to the interface of the hydrophobic domain, and negatively charged residues in the middle of the polar surface.

[0047] The term “Class Y alpha helices” refers to an alpha helices which are amphipathic, exhibiting a broad nonpolar surface and a hydrophilic domain that lacks interfacial positively charged residues.

Introduction

[0048] Cardiovascular disease is the number one cause of death in western societies and the prevalence of this disease is increasing worldwide. One of the strongest predictors of risk is the plasma concentration of high-density lipoprotein (HDL) which exhibits an inverse relationship. Despite the strong epidemiological data relating increased plasma HDL to protection from cardiovascular disease, a number of rare mutations in apolipoproteinA-I, the major protein of HDL, present an HDL deficiency and resistance to cardiovascular disease. Apolipoprotein A-IMilano (apoA-IMilano) and Apolipoprotein A-IParis (apoA-IParis) are rare, naturally occurring Arg→Cys substitutions in apoA-I primary sequence that manifest such a HDL deficiency, but affected subjects do not develop cardiovascular disease. The cysteine mutations enable apolipoprotein dimerization via a disulfide bridge. This dimerization limits HDL particle growth and facilitates the clearance of HDL from the circulation. Indeed, human carriers of apoA-IMilano exhibit a HDL deficiency and mild triglyceridemia. The paradox of HDL deficiency and protection from cardiovascular disease has led to the suggestion that the cysteine substitution for arginine in the lipid-binding domain of apoAl may impart redeeming qualities protecting the artery wall from athermatous lesion formation.

[0049] The inventor reported, for the first time, in Biochemistry 41, 2089-2096 (2002), a unique antiatherogenic function of apoA-IMilano and apoA-IParis related to antioxidant properties on phospholipid surfaces. The results of the studies indicate that apoA-IMilano and apoA-IParis, were potent inhibitors of lipid peroxidation protecting phospholipid surfaces from lipophilic, as well as, water soluble free radical initiators; whereas, apoA-IWT was a relatively poor inhibitor of oxidative events.

[0050] Using purified recombinant apolipoproteins and enzymatic methods of phospholipid peroxidation it was demonstrated that apoA-IMilano (R173C) and apoA-IParis (R151C) exhibited antioxidant activities not associated with wild-type apoA-I. This antioxidant activity was attributed to the monomeric form of apoAlMilano and apoA-Ipans and was found to be dependent on the presence of phospholipid. The latter is based on studies where apoA-IMilano and a synthetic peptide mimetic were unable to prevent superoxide anion induced reduction in Cylochrome C (FIG. 6C).

[0051] This observation has important implications regarding the underlying mechanism by which cysteine containing amphipathic alpha helices in the exchangeable apolipoproteins exert antioxidant activity and protect against inflammatory related disease. Xanthine/xanthine oxidase generates superoxide anion and hydroxyl radicals in the aqueous phase that mediate the oxidation of phospholipid. The fact that apoA-IMilano was unable to prevent xanthine/xanthine oxidase mediated reduction of cytochrome C suggests that in lipid-free form apoA-IMilano was unable to quench reactive oxygen species in lipid-free form. This indicates that apoA-IMilano (and its synthetic peptide mimetics) act on the phospholipid to inhibit the initiation/amphification of lipid peroxidation via a mechanism probably related to chain-breaking antioxidant activity. By extension of these observations, one can infer that antioxidant activity is directed toward lipid surfaces which links the antioxidant activity of apoA-IMilano to the ABCA1 transporter that is required for the lipidation of apolipoproteins in vivo. This provides a basis for drug development in which synthetic peptides can be engineered to possess both the lipidation properties of the native apolipoproteins and the novel antioxidant activity discovered for apoA-IMilano. As a result, it is feasible to specifically target the therapeutics to sites of inflammation and cholesterol deposition where there is an upregulation in the ABCA1 transporter to specifically deliver potent antioxidant activity to sites where its needed most to prevent inflammatory related disease initiation.

[0052] The gene for human Apolipoprotein A-I (apoA-I) had been previously cloned. The entire sequence for the human apoAl protein is found at GenBank Accession No.: P02647 locus APA1_HUMAN. Certain mutations in this gene have been identified at the molecular level, such as Apolipoprotein A-IMilano (A-IParis) (R173C) and Apolipoprotein A-IParis (apoA-IParis) (R151C).

[0053] Using the observation that a free thiol positioned at the hydrophobic/hydrophilic interface of an amphipathic alpha helix confers chain breaking antioxidant activity associated with an inhibition in lipid peroxide amplification on phospholipid surfaces, permits the design and synthesis of peptides that have anti-oxidant activity and which thereby allow prevention of inflammatory events associated with the onset of CVD and other diseases.

[0054] Designing peptides that exhibit antioxidant activity that are active only upon lipidation in effect enables these peptides to be targeted to areas where there is inflammation and cholesterol deposits. The synthetic peptides based on apoA-IMilano and apoA-IParis bind to lipid surfaces and exert the antioxidant characteristics of the full-length variants. These specific peptides do not promote cholesterol and phospholipid efflux from cells indicating that they are useful as anti-inflammatory agents directed towards preformed HDL and metabolic pathways linked to HDL metabolism. Utilizing information based on the position of the cysteine residue at the polar/nonpolar interface of amphipathic alpha helices, it is possible to create peptides derived from different amphipathic alpha helices of apoA-I which are known to exert cholesterol efflux properties. These specific sequences include, but are not limited to helix 1 (aa 44-65) and helix 10 (aa 209-238) which do mediate the lipidation process establishing a specific link to the ABCA1 transporter. This acts as an entrance to the pathway whereby the ABCA1 receptor, which is responsible for HDL assembly in the artery wall, is upregulated upon cholesterol enrichment of cells. As a result, it is possible to couple the lipidation properties of native apoA-I with the phospholipid directed antixodant activity of apoA-IMilano in the form of a synthetic peptide that specifically targets metabolically active sites of cholesterol deposition thus inhibiting inflammatory events involved in early disease progression.

[0055] Structural analyses revealed identical placement of a cysteine residue at the polar/nonpolar interface of an amphipathic alpha helix within apoE3 thus defining the mechanism for the antioxidant activity of apoE3. A similar “motif” in apoAV is also used to create new peptides. Thus the current peptides take advantage of a unifying structural domain that confers a newly discovered beneficial activity to a broad spectrum of apolipoproteins that exhibit diverse anti-inflammatory activities. This novel feature of specific cysteine placement within amphipathic alpha helices thus provides the basis for the development of therapeutic agents that have wide applicability to prevent the onset of a number of inflammatory related diseases including atherosclerosis, Alzheimer's disease, and osteoporosis.

[0056] Moreover, these peptides have been designed to have both native properties of HDL-associated proteins and the newly discovered antioxidant property. Because the present peptides are based on naturally occurring proteins, they are expected to have above average safety and efficacy profiles.

[0057] Additional peptides are derived from different amphipathic alpha helical repeats of apoA-I including Class Y helices, and combinations thereof, to create novel peptides that possess the native properties of apoA-I in promoting cellular cholesterol efflux in addition to the novel antioxidant activity of apoA-IMilano. This will effectively target peptides to sites of cholesterol deposition and inflammation where there is an upregulation in ABCA1 cholesterol transporter expression. (The ABCA1 transporter is a recently discovered HDL receptor located on aortic macrophages associated with Tangier's Disease and is responsible for the synthesis of HDL in the artery wall.) This is made possible by the unique antioxidant property of the synthetic peptides which is conferred upon lipidation on phospholipid surfaces as well as the ability to incorporate a free cysteine residue within different classes of amphipathic alpha helices. As a result the peptides can be used in combinations with other therapeutic regiments that promote an unregulation in ABCA1 to effectively target to therapeutic peptides to sites of inflammation and cholesterol deposition

[0058] A. Basic Peptide Mimetic Features

[0059] The helical wheel projections in FIG. 1 indicate the position of R173C and R151C within amphipathic alpha helices 6 (residues 145-166) and 7 (residues 167-188) of apoA-I. FIGS. 2 and 3 indicate similar positioning of R112C and C224 in amphipathic alpha helices of apoE3 and apoAV, respectively. Each helix is shown looking down the barrel of the helix. The hydrophobic/hydrophilic interface is drawn as the dotted line to separate the polar from the nonpolar face of the amiphipathic helix. The nonpolar face is the helix is the side and site of lipid interaction, while the polar face interacts with water and water-soluble free radical initiators.

[0060] The cysteine substitutions for arginine in FIG. 1 correspond to apoA-IMilano (helix 7) and apoA-IParis (helix 6). The substitutions are located near the interface of the hydrophobic/hydrophilic surfaces of their respective amphipathic alpha helices. This unique location of the cysteine residue near the hydrophobic/hydrophilic interface demonstrate that cysteine substitutions at strategic loci on the amphipathic alpha helices of HDL-associated proteins can protect phospholipids from water soluble free radical initiators in addition to lipoxygenase-mediated mechanisms which occur on phospholipid surfaces. Moreover, the apparent water accessibility of the free thiol allows for important interactions with water-soluble antioxidants enhancing the antioxidant activity of apoA-IMilano and its peptide mimetics.

[0061] Based on this model, the peptides of this invention take advantage of this observation and direct the design of peptides that have a cysteine residue present at the polar/nonpolar interface of an amphipathic alpha helix. The generated peptide should thereby exhibit an antioxidant property which can thus protect phospholipids from water soluble free radical initiators.

[0062] Since a goal of this invention is to form peptide mimetics that are derived from naturally occurring proteins, so as to be safe and not prone to eliciting an immune response in a patient, it is preferred that the present peptides are derived from HDL-associated proteins. Appropriate HDL-associated proteins include but are not limited to, Apolipoprotein A-I, Apolipoprotein A-V, Apoliproprotein E3, Apolipoprotein E4, Human Serum Paraoxonase and their variants. As each of these proteins is associated with HDL and exhibit the same apparent structural motif, the current peptides of this invention are derived from specific amphipathic alpha helices in these proteins that known to possess cysteine residues and/or derived from helical segments engineered to possess a free cysteine at the polar/nonpolar interface of amphipathic alpha helices.

[0063] It is also important that the peptides that are created meet with the following criteria, that they 1) exhibit antioxidant activity that is directed toward lipid surfaces, 2) be unable to quench water-soluble free radicals in the absence of lipids and 3) have potential interactions with water-soluble antioxidants. These peptides can be tested to meet this criteria through several simple experiments which take minutes to complete. FIG. 4 outlines these three simple assays used to define that antioxidant characteristics of each peptide. The assay to test antioxidant activity directed toward lipid surfaces can be test using the assay described in Example 3. The assay in Example 4 can be used to test potential interactions with other water-soluble antioxidants and the Cytochome C reduction assay. Example 7 can be used to test whether the peptide is unable to quench water-soluble free radicals absent lipids. Other assays that are disclosed in Examples 4 and 5, which test whether the peptide is capable of LCAT activation or cholesterol efflux from cells, are used to determine whether the peptides exhibit the native properties of the protein from which the peptides is based on.

[0064] Preferably these peptides should exhibit a percent protection of phospholipids from oxidation at a concentration that is at least the same level of the apoA-I Milano and Paris variants as shown in FIG. 5. Therefore, suitable peptides should exhibit at least a 50% protection against maximum accumulation of lipid peroxides at a concentration of preferably no more than 200 μg/mL. The peptides should more preferably exhibit 50% protection of maximum accumulation of phospholipids at a 250 μg/mL concentration, even more preferably no more than 300 μg/mL, and even more preferably no more than 400 μg/mL of peptide, and most preferably at concentrations no greater than 500 μg/mL.

[0065] The peptides may be made and purified by methods known in the art, preferably by in vitro automated synthesis, but also by recombinant DNA methods. Furthermore, these peptides can be synthesized using L-amino acids, non-natural or other modified amino acids, as is known in the art, in order to synthesize peptides which can act upon targets in the body and be degraded, yet do not interfere with normal protein function. The peptides can be stored in lyopholized form and dissolved in aqueous buffers or water prior to use. For the purposes of experimental use, the peptides are dissolved in sterilized degassed buffers to optimize biological activity which remains stable over 1-3 months at 4° C.

[0066] The synthetic peptides of the present invention could contain at least 1-4 cysteine residues. In place of cysteine, other residues may be employed that also contain a reducing moiety, namely a thiol (SH) group. Creating peptides with other thiol-bearing moieties would confer an increased nucleophilicity and thereby increase the ability to reduce free radicals. However, this could potentially interfere with important interactions with water-soluble antioxidants and thus limit potential use in humans.

[0067] B. Designing Peptides from HDL-Associated Proteins

[0068] The starting point for the present model is Apolipoprotein A-I and other HDL-assoociated proteins. As shown in the sub-sequences in Examples 7-13, the present peptides are derived from several regions of the wild-type apoA-I protein (GenBank Accession No. P02647 locus APA1_HUMAN), Apolipoprotein E3 (apoE3) as set forth in GenBank Accession No. P02649 locus APE_HUMAN, Apolipoprotein A-V (apoA-V) as set forth in GenBank Accession No. NP—44320 locus XP—443200 (Science 294:169-173) and Human Serum Paraoxonase (PON) as set forth in GenBank Accession No. XP—043694 locus XP—043694 (NCBI REFSEQ), particularly regions that have amphipathic alpha helices that contain a cysteine residue at the polar/nonpolar interface.

[0069] Good candidates for peptides useful in the invention are peptides based on protein domains of HDL-associated protein molecules that are amphipathic alpha helices having a cysteine at the hydrophic/hydrophilic interface because such regions are most likely to interact with lipid surfaces and be able to confer antioxidant activity. Generally, the helical segments are marked at their boundary by proline residues. Helical wheel projections as shown in FIGS. 1-3 are constructed to define the position of the cysteine residue at the polar/nonpolar interface of the helix. Additional helical segments not known to possess cysteine residues are used because their unique qualities that permit lipidation (i.e. allow the peptides to promote cholesterol efflux from cells such as J774 macrophages). Important helical segments are identified based on published information (J. Biol. Che,. 274: 2021-2028) include helix 1 (aa 44-65) and helix 10 (aa 209-238) of apoA-I that are important in the nucleation of lipidation.

[0070] C. Cysteine Placement that Influences Antioxidant Activity

[0071] Peptides based on helix 1 and helix 10 of apoA-I can be used in the form of a single 18-mer in which a cysteine residue has been strategically added to the polar/nonpolar interface of the amphipathic alpha helix as in apoA-IMilano. The position of the cysteine residue is set between 1 and 4 amino acids off the interface to mimic the position in the natural variants. In an idealized model peptide (18-mer) in which the numbering represents a consecutive sequence of amino acids 1-18, the cysteine can be placed at positions 7, 11, 18 or 5, 9, or 16 depending on the positioning of the nonpolar face of the helix in the native structures defined by the helices derived from apolipoproteins with known cysteine residues such as apoA-IMilano and apoA-IParis. However, this generalized numbering scheme may not apply to all helices such as helix 1 of apoA-I (aa 44-61) where the cysteine can be placed at positions 4, 11, 18 or 2, 6, 13 to mimic the natural positioning of the cysteine residue at the interface of amphipathic alpha helices. But, in generalized terms, the cysteine residue can be placed between 1-4 residues off the interface of the helix to confer thiol dependent antioxidant activity.

[0072] The ability to utilize amphipathic alpha helices that are not known to possess a cysteine residue, such as helix 1 and 10 of apoA-I, is based on the following observations: 1) the synthetic peptides derived from helix 6 and 7 of apoA-IMilano and apoA-IParis (peptides SEQ ID NO: 1 and SEQ ID NO: 9, respectively) as well as the peptide derived from apoE3 all possess the same antioxidant characteristics despite the fact that they differ in primary amino acid sequence (data shown in FIGS. 5, 6 and 9). This highlights the critical contribution of specific cysteine placement which is conversed in each of the parent apolipoproteins. 2) A cysteine residue added to the polar/nonpolar interface of helix 10 of apoA-I (peptide SEQ ID NO: 16) possesses antioxidant activity similar to the peptide derived from apoA-IMilano (data shown in Example 8 and FIG. 8).

[0073] Specific placements of a free cysteine residue render the amphipathic peptide relatively protected against free radical mediated oxidation of phospholipid. The free thiol group does not directly quench the free radical, but instead prevents the initiation/amphification of lipid peroxidation.

[0074] There are three categories that represent specific placement of cysteine residues around the face of an amphipathic alpha helix as shown in the helical wheel projection of an 18-mer peptide in FIGS. 1-3. First, placement of a cysteine residue at the polar/nonpolar interface of the helix (i.e. 1-3 amino acids away from the interface as judged by the two dimensional face of the wheel projection) renders the peptide fully effective as an antioxidant when oxidation is initiated in the water phase or on the phospholipid surface. Utilizing a generalized number scheme for the sequence of amino acids (1-18) in an 18 mer peptide, peptides can be engineered from the following apolipoproteins where the position of the cysteine residues in noted by numeric position: apoA-IMilano and apoA-IParis based peptides, positions 7, 11, 18 or 5, 9, 16; ApoE3 based peptides, positions 4, 8, 15 or 3, 7, 10; apoAV based peptides, positions 4, 11, 15 or 6, 13, 17; generic peptide I (18-A) based peptides, 4, 8, 15 or 2, 9, 13; and the generic sequence II, positions 7, 11, 18 or 5, 9, 16.

[0075] Second, placing the cysteine residue in middle of the hydrophobic or hydrophilic face of the helix may impart functionality or result in loss of functionality by disrupting salt bridges, etc., depending on what the goal of the peptide use is.

[0076] Third, peptides with multiple cysteines may be made by placing cysteines at the hydrophobic/hydrophilic interface, as well as in either the hydrophobic or hydrophilic faces of the helix. Such changes in placement of the cysteine residue is predicted to specifically alter the functionality of the peptides in a systematic fashion. This can lead to the creation of a generic amphipathic alpha helix for the development of tailor-made pharmacteuticals of defined functionality including specific antioxidant activity attributed to strategic cysteine placement, LCAT activation properties endowed via arginine clustering at the polar/nonpolar interface, and cholesterol efflux properties obtained by either phenylalanine placement or by combining unique helical segments.

[0077] The generalized placement of a cysteine at the polar/nonpolar interface of amphipathic alpha helix is represented in FIGS. 10 and 11. FIGS. 1-3 also illustrate the conservation of the cysteine placement in apoA-IMilano, apoA-IParis, apoE3 and apoAV. In some instances the specific cysteine placement disrupts salt bridges which may be important for allowing increased mobility of the free thiol at the hydrophobic/hydrophilic interface. This is true for peptides based on apoA-IMilano, apoA-IParis and apoE3 which result from R→C interchanges in class A amphipathic alpha helices.

[0078] By definition, class A amphipathic alpha helices possess positively charged amino acids such as arginine (R) at the interface of the polar/nonpolar surface of the helix which may be important in designing therapeutic peptides. In general, the positioning of the cysteine is positioned near the interface either 1-3 or 1-4 amino acids off the interface into the aqueous phase which may be utilized to generate a peptide with antioxidant activity. Antioxidant activity of peptides, wherein the cysteine is placed further away from the nonpolar surface, may be influenced by the overall lipid binding affinity of the helical segment and its ability to penetrate into phospholipid surfaces. Lipid binding affinity is influenced by the contribution of specific hydrophobic amino acids located in the nonpolar face of the helix as well as the distribution of positively charged residues such as lysine and arginine residue in the polar surface promote electrostatic interactions with phospholipids.

[0079] An important feature of some of the synthetic peptides is the movement of the cysteine to the middle of the nonpolar surface of the helix which, in and of itself, may not influence antioxidant activity of the individual synthetic peptide. But such peptides are theorized to lack specific interactions with water-soluble antioxidants that enhance antioxidant activity of the thiol-containing apolipoproteins, such as apoA-IMilano, and their peptide mimetics. Loss of such important interactions would generate an important series of peptide mimetics that could be used experimentally to determine whether such interactions are important in preventing inflammatory related diseases, thereby allowing the identification of new drug targets and permitting future drug design.

[0080] In general, the free cysteine residue can be moved around the face of the helix in single-turn fashion as illustrated in the peptide based on apoA-IMilano. The natural position of the cysteine residue in apoA-IMilano (helix 7, aa 167-184) is found at position 173. Movement of the thiol around the face of the helix is achieved via specific placement at 171 (located at the opposite interface), 172 (thiol positioned in the middle of the hydrophilic surface), and 174 (thiol positioned in the middle of the hydrophobic surface). However, it is possible that cysteine placement at other interfacial sites confers antioxidant activity and it is possible that other sites towards the middle of the hydrophobic surface are useful in designing peptides.

[0081] Moreover, various placement schemes are used to create peptides containing two or more free cysteine residues. For example, peptides can be engineered to possess two or three thiols: one in the water face of the helix, one in the lipid face, and one at the polar/nonpolar interface. General examples of the structural placement of multiple cysteine residues are represented in FIG. 11. This last strategy of cysteine placement is to be used with caution as the multiple cysteines may serve to dilute the specificity and unique properties of the peptides rather than enhance their antioxidant properties.

[0082] D. Extending the Peptides

[0083] The present peptides are based on a modeled amphipathic alpha helical structure. Accordingly, they may be from about 12 to 100 amino acids in length, preferably 18-40 amino acids in length, more preferably 18-20 amino acids in length. The peptide subsequences can be extended in either the amino and carboxy direction or both, with the sequence from the native protein from which the peptide was derived.

[0084] When extending the peptides, in one embodiment, beyond the peptide amphipathic helix in the amino and/or carboxy directions, it is preferred that the sequence of the native ApoA-I, as set forth in GenBank Accession No.: P02647, is used. In a separate preferred embodiment, the sequences of native Apolipoprotein E3 (apoE3) as set forth in GenBank Accession No.: P02649, Apolipoprotein A-V (apoAV) as set forth in GenBank Accession No.: NP—44320 or human serum paraoxonase (PON) as set forth in GenBank Accession No.: XP—043694, are used to extend the peptide.

[0085] The extended sequence need not be identical to the recited sequences above, however it should be substantially identical, preferably at least 80% homologous.

[0086] In another preferred embodiment, multiple amphipathic alpha helical peptides having cysteine substitutions can be used to extend the peptides to create a larger peptide in which multiple domains have antioxidant properties. See Example 8, specifically SEQ ID NOS: 30,31, 39-46.

[0087] Depending upon what the targeted disease, proteins and events are will dictate which sequence is used to extend the peptides. For example, if the targeted oxidation events are related to Alzheimer's Disease, then the peptide should probably be extended with sequence having homology to apoE3. If the targeted oxidation events are related to atherosclerosis, the peptide can be extended with the sequences homologous to apoA-I or PON.

[0088] E. Applications and Therapeutics

[0089] These peptides make feasible the preparation and administration (either orally or intravenously) of agents that carry the beneficial properties of the full length apoA-IMilano and apoA-IParis proteins. The present peptides may be used to prevent CVD in the general population, based on this newly described activity associated with the presence of the free thiol in the monomeric form of apoA-IMilano and its peptide mimetics. Therapeutics derived from the dimeric form of the variant, which lacks the antioxidant activity attributed to the monomeric form of the variant, is currently in pharmaceutical development. The present antioxidant peptides are also useful in preventing ischemia following bypass surgery and/or after myocardial infarction, since the present peptides move into and out of artieries with lipoproteins such as HDL. A recently discovered HDL receptor located on aortic macrophages is associated with Tangier's Disease (the ABCAI transporter) and is responsible for the synthesis of HDL in the artery wall. In one embodiment, the present peptides may be used to promote cellular cholesterol removal from macrophages in the arterial wall via ABCA1.

[0090] The present peptides may be prepared according to known pharmaceutical technology. They may be administered singly or in combination, and may further be administered in combination with other cardiovascular drugs. They may be conventionally prepared with excipients and stabilizers in sterilized, lyophilized,powdered form for injection, or prepared with stabilizers and peptidase inhibitors of oral-and gastrointestinal metabolism for oral administration.

EXAMPLE 1

Preparation of Synthetic Peptide Mimetics

[0091] Synthetic peptides were engineered from the monomeric forms of apoA-IMilano, apoA-IParis and apoE3. Peptides which lacked cysteine residues were developed from wild-type apoA-I and the apoE4 isoform and served as controls. All peptides were purchased from Biosynthesis Incorporated (Lewisville, Tex.) and were modified by an N-terminal acetyl group and C-terminal amide group. Peptides were dissolved in sterile, filtered, degassed 10 mM Tris-buffered (pH=8.0) Saline EDTA (2.7 mM) and stored at 4° C. The antioxidant activity of the thiol-containing peptides remained stable over a 3 month period when stored in this manner.

EXAMPLE 2

Micelle Substrate to Test the Antioxidant Activity of the Peptides

[0092] A schematic showing the assays used for determining the antioxidant activity of synthetic peptide mimetics is shown briefly at FIG. 4. Assays 1 and 2 utilize a micelle substrate composed of 1 mM 1-palmitoyl-2-linoleoylphosphatidylcholine dispersed in Borate (0.2 M) buffered (pH=9.0)/saline-EDTA (2.7 mM) containing deoxycholate. The buffer is made by adding 1.52 grams of deoxycholate to 50 ml of borate buffer/saline-EDTA. The phospholipid is dried on the surface of a glass tube and resuspended with borate/saline-EDTA/deoxycholate and vortexed to dissolve the lipid. The tube is incubated at 37° C. for 10 minutes and allowed to cool to 25° C. before use.

EXAMPLE 3

Assay to Test Perodixation of phospholipids in Presence of Peptides

[0093] The oxidation system consisted of a micelle substrate composed of 1-palmitoyl-2-linoleoylphosphatidycholine (3 mM) dispersed in borate (pH=9.0)/saline-EDTA (2.7 mM) and deoxycholate (6 mM) as described. Phospholipid micelles were used throughout most of these studies to optimize rates of lipid peroxidation catalyzed by specific enzymes. This permitted us to quantify initial rates reliably and in reproducible fashion. Soybean lipoxygenase (5 U/μl) and xanthine (0.2 mM)/xanthine oxidase (20 U/ml) were used to initiate lipid peroxidation following the addition of recombinant apolipoproteins to the phospholipid micelles. Increases in conjugated dienes (lipid peroxidation) were monitored by ultraviolet absorption spectroscopy (234 nm) at 25° C. The mass of phospholipid hydroperoxides was calculated using the molar absorptivity coefficient (ε=29,500 Lcm−1 mol−1) of conjugated dienes. This is made possible because the phospholipid used possesses only two carbon-carbon double bonds; as such, only one conjugated diene species is formed per phospholipid molecule. Initial rates of lipoxygenase mediated lipid peroxidation are calculated from the slopes of the linear portion of the oxidation curves and results can be expressed as nmoles of phospholipid peroxide formed/min.

[0094] Based on the maximum levels of lipid peroxide accumulation obtained in the absence of peptide (i.e. the plateau associated with the oxidation curves), it is possible to derive quantitative information regarding the potency of the peptide (i.e. the concentration of peptide resulting in 50% protection against lipid peroxidation), Thiol-containing peptides based on apoA-IMilano, apoA-IParis and apoE3 generally give 50% protection at a concentration of 200 μg/ml.

EXAMPLE 4

Assays to Test Potential Interaction of Peptides with Other Water-soluble Anti-oxidants

[0095] Interactions of between apoA-IMilano (and other peptides) with reduced glutathione were evaluated using phospholipid micelles and lipoxygenase (5 U/μl). The latter initiates lipid peroxidation on phospholipid surfaces. Glutathione (GSH) is used a concentration which range from 0.025 to 0.1 mM which is added to the phospholipid micelles before the addition of lipoxygenase. GSH is also added in combination with apoA-IMilano (or its peptide mimetics) and lipid peroxidation monitored at 234 nm. The capacity of GSH plus apoA-IMilano (or other peptides) to inhibit lipid peroxidation is compared to the inhibitory action of the thiol-containing compounds alone. Water-soluble free radicals useful for this assay include but are not limited to any known biologically effective antioxidant, such as GSH, vitamin C, vitamin E and N-acetyl cysteine (NAC).

EXAMPLE 5

Assay to Test LCAT Activation Properties of Synthetic Peptides

[0096] ApoA-1 is a cofactor of LCAT which esterifies cholesterol on HDL. The ability of synthetic peptides to activate Lecithin:Cholesterol Acyltransferase (LCAT) was examined using a standard proteoliposome substrate (Chen and Alber, J Lipid Res. 23:680-691). The substrate contained the synthetic peptide of interest, phosphatidylcholine (egg yolk PC) and unesterfied cholesterol at the following mole ratios: 15:250:12.5. Trace amounts of [14C]cholesterol are added to the proteoliposome during preparation. To monitor cholesterol esterification, reaction mixtures are prepared with the following constituents, [14C]cholesterol containing proteoliposomes (4.4×105 dpm/ml), 20 mM Tris (pH 8.0), 0.15 mM NaCl, 0.27 mM EDTA, 0.5% human serum albumin, 2.0 mM β-mercaptoethanol and recombinant human LCAT enzyme (20 μg/ml). Results are expressed as a percentage of [14C]cholesterol converted to [14C]cholesterol esters in a 30 minute assay at 37° C.

EXAMPLE 6

Assay to Determine Cellular Cholesterol Efflux Capability of the Peptides

[0097] The main function of ApoA-I is promoting cholesterol efflux from cells. This process results in formation of HDL particles. Therefore this assay is used to show that the peptides possess the native properties of apoA-I in promoting HDL assembly. The murine macrophage cell-line, J774, was used as cholesterol donors for efflux studies to lipid-free apolipoproteins or synthetic peptides. This cell-line was chosen because it has recently been shown to possess an active apolipoprotein-mediated efflux pathway involving ABCA1 which is up-regulated by the cAMP analog, 8-(4-chlorophenthio)adenosine 3′:5′-cyclic monophosphate. Briefly, 1×105 cells/ml were seeded into 24 well culture plates and labeled with 1 μCi/ml of [3H]cholesterol dispersed in RMPI 1640 medium containing 1% FBS. Confluent monolayers of radio-labeled cells were equilibrated (2 h) with RPMI containing 0.2% BSA and extensively rinsed with serum-free RPMI prior to addition of recombinant apolipoproteins or synthetic peptides. In some instances 0.3 mM of the cAMP analog was added to serum-free medium to upregulate cellular cholesterol efflux. Lipid-free apolipoproteins and/or synthetic peptides were added (25 μg/ml) to serum-free RPMI and applied to cells. At specified times, aliquots of medium were removed and cellular debris pelleted by centrifugation (1000×g, 10 min). Results were expressed as a percentage of the initial cellular [3H]cholesterol appearing in the medium at each time point.

EXAMPLE 7

Assay Confirming that Peptides do not Quench Water Soluble Reactive Oxygen Species

[0098] A stock solution (1 mg/ml) of cytochrome C is prepared and 50 μg/ml is added to 0.2 mM xanthine solution. Xanthine oxidase (20 U/ml) is added to generate water soluble reactive oxygen species (i.e. superoxide anion and hydroxyl radicals). Reduction of cytochrome C is followed at 550 nm over a time course to determine whether synthetic peptides directly quench the reactive oxygen species (ROS) in the absence of phospholipid.

[0099] The rate of reduction of cytochorme C was compared between X/Xo, X/Xo plus the control peptide (400 μg/ml), X/Xo plus the thiol-containing peptide (400 μg/ml) and SOD (superoxide dismutase) was used as a control. The synthetic peptides should fail to protect cytochrome C indicating that the thiol-containing peptide is unable to directly quench ROS in the aqueous phase.

EXAMPLE 8

Designing Peptides from Amphipathic Helices in ApoA-IMilano

[0100] The following lists sequences of amino acids used to prepare peptides which exhibited the newly discovered anti-oxidant activity of apoA-IMilano. Control sequences based on amphipathic alpha helices that lack cysteine residues are also listed. Alternative positions for the cysteine residues are listed for each sequence and are important both therapeutically and as biological tools to investigate the underlying basis of inflammatory related diseases.

[0101] Synthetic peptides based on the primary amino acid (aa) sequence (aa 167-184) where the R173C mutation can be found in apoA-IMilano. SEQ ID NO: 1 mimics the precise location of the cysteine residue in apoA-IMilano SEQ ID NO: 2 is peptide based on wild-type apoA-I which lacks a cysteine residue. The underlined residues in SEQ ID NO: 2 represent alternative positions for the cysteine residue. SEQ ID NOS: 3-8 show peptides made with the underlined cysteine substitutions.

SDELRQCLAARLEALKEN	167-R173C-184	SEQ ID NO:1
SDELRQRLAARLEALKEN	Control wild type
	167-184	SEQ ID NO:2
SDELCQRLAARLEALKEN	167-R171C-184	SEQ ID NO:3
SDELRCRLAARLEALKEN	167-Q172C-184	SEQ ID NO:4
SDELRQRCAARLEALKEN	167-L174C-184	SEQ ID NO:5
SDELRQRLCARLEALKEN	167-R175C-184	SEQ ID NO:6
SDELRQRLACRLEALKEN	167-A176C-184	SEQ ID NO:7
SDELRQRLAARLEACKEN	167-L181C-184	SEQ ID NO:8

[0102] Synthetic peptides based on the primary amino acid sequence (145-162) where the R151C mutation can be found in apoA-IParis. The sequence in SEQ ID NO: 9 mimics the precise location of the cysteine residue in apoA-Iparis. SEQ ID NO: 10, sequence of control peptide based on wild-type apoA-I which lacks a cysteine residue. The underlined residues in SEQ ID NO: 10 represent alternative positions for the cysteine residue. The sequences in SEQ ID NOS: 11-15 are peptides made with the underlined cysteine substitutions.

GEEMRDCARAHVDALRTH	145-R151C-162	SEQ ID NO:9
GEEMRDRARAHVDALRTH	Control wild type	SEQ ID NO:10
	145-162
GEEMCDRARAHVDALRTH	145-R149C-162	SEQ ID NO:11
GEEMRCRARAHVDALRTH	145-D150C-162	SEQ ID NO:12
GEEMRDRCRAHVDALRTH	145-A152C-162	SEQ ID NO:13
GEEMRDRACAHVDALRTH	145-R153C-162	SEQ ID NO:14
GEEMRDRARACVDALRTH	145-H155C-162	SEQ ID NO:15

[0103] Synthetic peptides based on amino acids 220-237 of wild-type apoA-I. SEQ ID NO: 16 lists a sequence that mimics the position of the cysteine residue at the polar/nonpolar interface of the amphipathic alpha helix as can be found in the apoA-IMilano based peptides (Line 1). SEQ ID NO: 17 corresponds to a control sequence based on wild-type apoA-I that lacks a cysteine residue. The underlined residues in SEQ ID NO: 17 represent alternative positions for the cysteine residues. The sequences in SEQ ID NOS: 18-29 are peptides made with the underlined cysteine substitutions.

PVLESFCVSFLSALEEYT	220-K226C-237	SEQ ID NO:16
PVLESFKVSFLSALEEYT	Control wild type	SEQ ID NO:17
	220-237
PVLCSFKVSFLSALEEYT	220-E223C-237	SEQ ID NO:18
PVLECFKVSFLSALEEYT	220-S224C-237	SEQ ID NO:19
PVLESCKVSFLSALEEYT	220-F225C-237	SEQ ID NO:20
PVLESFKCSFLSALEEYT	220-V227C-237	SEQ ID NO:21
PVLESFKVCFLSALEEYT	220-S228C-237	SEQ ID NO:22
PVLESFKVSCLSALEEYT	220-F229C-237	SEQ ID NO:23
PVLESFKVSFCSALEEYT	220-L230C-237	SEQ ID NO:24
PVLESFKVSFLCALEEYT	220-S231C-237	SEQ ID NO:25
PVLESFKVSFLSCLEEYT	220-A232C-237	SEQ ID NO:26
PVLESFKVSFLSALCEYT	220-E234C-237	SEQ ID NO:27
PVLESFKVSFLSALECYT	220-E235C-237	SEQ ID NO:28
PVLESFKVSFLSALEECT	220-Y236C-237	SEQ ID NO:29

[0104] SEQ ID NO: 30 corresponds to amino acids 209-241 of wild-type apoA-I in which a cysteine has been added to the polar/nonpolar interface of the amphipathic alpha helix. This peptide possesses both the native cholesterol efflux properties of apoA-I (J. Biol. Chem. 274:2021-2028) and has been endowed with thiol dependent antioxidant activity. SEQ ID NO: 31 corresponds to a control peptide that lacks a cysteine residues. The underlined residues represents alternative sites for cysteine substitutions either singly or in combination to make new peptides. The core sequence is identical to that listed above (SEQ ID NO: 17) and the position of the cysteine residue follow those listed for SEQ ID NO: 18-29).

PALEDLRQGLL PVLESFCVSFLSALEEYT KKLN SEQ ID NO:30
PALEDLRQGLL PVLESFKVSFLSALEEYT KKLN SEQ ID NO:31

[0105] Synthetic peptides based on amino acids 44-61 of wild-type apoA-I. SEQ ID NO: 32 lists a sequence of a peptide containing a cysteine residue at the polar/nonpolar interface of the amphipathic alpha helix just as in apoA-IMilano. SEQ ID NO: 33 corresponds to a control sequence based on wild-type apoA-I that lacks a cysteine residue. The underlined residues in SEQ ID NO: 33 represent alternative positions for the cysteine residue. SEQ ID NOS: 34-38 are peptides with the underlined cysteine substitutions.

LKLCDNWDSVTSTFSKLR	44-L47C-61	SEQ ID NO:32
LKLLDNWDSVTSTFSKLR	Control wild type	SEQ ID NO:33
	44-61
LCLLDNWDSVTSTFSKLR	44-K45C-61	SEQ ID NO:34
LKCLDNWDSVTSTFSKLR	44-L46C-61	SEQ ID NO:35
LKLLCNWDSVTSTFSKLR	44-D48C-61	SEQ ID NO:36
LKLLDCWDSVTSTFSKLR	44-N49C-61	SEQ ID NO:37
LKLLDNWDSVTSTFSCLR	44-K59C-61	SEQ ID NO:38

[0106] Synthetic peptides based on a combination of helices (209-220 plus 44-65) found in wild-type apoA-I. SEQ ID NO: 39 lists the sequence of a peptide containing a cysteine residue located at the polar/nonpolar interface of an amphipathic alpha helix just as in apoA-IMilano SEQ ID NO: 40 corresponds to a control sequence based on wild-type apoA-I that lacks cysteine residues. The underlined residues in SEQ ID NO: 40 represent alternative positions for the cysteine residue. SEQ ID NOS: 41-46 are peptides with those underlined cysteine substitutions.

PALEDLRQGLLP LKLCD	209/44-L47C-61	SEQ ID NO:39
NWDSVTSTFSKLR
PALEDLRQGLLP LKLLD	Control209/44-6l	SEQ ID NO:40
NWDSVTSTFSKLR
PALEDLCQGLLP LKLLD	209-R215C-220/44-61	SEQ ID NO:41
NWDSVTSTFSKLR
PALEDLRQGLLP LCLLD	209/44-K45C-61	SEQ ID NO:42
NWDSVTSTFSKLR
PALEDLRQGLLP LKCLD	209/44-L46C-61	SEQ ID NO:43
NWDSVTSTFSKLR
PALEDLRQGLLP LKLLC	209/44-D48C-61	SEQ ID NO:44
NWDSVTSTFSKLR
PALEDLRQGLLP LKLLD	209/44-N49C-61	SEQ ID NO:45
CWDSVTSTFSKLR
PALEDLRQGLLP LKLLD	209/44--K59C-61	SEQ ID NO:46
NWDSVTSTFSCLR

EXAMPLE 9

Amphipathic Antioxidant Peptide Based on Human Apolipoprotein E3

[0107] Synthetic peptides based on amino acids (105-122) of apoE3. The sequence in SEQ ID NO: 47 mimics the precise location of the cysteine residue in human apoE3. SEQ ID NO: 48 corresponds to a control peptide based on the primary amino acid sequence of apoE4 which lacks cysteine residues. The underlined residues in SEQ ID NO: 48 represent alternative positions for the cysteine residue. The sequences in SEQ ID NOS: 49-51 are peptides with those underlined cysteine substitutions.

GADMEDVCGRLVQYRGEV	105-R112C-122	SEQ ID NO:47
GADMEDVRGRLVQYRGEV	Control wild type	SEQ ID NO:48
	105-122
GADMEDCRGRLVQYRGEV	105-V111C-122	SEQ ID NO:49
GADMEDVRCRLVQYRGEV	105-G113C-122	SEQ ID NO:50
GADMEDVRGCLVQYRGEV	105-R114C-122	SEQ ID NO:51

EXAMPLE 10

Amphipathic Antioxidant Peptide Based on Apolipoprotein A-V

[0108] Synthetic peptides based on Apolipoprotein A-V. SEQ ID NO: 52 mimics the precise location of the cysteine residue in human apoAV amino acids 219-236. SEQ ID NO: 53 corresponds to a control peptide based on the same sequence as shown in SEQ ID NO: 52 except the cysteine residue has been replaced with a glycine residue to generate a peptide which lacks the cysteine. The underlined residues in SEQ ID NO: 53 represent alternative positions for the cysteine residue.

ARLSRCVQVLSRKLTLKA	219-G224C-236	SEQ ID NO:52
ARLSRGVQVLSRKLTLKA	Control wild type	SEQ ID NO:53
	219-236
ARLCRGVQVLSRKLTLKA	219-S222C-236	SEQ ID NO:54
ARLSCGVQVLSRKLTLKA	219-R223C-236	SEQ ID NO:55
ARLSRGCQVLSRKLTLKA	219-V225C-236	SEQ ID NO:56
ARLSRGVQVLSRKCTLKA	219-L232C-236	SEQ ID NO:57

[0109] SEQ ID NO: 58 lists a sequence of 36 amino acids (219-254) found in apoAV. SEQ ID NO: 59 is a control peptide based on peptide listed in SEQ ID NO: 58 except the cysteine has been replaced with a glycine residue. The underlined residues in SEQ ID NO: 58 represent alternative positions for the cysteine residue. SEQ ID NO: 60-63 are the peptides with the underlined cysteine substitutions.

ARLSRCVQVLSRKLTLKAKALHAR	219-G224C-254	SEQ ID NO:58
IQQNLDQLREEL
ARLSRGVQVLSRKLTLKAKALHAR	Control	SEQ ID NO:59
IQQNLDQLREEL	219-254
ARLCRGVQVLSRKLTLKAKALHAR	219-S222C-254	SEQ ID NO:60
IQQNLDQLREEL
ARLSCGVQVLSRKLTLKAKALHAR	219-R223C-254	SEQ ID NO:61
IQQNLDQLREEL
ARLSRGCQVLSRKLTLKAKALHAR	219-V225C-254	SEQ ID NO:62
IQQNLDQLREEL
ARLSRGVQVLSRKCTLKAKALHAR	219-L232C-254	SEQ ID NO:63
IQQNLDQLREEL

[0110] SEQ ID NO: 64 lists a sequence based on amino acids 51-72 of apoAV that has been engineered to possess a cysteine residue at the polar/nonpolar interface of the amphipathic alpha helix just as in apoA-IMilano The control peptide in SEQ ID NO: 65 does not contain a cysteine residue, but the underlined residues correspond to alternative sites for cysteine substitutions.

ATLKDSLCQDLNNMNKFLEKLR	51-E58C-72	SEQ ID NO:64
ATLKDSLEQDLNNMNKFLEKLR	Control	SEQ ID NO:65
	wild type
	51-72
ATLCDSLEQDLNNMNKFLEKLR	51-K54C-72	SEQ ID NO:66
ATLKDCLEQDLNNMNKFLEKLR	51-S56C-72	SEQ ID NO:67
ATLKDSCEQDLNNMNKFLEKLR	51-L57C-72	SEQ ID NO:68
ATLKDSLECDLNNMNKFLEKLR	51-Q59C-72	SEQ ID NO:69
ATLKDSLEQCLNNMNKFLEKLR	51-D60C-72	SEQ ID NO:70

EXAMPLE 11

Peptide Based on Human Serum Paraoxonase

[0111] Human serum paraoxonase (PON1A) possesses thiol-dependent antixoidant activity, however, the domain structure of the enzyme is not well defined. It has been reported previously that the enzyme can inhibit lipoxygenase mediated lipid peroxidation (Brushia et al, J. Lipid Res. 42:951-958) utilizing the protocols set forth in the Examples which indicate that peptides derived from aspects of paraoxonase secondary structure may be useful in the design of therapeutic agents. Not shown is a two dimensional wheel projection of the synthetic peptide that encompasses the antioxidant domain of human serum paraoxonase. Amino acid residues 276-293 form an amphipathic alpha helix having a cysteine located at the interface of the hydrophilic/hydrophobic interface. Below is a brief list of peptides that possess beneficial potential as antioxidants. The sequences were derived from the basic criteria established in this patent disclosure including specific cysteine placement within amino acid stretches separated by proline residues. Moreover, the native PON enzyme is an HDL-associated protein that appears to possess thiol-dependent antioxidant activity directed toward lipid surfaces.

ETGDLWVGCHP                      SEQ ID NO:71
ETGDLWVGCHPNGMKIFFYDSEN          SEQ ID NO:72
LKSLDFNTLVDNISVDP ETGDLWVGCHPNGMKIFFY SEQ ID NO:73
DSEN

EXAMPLE 12

Amphipathic Antioxidant Peptide Based on a Generic Peptide

[0112] The sequence of the published (generic) peptide by Segrest et al., in U.S. Pat. No. 4,643,988, DWLKAFYDKVAEKLKEAF, which codes for an alpha helix unrelated to apoA-I, can be modified for purposes of this invention. This peptide has been made to model apoA-I amphipathic alpha helices and used it extensively to study apoA-I structure and function (Yancey et al. Biochemistry, 1995, vol 34, 7955-7965). Because it has been used often to study alpha helices, cysteine residues can be introduced into this peptide to model antioxidant activity in a generic sequence. SEQ ID NO: 74 is the Segrest peptide with a Cysteine placed at the interface. Alternate residues of cysteine substitution are underlined in control peptide SEQ ID NO: 75. SEQ ID NOS: 76-81 are peptides having those underlined cysteine substitutions.

	DWLCAFYDKVAEKLKEAF	18A-K4C	SEQ ID NO:74
	DWLKAFYDKVAEKLKEAF	18A control	SEQ ID NO:75
	DCLKAFYDKVAEKLKEAF	18A-W2C	SEQ ID NO:76
	DWCKAFYDKVAEKLKEAF	18A-L3C	SEQ ID NO:77
	DWLKCFYDKVAEKLKEAF	18A-A5C	SEQ ID NO:78
	DWLKACYDKVAEKLKEAF	18A-F6C	SEQ ID NO:79
	DWLKAFYDKCAEKLKEAF	18A-V10C	SEQ ID NO:80
	DWLKAFYDKVCEKLKEAF	18A-A11C	SEQ ID NO:81

EXAMPLE 13

Amphipathic Antioxidant Peptide Based on a Generic Peptide II

[0113] The following peptides are hypothetical in nature but were engineered to possess unique structural aspects of apoA-I (helix 1, aa 44-65) that may be important in promoting cellular cholesterol efflux, as well as, a cluster of arginine residues based on helix 6 (aa 145-166) that play a role in LCAT activation. Moreover, the antioxidant activity of apoA-IMilano has been added to the peptide by virtue of the placement of a free cysteine residue at the polar/nonploar interface of the first helical segment (18-mer).

[0114] The peptide is arranged in a series of two 18-mers separated by a proline residue. The first and second 18-mers contain a non-polar face composed entirely of leucine residues. Conservative substitutions in these domains for isoleucine, phenylalanine, tryptophan, and/or methionine residues can be made to increase the hydrophobicity of the peptide to facilitate lipid interactions. The polar face of this first 18-mer is modeled from helix 1 of apoA-I and it lacks salt-bridge interactions within the peptide and the overall net charge is zero. Cysteine placement at position 7 within the peptide mimics the position of the thiol at the polar/nonpolar interface of an amphipathic alpha helix as found in apoA-IMilano, apoA-Iparis and apoE3. The second 18-mer connected in series via a proline residue is nearly an exact match to the first 18-mer except arginine residues have been added at positions 5 and 16 to mirror the precise arrangement of the conserved amino acids within helix 6 (aa 145-166) of apoA-I. The underlined serine residue can be replaced with a cysteine residue to add antioxidant properties to the second helical repeat.

[0115] The peptide can be used in combined form as shown in SEQ ID NO: 82 or as two singular 18-mers to separate biological activities. The underlined cysteine residue in SEQ ID NO: 83, can be replaced with a serine to remove thiol-dependent antioxidant activity. Conversely, thiol-dependent activity can be added to SEQ ID NO: 84 by replacing the serine with a cysteine residue. The unique feature of the peptides is the ability to precisely add or remove biological activities in a controlled manner to generate an array of biological tools to probe the complex etiology of inflammatory related diseases. This may permit the identification of specific biological activities that are most important in protecting against disease in various genetic models of atherosclerosis thus opening the door for the development of tailor-made pharmaceuticals to combat a variety of inflammatory diseases.

LEKLNSCLRDRLSALTDTPLEELRDSLRSRLDALRST SEQ ID NO:82
LEKLNSCLRDRLSALTDT               SEQ ID NO:83
LEELRDSLRSRLDALRST               SEQ ID NO:84

EXAMPLE 14

Antioxidant Activity of Synthetic Peptide Mimetics of apoA-IMilano

[0116] FIGS. 5A and 5B show the oxidation of phospholipid in the absence (squares) and presence of synthetic peptide mimetics based on apoA-IMilano and apoA-IParis, respectively. Oxidation of phospholipid was achieved by exposing phospholipid micelles to reactive oxygen specieis generated via xanthine/xanthine oxidase. Peptides based on apoA-IMilano (SEQ ID NO: 1) and apoA-IParis (SEQ ID NO: 9) inhibited the oxidation of phospholipid in a dose dependent manner where diamonds, circles, triangles and hatched squares correspond to 100, 200, 300 and 400 μg/ml, respectively. Note that 50% protection was achieved with approximately 200 μg/ml of peptides derived from either apoA-IMilano or apoA-IParis.

[0117] FIG. 6A shows that a peptide which lacks a cysteine residue (167-184, SEQ ID NO: 2) failed to inhibit oxidation of phospholipid induced by reactive oxygen species. Peptide 167-R173C-184 based on apoA-IMilano effectively inhibited lipid peroxidation (FIG. 6B), but the thiol-containing peptide was unable to directly quench water-soluble reactive oxygen species in the aqueous phase as determined using the cytochrome C assay (FIG. 6C). This indicates that the antioxidant activity of the peptide mimetic of apoA-IMilano (SEQ ID NO: 1) was directed toward phospholipid similar to the full-length cysteine variant.

[0118] Phospholipid micelles were exposed to xanthine/xanthine oxidase (X/Xo, 20 U/ml) in the absence of peptide (squares, FIGS. 6A & 6B). FIG. 6A shows the results of a cysteine-free peptide (167-184) where diamonds, circles, triangles, and hatched squares correspond to 100, 200, 300 and 400 μml respectively. FIG. 6B shows results using the thiol-containing peptide (167-R173C-184); doses and symbols are the same as in FIG. 6A. FIG. 6C shows reduction of cytochrome C (no phospholipids) with X/Xo (squares); triangles, X/Xo plus the apoA-IMilano 167-R173C-184 peptide (400 μg/ml); circles, X/Xo plus the apoA-IMilano 167-184 peptide (400 μg/ml). Note the synthetic peptides failed to protect cytochrome C indicating that the thiol-containing apoA-IMilano peptide is unable to directly quench ROS in the aqueous phase. SOD (superoxide dismutase) was used as a control.

[0119] Interaction of apoA-IMilano peptide 167-R173C-184 with GSH is shown in FIG. 7A. Squares show the oxidation of PL-micelles with lipoxygenase; diamonds, circles, and triangles correspond to oxidation in the presence of GSH alone (100 μM), peptide alone (200 μg/ml) and peptide plus GSH, respectively. FIG. 7A indicates that the apoA-IMilano peptide mimetic (SEQ ID NO: 1) was able to interact synergistically with reduced GSH to inhibit lipoxygenase-mediated lipid peroxidation. In the absence of thiol compound, lipoxygenase caused a rapid induction of lipid peroxidation (squares). The presence of reduced glutathione (0.1 mM, diamonds) was unable to effectively inhibit lipid peroxidation compared to incubations with the peptide mimetic of apoA-IMilano (200 μg/ml, circles). However, the combination of glutathione plus the peptide (triangles) provided even greater protection against oxidation compared to peptide alone.

[0120] FIG. 7B desmonstrates LCAT activation using a standard proteoliposome substrate composed of peptides: egg-yolk PC:unesterified cholesterol (15:250:12.5 mole ratios). Results are expressed as a percentage of activation obtained with apoA-IWT. The synthetic peptide based on apoA-IParis (SEQ ID NO: 9) was able to activate the ernzyme LCAT while the apoA-IMilano based peptide (SEQ ID NO: 1) failed in this regard (FIG. 7B). This is probably related to the cluster of three positively charged arginine residues (149, 153 and 160) associated with helix 6 (aa 145-162) of apoA-I that have been shown to play a role in LCAT activation. This series of arginine residues is present in the peptide based on apoA-IParis but it is absent in the peptide based on apoA-IMilano.

[0121] Both peptides based on apoA-IMilano and apoA-IParis were unable to stimulate cholesterol efflux from J774 macrophages as shown in FIG. 7C. FIG. 7C: Cholesterol efflux from J774 macrophages; squares, serum free medium (□); circles, apoA-IMilano peptide (167-R173C-184, SEQ ID NO: 1) (◯); triangles (behind squares), apoA-IParis peptide (145-R151C-162, SEQ ID NO: 9) (Δ); diamonds, apoA-IWT (⋄).

EXAMPLE 15

Antioxidant Activity of Synthetic Peptide Mimetics Based on Helix 10 of apoA-I

[0122] In FIG. 8A, PL micelles were exposed to lipoxygenase (5 U/μL) in the absence (squares) and presence of a synthetic peptide (220-E224C-237) based on helix 10 of apoA-I; diamonds, circles and triangles correspond to 100, 200, 300 μg/mL. For comparative purposes, the ability of peptide 167-R173C-184 to inhibit lipoxygenase-mediated lipid peroxidation is shown in FIG. 8B. Squares represent oxidation of phospholipid in the absence of peptide; diamonds, circles, triangles, and half-darkened squares correspond to 100, 200, 300, and 400 μg/ml of the apoA-IMilano based peptide, respectively. Note that both peptides inhibited lipid peroxidation over the same relative dose range indicating that incorporation of a cysteine residue within helix 10 of apoA-I, which is a Class Y amphipathic alpha helix, is able to confer antioxidant activity like the peptide mimetic of apoA-IMilano.

EXAMPLE 16

Antioxidant Activity of Synthetic Peptide Mimetics of apoE3

[0123] Antioxidant activity of synthetic peptide, GADMEDVCGRLVQYRGEV (SEQ ID NO: 47), based on helix 3 of apolipoprotein E3 (apoE3) is shown in FIG. 9 . Phospholipid micelles were exposed to xanthine/xanthine oxidase (X/Xo, 20 U/ml) in the absence of peptide (squares, panels A & B). FIG. 9A shows the results of a cysteine-free peptide (105-122) based on the apolipoproteinE4 (apoE4) isoform where diamonds, circles, triangles, and hatched squares correspond to 100, 200, 300 and 400 μg/ml. . Note the control peptide (105-122, SEQ ID NO: 48) derived from apoE4 did not inhibit oxidation. FIG. 9B shows results using the thiol-containing peptide (105-R112C-122 , SEQ ID NO: 47) based on apoE3; doses and symbols are the same as in FIG. 9A. In contrast to the peptide based on apoE4, peptide 105-R112C-122 based on apoE3 inhibited oxidation in a dose dependent manner similar to the peptides based on apoA-IMilano and apoA-IParis. FIG. 9C shows reduction of cytochrome C (no phospholopids) with X/Xo (squares); triangles X/Xo plus the apoE4 peptide (SEQ ID NO: 48) (400 μg/ml); circles, X/Xo plus the apoE3 peptide (SEQ ID NO: 47) (400 μg/ml). Note the synthetic peptides failed to protect cytochrome C indicating that the thiol-containing peptide (SEQ ID NO: 47) was unable to directly quench ROS in the aqueous phase. The asterisks denote the control SOD (superoxide dismutase).

[0124] The present examples, methods, procedures, treatments, specific compounds and molecules are meant to exemplify and illustrate the invention and should in no way be seen as limiting the scope of the invention. Any patents or publications mentioned in this specification are indicative of levels of those skilled in the art to which the patent pertains and are hereby incorporated by reference to the same extent as if each was specifically and individually incorporated by reference. The SEQUENCE LISTING accompanying this specification is also hereby incorporated by reference in its entirety.

Claims

1. A peptide comprising a sequence selected from the following group SDELRQRLAARLEALKN (SEQ ID NO: 2); GEEMRDRARAHVDALRTH (SEQ ID NO: 10); PVLESFKVSFLSALEEYT (SEQ ID NO: 17); LKLLDNWDSVTSTFSKLR (SEQ ID NO: 33); and PALEDLRQGLLPVLESFKVSFLSALEEYTKKLN (SEQ ID NO: 31); PALEDLRQGLLP LKLLDNWDSVTSTFSKLR (SEQ ID NO: 40), wherein said peptide has one to four cysteine residues substituted for the amino acids shown in bold and underlined.

2. The peptide of claim 1, having from 18 to 100 amino acids, wherein the amino acid sequence is at least 80% homologous with the corresponding native Apolipoprotein A-I human protein sequence as set forth in GenBank Accession Number P02647.

3. A peptide comprising the sequence of GADMEDVRGRLVQYRGEV (SEQ ID NO: 48), wherein said peptide has one to four cysteine residues substituted for the amino acids shown in bold and underlined.

4. The peptide of claim 3, having from 18 to 100 amino acids, wherein the amino acid sequence is at least 80% homologous with the corresponding native Apolipoprotein E3 human protein sequence as set forth in GenBank Accession Number P02649.

5. A peptide comprising a sequence selected from the following group ARLSRGVQVLSRKLTLKA (SEQ ID NO: 53); ARLSRGVQVLSRKLTLKAKALHARIQQNLDQLREEL (SEQ ID NO: 59); and ATLKDSLEODLNNMNKFLEKLR (SEQ ID NO: 65), wherein said peptide has one to four cysteine residues substituted for the amino acids shown in bold and underlined.

6. The peptide of claim 5, having from 18 to 100 amino acids, wherein the amino acid sequence is at least 80% homologous with the corresponding native Apolipoprotein A-V human protein sequence as set forth in GenBank Accession Number NP—44320.

7. A peptide comprising a sequence selected from the following group ETGDLWVGCHP (SEQ ID NO: 71); ETGDLWVGCHPNGMKIFFYDSEN (SEQ ID NO: 72); LKSLDFNTLVDNISVDP ETGDLWVGCHPNGMKIFFYDSEN (SEQ ID NO: 73).

8. The peptide of claim 7, having from 18 to 100 amino acids, wherein the amino acid sequence is at least 80% homologous with the corresponding native Human Serum Paraoxonase as set forth in GenBank Accession Number XP—043694.

9. A peptide comprising the sequence selected from the following group DWLKAFYDKVAEKLKEAF (SEQ ID NO: 75); LEKLNSCLRDRLSALTDTPLEELRDSLRSRLDALRST (SEQ ID NO: 82); and LEKLNSSLRDRLSALTDT (SEQ ID NO: 84); wherein said peptide has one to four cysteine residues substituted for the amino acids shown in bold and underlined.

10. The peptide of claim 9, having from 18 to 100 amino acids, wherein the amino acid sequence is at least 80% homologous with the corresponding native Apolipoprotein A-I human protein sequence as set forth in GenBank Accession Number P02647.

11. A method of making an anti-oxidant peptide, comprising the steps of: (a) identifying an amphipathic helix in a Human HDL-associated protein, said helix having between 10 and 100 amino acids and further having a hydrophobic side and a hydrophilic side when viewed axially through the helix; (b) modifying at least one residue near the amphipathic interface from the naturally occurring amino acid to a cysteine residue to create a modified helix peptide; (c) selecting a modified helix peptide that has at least twice the antioxidant activity as the unmodified peptide; and (d) synthesizing said peptide in sufficient quantity for antioxidant use.

12. The method of claim 11, wherein the Human HDL-associated protein is selected from the group consisting of apoAI, apoE3, apo AV and paroxonase.

13. The method of claim 11, wherein the anti-oxidant activity is measured by the ability of the modified helix peptide to inhibit lipid peroxidation by soybean lipoxygenase.

14. The method of claim 13, wherein the antioxidant activity is further measured by the ability of the modified helix peptide to inhibit lipid peroxidation by xanthine oxidase and to (not) inhibit xanthine/xanthine oxidase mediated reduction of cytochrome C.

15. The method of claim 14, wherein the anti-oxidant activity is at least a 50% protection against maximum accumulation of lipid peroxides at a concentration of no more than 500 micrograms per mL.

15. The method of claim 13 wherein the peptide is selected from helix 1 (amino acids 44-65), helix 6 (amino acids 145-162) and helix 10 (amino acids 209-238) of apoAI, helix 7 (amino acids 167-184) of apoAI, the helix spanning amino acids 105-122 of apoE3, and amino acids 219-236 of apo AV.

16. A method for preventing oxidation of a phospholipid, comprising contacting the phospholipid with a peptide of claim 1.

17. The method of claim 16 further comprising the addition of a water soluble antioxidant.

18. The method of claim 17 where the water soluble antioxidant is selected from the group consisting of GSH, vitamin C, vitamin E and N-acetyl cysteine.

19. A method for preventing oxidation of a phospholipid, comprising contacting the phospholipid with a peptide of claim 3.

20. The method of claim 19 further comprising the addition of a water soluble antioxidant.

21. The method of claim 19 where the water soluble antioxidant is selected from the group consisting of GSH, vitamin C, vitamin E and N-acetyl cysteine.

22. A method for preventing oxidation of a phospholipid, comprising contacting the phospholipid with a peptide of claim 5.

23. The method of claim 22 further comprising the addition of a water soluble antioxidant.

24. The method of claim 23 where the water soluble anti-oxidant is selected from the group consisting of GSH, vitamin C, vitamin E and N-acetyl cysteine.

25. A method for preventing ocidation of a phospholipid, comprising contacting the phospholipid with a peptide if claim 9.

26. The method of claim 25 further comprising the addition of the water soluble antioxidant.

27. The method of claim 26 where the water soluble antioxisant is selected from the group consisting of GSH, vitamin C, vitamin E and N-acetyl cysteine.