US 5,652,332 AGrant

Biologically Active Peptides From Functional Domains of Bactericidal/Permeability-Increasing Protein and Uses Thereof

Issue Date:1997-07-29

•4 Claims

•78 Drawing Sheets

Abstract

The present invention provides peptides having an amino acid sequence that is the amino acid sequence of a human bactericidal/permeability-increasing protein (BPI) functional domain or a subsequence thereof, and variants of the sequence or subsequence thereof, having at least one of the BPI biological activities, such as heparin binding, heparin neutralization, LPS binding, LPS neutralization or bactericidal activity. The invention provides peptides and pharmaceutical compositions of such peptides for a variety of therapeutic uses.

Metadata

Assignee

XOMA

Inventor

Roger G. Little, II

Application Information

Application Number:US 3064735

Filing Date:1994-09-15

Priority Date:1993-03-12

Art Unit:181

Classifications

IPC:

A61K 3800C07K 500C07K 700C07K 1600

Field of Search:

51453012;13;14;15324;325;326;327;328;300

Patent Drawings (78 sheets)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1a and 1b depict HPLC absorbance spectra for cyanogen bromide and proteolytic fragments of rBPI.sub.23 ;

FIG. 2 is a graph of LAL inhibition assay results for proteolytic fragments of rBPI.sub.23 ;

FIG. 3 is a graph of a heparin binding assay results using 15-mer BPI peptides;

FIG. 4 is a graph of a Limulus Amoebocyte Lysate (LAL) inhibition assay results using 15-mer BPI peptides;

FIG. 5 is a graph of a radial diffusion bactericidal assay results using 15-mer BPI peptides;

FIG. 6 is a graph showing the effect of BPI functional domain peptides in a heparin binding assay;

FIGS. 7a and 7b are graphs showing the effects of BPI functional domain peptides on ATIII/heparin inhibition of thrombin;

FIGS. 8a and 8b are graphs showing the results of BPI functional domain peptides in an LAL inhibition assay;

FIGS. 9a, 9b, 9c, and 9d are graphs showing the results of BPI functional domain peptides in radial diffusion bactericidal assays;

FIGS. 9e and 9f are graphs showing the results of BPI functional domain peptides in E. coli broth assays;

FIGS. 10a, 10b, 10c, 10d and 10e are graphs showing the results of BPI functional domain combination peptides in radial diffusion bactericidal assays;

FIGS. 11a, 11b, 11c, 11d, 11e, 11f, 11g, 11h and 11i are graphs showing the results of BPI functional domain peptides in radial diffusion bactericidal assays;

FIGS. 11j and 11k are graphs showing the results of BPI functional domain peptides in bactericidal assays on bacterial cells growing in broth media;

FIG. 11l is a graph showing the results of BPI functional domain peptide BPI.30 in bactericidal assays performed in human serum;

FIGS. 11m and 11n are graphs showing the results of BPI functional domain peptides in radial diffusion bactericidal assays using Gram-positive bacteria;

FIG. 11o is a graph showing the results of BPI functional domain peptides in radial diffusion bactericidal assays in comparison with gentamicin and vancomycin using S. aureus cells;

FIGS. 11p and 11q are graphs showing the results of BPI functional domain peptides in cytotoxicity assays using C. albicans cells growing in broth media;

FIGS. 12a, 12b, 12c, 12d, 12e, 12f, and 12g are graphs showing the results of a heparin neutralization assay using BPI functional domain peptides;

FIG. 13 is a schematic diagram of the structure of BPI domain II peptide BPI.2 (amino acid sequence 85-99 of the BPI sequence, SEQ ID NO:7);

FIG. 14 is a schematic diagram of the structure of BPI domain III peptide BPI.1.sup.. 1 (amino acid sequence 148-161 of the BPI sequence, SEQ ID NO: 13);

FIGS. 15a, 15b, 15c, 15d and 15e are graphs showing the results of heparin binding assays using BPI functional domain substitution peptides;

FIG. 16 is a graph showing the results of heparin binding experiments using a variety of BPI functional domain peptides;

FIGS. 17a and 17b are graphs of the results of Lipid A binding competition assays between synthetic BPI functional domain peptides and radiolabeled rBPI.sub.23 ;

FIG. 18 is a graph of the results of Lipid A binding competition assays between synthetic BPI.10 peptide and radiolabeled rBPI.sub.23 in blood or phosphate buffered saline;

FIG. 19 is a graph of the results of Lipid A binding competition assays between synthetic BPI peptides BPI.7, BPI.29 and BPI.30 versus radiolabeled rBPI.sub.23 ;

FIGS. 20a and 20b are graphs of the results of Lipid A binding competition assays between BPI functional domain peptides and radiolabeled rLBP.sub.25 ;

FIG. 21 is a graph of the results of radiolabeled RaLPS binding experiments using BPI functional domain peptides pre-bound to HUVEC cells;

FIGS. 22a, 22b, 22c, 22d, 22e, 22f, 22g, and 22h are graphs showing the various parameters affecting a cellular TNF cytotoxicity assay measuring the LPS neutralization activity of BPI;

FIGS. 23a, 23b and 23c are graphs showing the dependence of NO production on the presence of .gamma.-interferon and LBP in LPS-stimulated RAW 264.7 cells and inhibition of such NO production using rBPI.sub.23 ;

FIGS. 24a, 24b, 24c, 24d, 24e, and 24f are graphs showing LPS neutralization by BPI functional domain peptides reflected in their capacity to inhibit NO production by RAW 264.7 cells stimulated by zymosan or LPS;

FIG. 24g is a graph showing the IC.sub.50 values of synthetic BPI peptides for inhibition of LPS- or zymosan-stimulated NO production by RAW 264.7 cells;

FIG. 25 is a schematic of rBPI.sub.23 showing three functional domains;

FIG. 26a is a graph showing the dependence of LPS-mediated inhibition of RAW 264.7 cell proliferation on the presence of rLBP;

FIGS. 26b and 26c are graphs showing patterns of BPI functional domain peptides using the assay of Example 20D;

FIG. 27 is a graph showing a comparison of TNF inhibition in whole blood by various BPI functional domain peptides using the assay of Example 20E; and

FIG. 28 is a graph showing the results of the thrombin clotting time assay described in Example 20G using various BPI functional domain peptides.

FIGS. 29(a-h) are graphs showing the results of BPI functional domain peptides the radial diffusion bactericidal assays described in Example 27.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

This invention provides peptides having an amino acid sequence that is the amino acid sequence of at least one functional domain or subsequence thereof and variants of the sequence or subsequence of BPI. For the purposes of this invention, the term "functional domain" is intended to designate a region of the amino acid sequence of BPI that contributes to the total biological activity of the protein. These functional domains of BPI are defined by the activities of proteolytic cleavage fragments, overlapping 15-mer peptides and other synthetic peptides.

Domain I is defined as the amino acid sequence of BPI comprising from about amino acid 17 to about amino acid 45. Peptides based on this domain are moderately active in both the inhibition of LPS-induced LAL activity and in heparin binding assays, and do not exhibit significant bactericidal activity. Domain II is defined as the amino acid sequence of BPI comprising from about amino acid 65 to about amino acid 99. Peptides based on this domain exhibit high LPS and heparin binding capacity and are bactericidal. Domain III is defined as the amino acid sequence of BPI comprising from about amino acid 142 to about amino acid 169. Peptides based on this domain exhibit high LPS and heparin binding activity and are bactericidal.

The functional domains as herein defined include the continuous domains I, II and III, i.e., domains comprised of a continuous portion of the BPI amino acid sequence. However, the invention also includes peptides comprising portions of BPI which are not continuous, i.e., that are separated in the BPI sequence. It is recognized that some non-continuous stretches of amino acid sequence may be folded in the native protein to make such amino acid regions contiguous or in proximity, which structure can be mimicked in the peptides of the invention by covalently linking together peptides from non-continuous regions.

Peptides containing non-continuous regions of BPI amino acid sequence are one example of combination peptides provided by the invention. For the purposes of this invention, combination peptides are intended to include linear, cyclized or branched-chain peptides comprised of two or more peptides having an amino acid sequence from the same or different functional domains of BPI and subsequences thereof. Specifically encompassed in this definition are combinations containing from two to about 10 functional domain peptides or subsequence thereof, preferably combinations of two or three functional domain peptides (for example, homodimers, homotrimers, heterodimers and heterotrimers). Each of the component peptides comprising such combinations may have an amino acid sequence from any particular BPI functional domain amino acid sequence or subsequence thereof.

For purposes of this invention, the term "a biological activity of BPI" is intended to include, but is not limited to the biological activities of a human bactericidal/permeability-increasing protein (BPI), including, for example, a recombinant BPI holoprotein such rBPI (SEQ ID NO:69), an amino-terminal fragment of BPI such as rBPI.sub.23, and mutated amino-terminal fragments of BPI such as rBPI.sub.21 .DELTA.cys (designated rBPI (1-193) ala.sup.132 in copending and co-assigned U.S. patent application Ser. No. 08/013,801, filed Feb. 2, 1993, and corresponding PCT application No. US94/01235 filed Feb. 2, 1994, incorporated by reference). As disclosed in copending and co-assigned U.S. patent application Ser. No. 08/093,202, incorporated by reference, rBPI has been produced having the sequence set out as SEQ ID NO:69 as shown in Gray et al. (supra) except that valine at position 151 is specified by GTG rather than GTC, and residue 185 is glutamic acid (specified by GAG) rather than lysine (specified by AAG). In addition, rBPI.sub.23 (see also, Gazzano-Santoro et al., 1992, Infect. Immun. 60:4754-4761) has been produced using an expression vector containing the 31-residue signal sequence and the first 199 amino acids of the sequence of rBPI with the exceptions from the Gray et al. (supra) sequence as noted above. Such biological activities include LPS binding, LPS neutralization, heparin binding and heparin neutralization, and bactericidal activity. Specifically included is a biological activity of any peptide of this invention that is between 0.1 and 10 times the activity of BPI or of a corresponding peptide encompassing a corresponding functional domain of BPI. Also expressly included in this definition of the "biological activity of BPI" is a biological activity, for example bactericidal activity, that is qualitatively different than the activity of BPI or the corresponding peptide encompassing the entire corresponding domain of BPI. For example, such qualitative differences include differences in the spectrum of bacteria or other microorganisms against which the peptide is effective, relative to the amino acid sequence of the corresponding functional domain of BPI. This definition thus encompasses peptide activities, such as bactericidal activity against Gram-positive bacteria and fungicidal activity, not previously reported for BPI.

The invention provides peptides each of which has an amino acid sequence that is the amino acid sequence of one of the functional domains of human BPI or a subsequence thereof. Embodiments of such peptides include the following exemplary domain I peptides {single-letter abbreviations for amino acids can be found in G. Zubay, Biochemistry (2d. ed.), 1988 (MacMillen Publishing: N.Y.), p.33}:

the following exemplary domain II peptides:

and the following exemplary domain III peptides:

It will be recognized that BPI.14, BPI.12 and BPI.55 are examples of addition variants.

The invention also provides linear and branched-chain combinations of the same or different peptides, wherein each of the peptides of the combination has an amino acid sequence that is the amino acid sequence of one of the functional domains of human BPI or a subsequence thereof. Embodiments of such peptides include the following exemplary combination domain II peptides:

and the following exemplary branched-chain domain II peptide:

MAP.1 (.beta.-alanyl-N.alpha.,N.epsilon.-substituted-{N.alpha.,N.epsilon.(BPI.2) lysyl}lysine);

and the following exemplary combination domain III peptide:

and the following exemplary branched-chain domain III peptide:

MAP.2 (.beta.-alanyl-N.alpha.,N.epsilon.-substituted-{N.alpha.,N.epsilon.(BPI.13 )lysyl}lysine);

and the following exemplary domain II-domain III interdomain combination peptides:

Amino acid substitution variants are also provided, wherein the amino acid residue at one or more positions in each of the peptides is a residue different from the amino acid found in the corresponding position of the BPI functional domain from which that specific peptide is derived. For example, in one embodiment of this aspect of the invention, one position in the peptide is substituted with an alanine residue for the amino acid found at the corresponding position in the BPI amino acid sequence. In other embodiments, one position in the peptide is substituted with e.g., a phenylalanine, leucine, lysine or tryptophan residue for the amino acid found at the corresponding position in the BPI amino acid sequence. Embodiments of these peptides include the following exemplary substitution domain II peptides:

and the following exemplary substitution domain III peptides:

A particular utility of such single amino acid-substituted BPI functional domain peptides provided by the invention is to identify critical residues in the peptide sequence, whereby substitution of the residue at a particular position in the amino acid sequence has a detectable effect on at least one of the biological activities of the peptide. Expressly encompassed within the scope of this invention are embodiments of the peptides of the invention having substitutions at such critical residues so identified using any amino acid, whether naturally-occurring or atypical, wherein the resulting substituted peptide has biological activity as defined herein.

Substituted peptides are also provided that are multiple substitutions, i.e., where two or more different amino acid residues in the functional domain amino acid sequence are each substituted with another amino acid. For example, in embodiments of such doubly-substituted peptides, both positions in the peptide are substituted e.g., with alanine, phenylalanine or lysine residues for the amino acid found at the corresponding positions in the BPI amino acid sequence. Examples of embodiments of these peptides include the multiply substituted domain II peptides:

and the exemplary multiply substituted domain III peptide:

and the following exemplary multiply substituted domain II substitution combination peptide:

and the following exemplary multiply substituted domain II-domain III interdomain substitution combination peptide:

Another aspect of such amino acid substitution variants are those where the substituted amino acid residue is an atypical amino acid. Specifically encompassed in this aspect of the peptides of the invention are peptides containing D-amino acids, modified or non-naturally-occurring amino acids, and altered amino acids to provide peptides with increased stability, potency or bioavailability. Embodiments of these peptides include the following exemplary domain II peptides with atypical amino acids:

the exemplary domain III peptide with atypical amino acids:

and the exemplary domain II-domain III interdomain combination peptides with atypical amino acids:

Linear and branched-chain combination embodiments of the amino acid substitution variant peptides, which create multiple substitutions in multiple domains, are also an aspect of this invention. Embodiments of these peptides include the following exemplary combination/substitution domain II peptides:

Dimerized and cyclized embodiments of each of the aforementioned BPI functional domain peptides are also provided by this invention. Embodiments of these peptides include the following exemplary cysteine-modified domain II peptides: ##STR1##

Thus, the invention includes novel chemical compounds which are peptides based upon or related to, domains I, II, and III of BPI, respectively identified as Group I, Group II, and Group III:

The invention includes peptides having a portion from different domains identified as Group IV:

BPI functional domain peptides described herein are useful as potent anti-bacterial agents for Gram-negative bacteria and for neutralizing the adverse effects of LPS associated with the cell membranes of Gram-negative bacteria. The peptides of the invention have, in varying amounts, additional activities of BPI, including activities not directly associated with the Gram-negative bacterial infection, such as heparin binding and neutralization. Peptides provided by this invention also may have biological activities distinct from the known biological activities of BPI. For example, some embodiments of the peptides of the invention surprisingly have been found to have a biological target range for bactericidal activity that is broader than BPI and exhibits bactericidal activity against Gram-positive as well as Gram-negative bacteria. Some embodiments of the invention have surprisingly been found to have fungicidal activity. Thus, the invention advantageously provides peptides having amino acid sequences of the biologically functional domains of BPI having distinct antimicrobial activities. Peptides of this invention that possess the dual anti-bacterial and anti-endotoxic properties of BPI, including those with an increased antibiotic spectrum, represent a new class of antibiotic molecules.

BPI functional domain peptides of the invention will have biological therapeutic utilities heretofor recognized for BPI protein products. For example, co-owned, copending U.S. patent application Ser. No. 08/188,221 filed Jan. 24, 1994, addresses use of BPI protein products in the treatment of humans exposed to Gram-negative bacterial endotoxin in circulation. For example co-owned, copending U.S. patent application Ser. No. 08/031,145 filed Mar. 12, 1993 and PCT/US94/02463 filed Mar. 11, 1994, addresses administration of BPI protein products for treatment of mycobacterial diseases. Co-owned, copending U.S. patent application Ser. No. 08/132,510, filed Oct. 5, 1993, addresses use of BPI protein products in the treatment of conditions involving depressed reticuloendothelial system function. For example co-owned, copending U.S. patent application Ser. No. 08/125,651, filed Sep. 22, 1993, addresses synergistic combinations of BPI protein products and antibiotics. For example co-owned, copending U.S. patent application Ser. No. 08/093,201 filed Jul. 14, 1993 and PCT/US94/07834 filed Jul. 14, 1994, addresses methods of potentiating BPI protein product bactericidal activity by administration of LBP protein products. The disclosures of the above applications are specifically incorporated by reference herein for the purpose of exemplifying therapeutic uses for BPI functional domain peptides of the invention. The BPI functional domain peptides of the invention also have therapeutic utility for the treatment of pathological conditions and disease states as disclosed in the above identified U.S. patent application Ser. Nos. 08/030,644, 08/093,202, 08/183,222 and 08/209,762 parent applications and corresponding PCT/US94/02465 filed Mar. 11, 1994.

BPI functional domain peptides of the invention are thus useful in methods for: neutralizing the anti-coagulant effect of heparin; inhibiting angiogenesis (especially angiogenesis associated with ocular retinopathy); inhibiting endothelial cell proliferation (especially endometriosis and proliferation associated with implantation of fertilized ova); inhibiting malignant tumor cell proliferation (especially Kaposi's sarcoma proliferation); treating chronic inflammatory disease states (such as arthritis and especially reactive and rheumatoid arthritis); treating Gram-negative bacterial infection and the sequelae thereof; treating the adverse effects (such as increased cytokine production) of Gram-negative endotoxin in blood circulation; killing Gram-negative bacteria; treating adverse physiological effects associated with depressed reticuloendothelial system function (especially involving depressed function of Kupffer cells of the liver such as results from physical, chemical and biological insult to the liver); treating, in synergistic combination with antibiotics (such as gentamicin, polymyxin B and cefamandole nafate) Gram-negative bacterial infection and the sequelae thereof; killing Gram-negative bacteria in synergistic combination with antibiotics; treating, in combination with LBP protein products, Gram-negative bacterial infection and the sequelae thereof; killing Gram-negative bacteria in combination with LBP protein products; treating, alone or in combination with antibiotics and/or bismuth, Mycobacteria infection (especially infection by M. tuberculosis, M. leprae and M. avium); treating adverse physiological effects (such as increased cytokine production) of lipoarabinomannan in blood circulation; decontaminating fluids (such as blood, plasma, serum and bone marrow) containing lipoarabinomannan; and, treating disease states (such as gastritis and peptic, gastric and duodenal ulcers) associated with infection by bacteria of the genus Helicobacter. The present invention also provides pharmaceutical compositions for oral, parenteral, topical and aerosol administration comprising BPI functional domain peptides in amounts effective for the uses noted above and especially compositions additionally comprising pharmaceutically acceptable diluents, adjuvants or carriers.

With respect to uses of BPI functional domain peptides in combination with LBP protein products, as used herein, "LBP protein product" includes naturally and recombinantly product lipopolysaccharide binding protein; natural, synthetic, and recombinant biologically active polypeptide fragments and derivatives of lipopolysaccharide binding protein; and biologically active polypeptide analogs, including hybrid fusion proteins, of either LBP or biologically active fragments thereof. LBP protein products useful according to the methods of the present invention include LBP holoprotein which can be produced by expression of recombinant genes in transformed eucaryotic host cells such as described in co-owned and copending U.S. patent application Ser. No. 08/079,510 filed Jun. 17, 1993, U.S. patent application Ser. No. 08/261,660 and corresponding PCT/US94/06931 both filed Jun. 17, 1994, and designated rLBP. Also described in that application are preferred LBP protein derivatives which lack CD14-mediated inflammatory properties and particularly the ability to mediate LPS activity through the CD14 receptor. Such LBP protein products are preferred for use according to the present invention because excessive CD14-mediated immunostimulation is generally considered undesirable, and is particularly so in subjects suffering from infection.

Preferred LBP protein derivatives are characterized as amino-terminal fragments having a molecular weight of about 25 kD. Most preferred are LBP amino-terminal fragments characterized by the amino acid sequence of the first 197 amino acids of the amino-terminus of LBP, as set out in SEQ ID NOS:97 and 98, designated rLBP.sub.25, the production of which is described in previously-noted co-owned and copending U.S. patent application Ser. Nos. 08/079,510, 08/261,660 and corresponding PCT/US94/06931. It is contemplated that LBP protein derivatives considerably smaller than 25 kD and comprising substantially fewer than the first 197 amino acids of the amino-terminus of the holo-LBP molecule are suitable for use according to the invention provided they retain the ability to bind to LPS. Moreover, it is contemplated that LBP protein derivatives comprising greater than the first 197 amino acid residues of the holo-LBP molecule including amino acids on the carboxy-terminal side of first 197 amino acids of the rLBP as disclosed in SEQ ID NOS: 97 and 98 will likewise prove useful according to the methods of the invention provided they lack an element that promotes CD14-mediated immunostimulatory activity. It is further contemplated that those of skill in the art are capable of making additions, deletions and substitutions of the amino acid residues of SEQ ID NOS: 97 and 98 without loss of the desired biological activities of the molecules. Still further, LBP protein products may be obtained by deletion, substitution, addition or mutation, including mutation by site-directed mutagenesis of the DNA sequence encoding the LBP holoprotein, wherein the LBP protein product maintains LPS-binding activity and lacks CD14-mediated immunostimulatory activity. Specifically contemplated are LBP hybrid molecules and dimeric forms which may result in improved affinity of LBP for bacteria and/or increased stability in vivo. These include LBP/BPI hybrid proteins and LBP-Ig fusion proteins. Such hybrid proteins further include those using human gamma 1 or gamma 3 hinge regions to permit dimer formation. Other forms of dimer contemplated to have enhanced serum stability and binding affinity include fusions with Fc lacking the CH.sub.2 domain, or hybrids using leucine or helix bundles.

BPI functional domain peptides of the invention may be generated and/or isolated by any means known in the art, including by means of recombinant production. Co-owned U.S. Pat. No. 5,028,530, issued Jul. 2, 1991, co-owned U.S. Pat. No. 5,206,154, issued Apr. 27, 1993, and co-owned, copending U.S. patent application Ser. No. 08/010,676, filed Jan. 28, 1993, all of which are hereby incorporated by reference, disclose novel methods for the recombinant production of polypeptides, including antimicrobial peptides. Additional procedures for recombinant production of antimicrobial peptides in bacteria have been described by Piers et al., 1993, Gene 134:7-13. Co-owned, copending U.S. patent application Ser. No. 07/885,501, filed May 19, 1992, a continuation-in-part thereof, U.S. patent application Ser. No. 08/072,063, filed May 19, 1993 and corresponding PCTFUS93/04752, which are hereby incorporated by reference, disclose novel methods for the purification of recombinant BPI expressed in and secreted from genetically transformed mammalian host cells in culture and discloses how one may produce large quantities of recombinant BPI suitable for incorporation into stable, homogeneous pharmaceutical preparations.

BPI functional domain peptides may also be advantageously produced using any such methods. Those of ordinary skill in the art are able to isolate or chemically synthesize a nucleic acid encoding each of the peptides of the invention. Such nucleic acids are advantageously utilized as components of recombinant expression constructs, wherein the nucleic acids are operably linked with transcriptional and/or translational control elements, whereby such recombinant expression constructs are capable of expressing the peptides of the invention in cultures of prokaryotic, or preferably enkaryotic cells, most preferably mammalian cells, transformed with such recombinant expression constructs.

Peptides of the invention may be advantageously synthesized by any of the chemical synthesis techniques known in the art, particularly solid-phase synthesis techniques, for example, using commercially-available automated peptide synthesizers. Such peptides may also be provided in the form of combination peptides, wherein the peptides comprising the combination are linked in a linear fashion one to another and wherein a BPI sequence is present repeatedly in the peptide, with or without separation by "spacer" amino acids allowing for selected conformational presentation. Also provided are branched-chain combinations, wherein the component peptides are covalently linked via functionalities in amino acid sidechains of the amino acids comprising the peptides.

Functional domain peptides of this invention can be provided as recombinant hybrid fusion proteins comprising BPI functional domain peptides and at least a portion of at least one other polypeptide. Such proteins are described, for example, by Theofan et al. in co-owned, copending U.S. patent application Ser. No. 07/885,911, filed May 19, 1992, a continuation-in-part application thereof, U.S. patent application Ser. No. 08/064,693, filed May 19, 1993 and corresponding PCT/US93/04754, which are incorporated herein by reference in their entirety.

Generally, those skilled in the art will recognize that peptides as described herein may be modified by a variety of chemical techniques to produce compounds having essentially the same activity as the unmodified peptide, and optionally having other desirable properties. For example, carboxylic acid groups of the peptide, whether carboxyl-terminal or sidechain, may be provided in the form of a salt of a pharmaceutically-acceptable cation or esterified to form a C.sub.1 -C.sub.16 ester, or converted to an amide of formula NR.sub.1 R.sub.2 wherein R.sub.1 and R.sub.2 are each independently H or C.sub.1 -C.sub.16 alkyl, or combined to form a heterocyclic ring, such as 5- or 6-membered. Amino groups of the peptide, whether amino-terminal or sidechain, may be in the form of a pharmaceutically-acceptable acid addition salt, such as the HCl, HBr, acetic, benzoic, toluene sulfonic, maleic, tartaric and other organic salts, or may be modified to C.sub.1 -C.sub.16 alkyl or dialkyl amino or further converted to an amide. Hydroxyl groups of the peptide sidechain may be converted to C.sub.1 -C.sub.16 alkoxy or to a C.sub.1 -C.sub.16 ester using well-recognized techniques. Phenyl and phenolic rings of the peptide sidechain may be substituted with one or more halogen atoms, such as fluorine, chlorine, bromine or iodine, or with C.sub.1 -C.sub.16 alkyl, C.sub.1 -C.sub.16 alkoxy, carboxylic acids and esters thereof, or amides of such carboxylic acids. Methylene groups of the peptide sidechains can be extended to homologous C.sub.2 -C.sub.4 alkylenes. Thiols can be protected with any one of a number of well-recognized protecting groups, such as acetamide groups. Those skilled in the art will also recognize methods for introducing cyclic structures into the peptides of this invention to select and provide conformational constraints to the structure that result in enhanced binding and/or stability. For example, a carboxyl-terminal or amino-terminal cysteine residue can be added to the peptide, so that when oxidized the peptide will contain a disulfide bond, thereby generating a cyclic peptide. Other peptide cyclizing methods include the formation of thioethers and carboxyl- and amino-terminal amides and esters.

Peptidomimetic and organomimetic embodiments are also hereby explicitly declared to be within the scope of the present invention, whereby the three-dimensional arrangement of the chemical constituents of such peptido-and organomimetics mimic the three-dimensional arrangement of the peptide backbone and component amino acid sidechains in the peptide, resulting in such peptido- and organomimetics of the peptides of this invention having substantial biological activity. It is implied that a pharmacophore exists for each of the described activities of BPI. A pharmacophore is an idealized, three-dimensional definition of the structural requirements for biological activity. Peptido- and organomimetics can be designed to fit each pharmacophore with current computer modelling software (computer aided drug design). The degree of overlap between the specific activities of pharmacophores remains to be determined.

The administration of BPI functional domain peptides is preferably accomplished with a pharmaceutical composition comprising a BPI functional domain peptide and a pharmaceutically acceptable diluent, adjuvant, or carrier. The BPI functional domain peptide composition may be administered without or in conjunction with known antibiotics, surfactants, or other chemotherapeutic agents. Examples of such combinations are described in co-owned, copending, U.S. patent application Ser. No. 08/012,360, filed Feb. 2, 1993, continuation-in-part U.S. patent application Ser. No. 08/190,869, filed Feb. 2, 1994 and corresponding PCT/US94/01239 filed Feb. 2, 1994, the disclosures of which are incorporated herein by reference.

Effective doses of BPI functional domain peptides for bactericidal activity, partial or complete neutralization of the anti-coagulant activity of heparin, partial or complete neutralization of LPS and other effects described herein may be readily determined by those of skill in the art according to conventional parameters, each associated with the corresponding biological activity, including, for example, the size of the subject, the extent and nature of the bacterial infection, the extent and nature of the endotoxic shock, and the quantity of heparin administered to the subject and the time since administration of the heparin. Similar determinations will be made by those of skill in this art for using the peptide embodiments of this invention for therapeutic uses envisioned and described herein.

Embodiments of the invention comprising medicaments can be prepared for oral administration, for injection, or other parenteral methods and preferably include conventional pharmaceutically acceptable carriers, adjuvents and counterions as would be known to those of skill in the art. The medicaments are preferably in the form of a unit dose in solid, semi-solid and liquid dosage forms such as tablets, pills, powders, liquid solutions or suspensions, and injectable and infusible solutions. Effective dosage ranges from about 100 .mu.g/kg to about 10 mg/kg of body weight are contemplated.

The Examples which follow are illustrative of specific embodiments of the invention, and various uses thereof. Example 1 describes the preparation of proteolytic fragments of BPI; Example 2 describes the results of bactericidal assays of the proteolytic fragments of Example 1; Example 3 describes the results of heparin binding assays using the proteolytic fragments of Example 1; Example 4 describes the results of experiments using Limulus amebocyte lysates to assay the LPS binding activity of the proteolytic fragments of Example 1; Example 5 describes the preparation of 15-mer peptides of BPI; Example 6 describes the results of heparin binding assays using the 15-mer peptides of Example 5; Example 7 describes the results of Limulus amebocyte lysates assays using the 15-mer peptides of Example 5; Example 8 describes the results of bactericidal assays of the 15-mer peptides of Example 5; Example 9 describes the preparation of BPI individual functional domain peptides; Example 10 describes the results of heparin binding assays using the BPI individual functional domain peptides of Example 9; Example 11 describes the results of heparin neutralization assays using the BPI individual functional domain peptides of Example 9; Example 12 describes the results of Limulus amebocyte lysates assays of LPS neutralization activity using the BPI individual functional domain peptides of Example 9; Example 13 describes the results of bactericidal assays of the BPI individual functional domain peptides of Example 9; Example 14 describes the preparation of BPI combination functional domain peptides; Example 15 describes the results of bactericidal activity assays of the BPI combination functional domain peptides of Example 14; Example 16 describes the results of additional bactericidal activity assays of the BPI combination functional domain peptides of Example 14; Example 17 describes the results of in vivo and in vitro heparin neutralization assays using the BPI combination functional domain peptides of Example 14; Example 18 describes the preparation and functional activity analysis of bactericidal activity, heparin binding activity and LPS neutralization activity assays of BPI substitution variant functional domain peptides; Example 19 provides a summary of the results of bactericidal and heparin binding assays using representative BPI functional domain peptides; Example 20 describes analysis of BPI functional domain peptides in a variety of binding and neutralization assays; Example 21 addresses a heparin neutralization assay; Example 22 describes administration of BPI functional domain peptides in model systems of collagen and bacteria-induced arthritis animal model systems exemplifying treatment of chronic inflammatory disease states; Example 23 illustrates testing of BPI functional domain peptides for angiostatic effects in a mouse malignant melanoma metastasis model system; Example 24 addresses effects of BPI functional domain peptides on endothelial cell proliferation; Example 25 describes analysis of BPI functional domain peptides in animal model systems; and Example 26 describes a protocol for testing the anti-endotoxin effects of BPI functional domain peptides of the invention in vivo in humans; Example 27 describes the administration of BPI functional domain peptides to test for their anti-microbial effects against antibiotic resistant strains.

EXAMPLE 1

Preparation of BPI Proteolytic Fragments

Chemical cleavage and enzymatic digestion processes were applied to rBPI.sub.23 to produce variously-sized proteolytic fragments of the recombinant BPI protein.

rBPI.sub.23 protein was reduced and alkylated prior to proteolysis by cyanogen bromide (CNBr) or endoproteinase Asp-N. The protein was alesalted by overnight precipitation upon the addition of cold (4.degree. C.) acetone (1:1 v/v) and the precipitated protein recovered by pelleting under centrifugation (5000.times.g) for 10 minutes. The rBPI.sub.23 protein pellet was washed twice with cold acetone and dried under a stream of nitrogen. An rBPI.sub.23 solution was then reconstituted to a final concentration of 1 mg protein/mL in 8 M urea/0.1 M Tris-HCl (pH 8.1) and reduced by addition of 3.4 mM dithiothreitol (Calbiochem, San Diego, Calif.) for 90 minutes at 37.degree. C. Alkylation was performed by the addition of iodoacetamide (Sigma Chemical Co., St. Louis, Mo.) to a final concentration of 5.3 millimolar and incubation for 30 minutes in the dark at room temperature. The reduced and alkylated protein was acetone-precipitated, centrifuged and washed as described above and the pellet was redissolved as described below for either CNBr or Asp-N digestion.

For CNBr-catalyzed protein fragmentation, the washed pellet was first dissolved in 70% trifluoroacetic acid (TFA) (Protein Sequencing Grade, Sigma Chemical Co., St. Louis, Mo.) to a final protein concentration of 5 mg/mL. Cyanogen bromide (Baker Analyzed Reagent, VWR Scientific, San Francisco, Calif.) dissolved in 70% TFA was added to give a final ratio of 2:1 CNBr to protein (w/w). This ratio resulted in an approximately 75-fold molar excess of CNBr relative to the number of methionine residues in the rBPI.sub.23 protein. The reaction was purged with nitrogen and allowed to proceed for 24 hours in the dark at room temperature. The reaction was terminated by adding 9 volumes of distilled water, and followed by freezing (-70.degree. C.) and lyophilization.

For endoproteinase digestion, the reduced and alkylated rBPI.sub.23 was solubilized at a concentration of 5.0 mg/mL in 8 M urea/0.1 M Tris-HCl (pH 8.1). An equal volume of 0.1 M Tris-HCl (pH 8.1) was then added so that the final conditions were 2.5 mg/mL protein in 5 M urea/0.1 M Tris-HCl (pH 8.1). Endoproteinase Asp-N from Pseudomonas fragi (Boehringer-Mannheim, Indianapolis, Ind.) was added at a 1:1000 (w/w, enzyme:substrate) ratio, and digestion was allowed to proceed for 6 hours at 37.degree. C. The reaction was terminated by addition of TFA to a final concentration of 0.1% and the samples were then fractionated by reverse phase HPLC.

The CNBr and Asp-N fragment mixtures were purified on a Zorbax Protein Plus C.sub.3 column (4.6.times.250 mm, 300.ANG. pore size, MACMOD Analytical Inc, Chadsford, Pa.). A gradient ranging from 5% acetonitrile in 0.1% TFA to 80% acetonitrile in 0.1% TFA was run over this column over a 2 hour elution period at a flow rate of 1.0 mL/min. Fragment elution was monitored at 220 nm using a Beckman System Gold HPLC (Beckman Scientific Instruments, San Ramon, Calif.). The column heating compartment was maintained at 35.degree. C. and the fractions were collected manually, frozen at -70.degree. C. and dried in a Speed Vac concentrator. Fragments were then solubilized in a solution of 20 mM sodium acetate (pH 4.0)/0.5 M NaCl prior to use.

Electrospray ionization mass spectrometry (ESI-MS) was performed on a VG Bio-Q mass spectrometer by Dr. Francis Bitsch and Mr. John Kim in the laboratory of Dr. Cedric Shackleton, Children's Hospital-Oakland Research Institute. Molecular masses were obtained by mathematical transformation of the data.

Although the DNA sequence for rBPI.sub.23 encodes amino acid residues 1-199 of the mature protein, a significant portion of the protein that is produced is truncated at Leu-193 and Val-195, as determined by ESI-MS. The existence of these carboxyl-terminal truncations were verified by isolating the carboxyl-terminal tryptic peptides, which were sequenced and analyzed by ESI-MS.

There are six methionine residues in the rBPI.sub.23 protein, at positions 56, 70, 100, 111, 170, and 196, and chemical cleavage by cyanogen bromide produced six major peptide fragments as predicted. The results of the CNBr cleavage experiments are summarized in Table I. The fragments were isolated by reverse phase (C.sub.3) HPLC (FIG. 1a) and their amino-terminal sequences were determined by Edman degradation. The two largest fragments (C1 and C5) were not resolved by the C.sub.3 HPLC column and further attempts to resolve them by ion exchange chromatography were unsuccessful, presumably because they are similar in length and isoelectric point. The identities of the C1, C5 fragments within the mixture were determined by ESI-MS. The predicted mass of C1 is 269 (Table I), taking into account the loss of 30 a.m.u. resulting from the conversion of the carboxyl-terminal methionine to homoserine during the CNBr cleavage reaction. The observed mass of 6251.51.+-.0.34 is consistent with the loss of a water molecule (18 a.m.u.) in a homoserine lactone intermediate, which may be favored over the formation of the homoserine because of the hydrophobicity of the C1 fragment C-terminal amino acids. The predicted mass of the C5 fragment is 6487 and the observed mass is 6385.84.+-.0.39 (Table I). For the C5 fragment, the C-terminal amino acids are hydrophilic, so the hydrolysis of the homoserine lactone intermediate is probably favored. From both the amino-terminal sequencing and the mass spectrum data, the C5 component represents approximately 10-25% of the material in the C1/C5 mixture.

Proteolytic cleavage with endoproteinase Asp-N was performed to provide additional fragments for the regions contained within the CNBr C1/C5 mixture. There are six aspartic acid residues within the rBPI.sub.23 sequence at positions 15, 36, 39, 57, 105, and 16. The six major Asp-N fragments isolated by C.sub.3 HPLC (FIG. 1b) were sequenced and masses were determined by ESI-MS (Table I). A short duration digest at a 1:1000 (w/w, enzyme:substrate) ratio was used to eliminate potential non-specific cleavages, particularly at glutamic acid residues. It is evident that this digestion did not continue until completion, as one fragment (1-38) was isolated where Asp residues (amino acids 15 and 35) were not cleaved. The mass spectra of the Asp-N fragments were consistent with the predicted masses for each individual fragment. Unlike the CNBr cleavage, where the carboxyl-terminal fragment was poorly resolved, the Asp-N fragment from amino acid 116 to the carboxyl-terminus was well resolved from all of the other Asp-N fragments.

EXAMPLE 2

Bactericidal Effects of BPI Proteolytic Fragments

BPI proteolytic fragments produced according to Example 1 were screened for bactericidal effects using rough mutant E. coli J5 bacteria in a radial diffusion assay. Specifically, an overnight culture of E. coli J5 was diluted 1:50 into fresh tryptic soy broth and incubated for 3 hours at 37.degree. C. to attain log phase growth of the culture. Bacteria were then pelleted at 3,000 rpm for 5 minutes in a Sorvall RT6000B centrifuge (Sorvall Instruments, Newton, Conn.). 5 mL of 10 mM sodium phosphate buffer (pH 7.4) was added and the preparation was re-pelleted. The supernatant was decanted and 5 mL of fresh buffer was added, the bacteria were resuspended and their concentration was determined by measurement of absorbance at 590 nm (an Absorbance value of 1.00 at this wavelength equals a concentration of 1.25.times.10.sup.9 CFU/mL in suspension). The bacteria were diluted to 4.times.10.sup.6 CFU/mL in 10 mL of molten underlayer agarose (at approximately 45.degree. C.) and inverted repeatedly to mix in 15 mL polypropylene tubes conventionally used for this purpose.

The entire contents of such tubes were then poured into a level square petri dish and distributed evenly by rocking the dish side-to-side. The agarose hardened in less than 30 seconds and had a uniform thickness of about 1 mm. A series of wells were then punched into the hardened agarose using a sterile 3 mm punch attached to a vacuum apparatus. The punch was sterilized with 100% alcohol and allowed to air dry prior to use to avoid contaminating the bacterial culture.

5 or 10 gL of each of the BPI fragments were carefully pipetted into each well. As a negative control, dilution buffer (pH 8.3) was added to a separate well, and rBPI.sub.23 at concentrations of 5 .mu.g/mL and 1 .mu.g/mL were also added as positive controls. Each plate was incubated at 37.degree. C. for 3 hours, and then 10 mL of molten overlayer agarose (at approximately 45.degree. C.) was added into the level petri dish, allowed to harden and incubated overnight at 37.degree. C. The next day, a clear zone was seen against the lawn of bacteria in those wells having bactericidal activity. In order to visually enhance this zone, a dilute Coomassie solution (consisting of 0.002% Coomassie Brilliant Blue, 27% methanol, 15% formaldehyde (37% stock solution) and water) was poured over the agar and allowed to stain for 24 hours. The bacterial zones were measured with a micrometer.

No bactericidal activity was discerned for the rBPI.sub.23 fragments generated by CNBr or by Asp-N digestion, when tested at amounts up to 25 pmol/well. In contrast, this assay detected measurable bactericidal activity using rBPI.sub.23 in amounts as low as 0.75 pmol/well. Reduced and alkylated rBPI.sub.23, on the other hand, also was not bactericidal at amounts up to 100 pmol/well, while alkylated rBPI.sub.23 retained bactericidal activity equivalent to rBPI.sub.23.

EXAMPLE 3

Heparin Bindine by BPI Proteolytic Fragments

rBPI.sub.23 and the BPI proteolytic fragments produced according to Example 1 were evaluated in heparin binding assays according to the methods described in Example 1 in copending U.S. patent application Ser. No. 08/093,202, filed Jul. 15, 1993 and incorporated by reference. Briefly, each fragment was added to wells of a 96-well microliter plate having a polyvinylidene difluoride membrane (Immobilon-P, Millipore, Bedford, Mass.) disposed at the bottom of the wells. Heparin binding of CNBr fragments was estimated using 100 picomoles of each fragment per well with a saturating concentration of .sup.3 H-heparin (20 .mu.g/mL). Positive control wells contained varying amounts of rBPI.sub.23. The wells were dried and subsequently blocked with a 0.1% bovine serum albumin (BSA) in phosphate buffered saline, pH 7.4 (blocking buffer). Dilutions of .sup.3 H-heparin (0.03-20 .mu.Ci/ml, avg. M.W.=15,000; DuPont-NEN, Wilmington, Del.) were made in the blocking buffer and incubated in the BPI peptide-containing wells for one hour at 4.degree. C. The unbound heparin was aspirated and the wells were washed three times with blocking buffer, dried and removed for quantitation in a liquid scintillation counter (Model 1217, LKB, Gaithersburg, Md.). Although BSA in the blocking buffer did show a low affinity and capacity to bind heparin, this was considered physiologically irrelevant and the background was routinely subtracted from the test compound signal. The specificity of fragment-heparin binding was established by showing that the binding of radiolabeled heparin was completely inhibited by a 100-fold excess of unlabeled heparin (data not shown).

The results, shown in Table II (as the mean values of duplicate wells + the range between the two values), indicated that the CNBr fragments containing the amino acids 71-100 (C3) and 1-56 and 112-170 (C1,5) bound heparin to a similar extent. The CNBr fragment 171-196 also bound more heparin than the control protein (thaumatin, a protein of similar molecular weight and charge to rBPI.sub.23).

The Asp-N fragments also demonstrated multiple heparin binding regions in rBPI.sub.23. As seen in Table II, the 57-104 Asp-N fragment bound the highest amount of heparin, followed by the 1-38 and 116-193 fragments. These data, in combination with the CNBr fragment data, indicate that there are at least three separate heparin binding regions within rBPI.sub.23, as demonstrated by chemically or enzymatically-generated fragments of rBPI.sub.23, with the highest heparin binding capacity residing within residues 71-100.

EXAMPLE 4

Effect of BPI Proteolytic Fragments on an LAL Assay

BPI proteolytic fragments produced according to Example 1 were subjected to a Limulus Amoebocyte Lysate (LAL) inhibition assay to determine LPS binding properties of these fragments. Specifically, each of the fragments were mixed in Eppendorf tubes with a fixed concentration of E. coli 0113 LPS (4 ng/mL final concentration) and incubated at 37.degree. C. for 3 hours with occasional shaking. Addition controls comprising rBPI.sub.23 at 0.05 .mu.g/mL were also tested. Following incubation, 360 .mu.L of Dulbecco's phosphate buffered saline (D-PBS; Grand Island Biological Co. (GIBCO), Long Island, N.Y.) were added per tube to obtain an LPS concentration of 200 pg/mL for the LAL assay. Each sample was then transferred into Immulon II strips (Dynatech, Chantilly, Va.) in volumes of 50 .mu.l per well.

Limulus amoebocyte Lysate (Quantitative Chromogenic LAL kit, Whitaker Bioproducts, Inc., Walkersville, Md.) was added at 50 .mu.L per well and the wells were incubated at room temperature for 25 minutes. Chromogenic substrate was then added at a volume of 100 .mu.L per well and was well mixed. After incubation for 20 to 30 minutes at room temperature, the reaction was stopped with addition of 100 .mu.L of 25% (v/v) acetic acid. Optical density at 405 nm was then measured in a multiplate reader (Model Vmax, Molecular Dynamics, Menlo Park, Calif.) with the results shown in FIG. 2 in terms of percent inhibition of LPS. In this Figure, the filled circle represents rBPI.sub.23 ; the open circle represents Asp-N fragment A3; the x represents Asp-N fragment A2; the filled square represents Asp-N fragment A4; the filled triangle represents Asp-N fragment A1A2; the open square represents Asp-N fragment A6a; the small open triangle represents CNBr fragment C3; and the small filled square represents CNBr fragment C1/C5.

The CNBr digest fraction containing amino acid fragments 1-56 and 112-170 inhibited the LPS-induced LAL reaction with an IC.sub.50 of approximately 100 nM. This IC.sub.50 is approximately 10-fold higher than the IC.sub.50 for intact rBPI.sub.23 (9 nM) in the same assay. The other CNBr digest fragments were found to be non-inhibitory.

A slightly different result was observed with fragments generated from the Asp-N digest, where three fragments were found to be inhibitory in the LAL assay. The fragment corresponding to amino acids 116-193 exhibited LAL inhibitory activity similar to intact rBPI.sub.23 with complete inhibition of the LPS-induced LAL reaction at 15 nM. The fragments corresponding to amino acids 57-104 and 1-38 also inhibited the LAL assay, but required 10-fold higher amounts. These results, in combination with the CNBr digest results, further supported the conclusion from previously-described experimental results that at least three regions of the rBPI.sub.23 molecule have the ability to neutralize LBS activation of the LAL reaction, with the most potent region appearing to exist within the 116-193 amino acid fragment.

Immunoreactivity studies of the proteolytic fragments of rBPI.sub.23 described in Example 1 were performed using ELISA assays. In such assays, a rabbit polyclonal antirBPI.sub.23 antibody, capable of blocking rBPI.sub.23 bactericidal and LAL inhibition properties, and two different, non-blocking mouse anti-rBPI.sub.23 monoclonal antibodies were used to probe the rBPI.sub.23 proteolytic fragments. The polyclonal antibody was found to be immunoreactive with the 116-193 and 57-104 Asp-N fragments and with the 1-56 and 112-170 CNBr fragments, while the murine monoclonal antibodies reacted only with an Asp-N fragment representing residues 1-14 of rBPI.sub.23.

EXAMPLE 5

Preparation of 15-mer Peptides of BPI

In order to further assess the domains of biological activity detected in the BPI fragment assays described in Examples 1-4, 15-mer synthetic peptides comprised of 15 amino acids derived from the amino acid sequence of the 23 kD amino terminal fragment of BPI were prepared and evaluated for heparin-binding activity, activity in a Limulus Amoebocyte Lysate Inhibition (LAL) assay and bactericidal activity. Specifically, a series of 47 synthetic peptides were prepared, in duplicate, each comprising 15 amino acids and synthesized so that each peptide shared overlapping amino acid sequence with the adjacent peptides of the series by 11 amino acids, based on the sequence of rBPI.sub.23 as previously described in copending U.S. patent application Ser. No. 08/093,202, filed Jul. 15, 1993. Peptides were simultaneously synthesized according to the methods of Maeji et al. (1990, Immunol. Methods 134:23-33) and Gammon et al. (1991, J. Exp. Med. 173:609-617), utilizing the solid-phase technology of Cambridge Research Biochemicals Ltd. under license of Coselco Mimotopes Ply. Ltd. Briefly, the sequence of rBPI.sub.23 (1-199) was divided into 47 different 15-mer peptides that progressed along the linear sequence of rBPI.sub.23 by initiating a subsequent peptide every fifth amino acid. This peptide synthesis technology allows for the simultaneous small scale synthesis of multiple peptides on separate pins in a 96-well plate format. Thus, 94 individual pins were utilized for this synthesis and the remaining two pins (B,B) were subjected to the same steps as the other pins without the addition of activated FMOC-amino acids. Final cleavage of the 15-mer peptides from the solid-phase pin support employed an aqueous basic buffer (sodium carbonate, pH 8.3). The unique linkage to the pin undergoes a quantitative diketopiperazine cyclization under these conditions resulting in a cleaved peptide with a cyclo(lysylprolyl) moiety on the carboxyl-terminus of each peptide. The amino-termini were not acetylated so that the free amino group could potentially contribute to anion binding reactions. An average of about 15 .mu.g of each 15-mer peptide was recovered per well.

EXAMPLE 6

Heparin Binding by 15-mer Peptides of BPI

The BPI 15-mer peptides described in Example 5 were subjected to a heparin binding assay according to the methods described in Example 3.

The results of these experiments are shown in FIG. 3, expressed as the total number of cpm bound minus the cpm bound by control wells which received blocking buffer only. These results indicated the existence of three distinct subsets of heparin-binding peptides representing separate heparin-binding functional domains in the rBPI.sub.23 sequence. In the BPI sequence, the first domain was found to extend from about amino acid 21 to about amino acid 55; the second domain was found to extend from about amino acid 65 to about amino acid 107; and the third domain was found to extend from about amino acid 137 to about amino acid 171. Material from the blank control pins showed no heparin binding effects.

EXAMPLE 7

Effect of 15-mer Peptides of BPI on an Limulus Amoebocyte Lysate (LAL) Assay

The 15-mer peptides described in Example 5 were assayed for LPS binding activity using the LAL assay described in Example 4.

The results of these experiments are shown in FIG. 4. The data in FIG. 4 indicated at least three major subsets of peptides representing three distinct domains of the rBPI.sub.23 protein having LPS-binding activity resulting in significant LAL inhibition. The first domain was found to extend from about amino acid 17 to about amino acid 55; the second domain was found to extend from about amino acid 73 to about amino acid 99; and the third domain was found to extend from about amino acid 137 to about amino acid 163. In addition, other individual peptides also exhibited LAL inhibition, as shown in the Figure. In contrast, material from blank control pins did not exhibit any LPS neutralizing effects as measured by the LAL assay.

EXAMPLE 8

Bactericidal Effects of 15-mer Peptides of BPI

The 15-mer peptides described in Example 5 were tested for bactericidal effects against the rough mutant strain of E. coli bacteria (J5) in a radial diffusion assay as described in Example 2. Products from the blank pins (B, B) were tested as negative controls.

The results of the assay are shown in FIG. 5. The only 15-mer peptide found to have bactericidal activity was a peptide corresponding to amino acids 85-99 of the BPI protein. As is seen in FIG. 5, the positive control wells having varying amounts of rBPI.sub.23 also showed bactericidal activity, while the buffer and blank pin controls did not.

The results of these bactericidal assays, along with the heparin binding and LAL assays described in the above Examples, indicate that there exist discrete functional domains in the BPI protein.

The results shown in Examples 1-8 above indicate that rBPI.sub.23 contains at least three functional domains that contribute to the total biological activity of the molecule. The first domain appears in the sequence of amino acids between about 17 and 45 and is destroyed by Asp-N cleavage at residue 38. This domain is moderately active in both the inhibition of LPS-induced LAL activity and heparin binding assays. The second functional domain appears in the region of amino acids between about 65 and 99 and its inhibition of LPS-induced LAL activity is diminished by CNBr cleavage at residue 70. This domain also exhibits the highest heparin binding capacity and contains the bactericidal peptide, 85-99. The third functional domain, between about amino acids 142 and 169, is active in the inhibition of LPS-induced LAL stimulation assay and exhibits the lowest heparin binding capacity of the three regions.

EXAMPLE 9

Preparation of BPI Individual Functional Domain Peptides

Based on the results of testing the series of overlapping peptides described in Examples 5 through 8, BPI functional domain peptides from each of the functionally-defined domains of the BPI protein were prepared by solid phase peptide synthesis according to the methods of Merdfield, 1963, J. Am. Chem. Soc. 8:5:2149 and Merrifield et al., 1966, Anal. Chem. 38:1905-1914 using an Applied Biosystems, Inc. Model 432 peptide synthesizer. BPI functional domain peptides were prepared having the amino acid sequences of portions of amino acid residues 1-199 of BPI as set out in Table III below and designated BPI.2 through BPI.5 and

EXAMPLE 10

Heparin Binding Activity by BPI Individual Functional Domain Peptides

BPI individual functional domain peptides BPI.2, BPI.3, and BPI.8, along with rBPI.sub.21 .DELTA.cys were assayed for heparin binding activity according to the methods described in Example 3. The results are shown in FIG. 6 and indicate that BPI.3 and rBPI.sub.21 .DELTA.cys had moderate heparin binding activity and BPI.2 and BPI.8 had little or no heparin binding activity.

EXAMPLE 11

Heparin Neutralization Activity of BPI Individual Functional Domain Peptides

BPI functional domain peptides BPI.2, BPI.3, BPI.4, BPI.5, BPI.6, and BPI.8, along with rBPI.sub.23 as a positive control, were assayed for their effect on thrombin inactivation by ATIII/heparin complexes according to the method of Example 3 in copending and co-assigned U.S. patent application Ser. No. 08/093,202, filed Jul. 15, 1993, incorporated by reference. Specifically, a Chromostrate.TM. anti-thrombin assay kit (Organon Teknika Corp., Durham, N.C.) was used to examine the inhibition of purified thrombin by preformed ATIII/heparin complexes in plasma.

Briefly, the assay was performed in 96 well microtiter plates in triplicate with a final volume per well of 200 .mu.L. Varying concentrations of the BPI functional domain peptides ranging from 1.0 .mu.g/mL to 100 .mu.g/mL were assayed to determine their effect on thrombin inhibition in the presence of pre-formed ATIII/heparin complexes. The order of addition of assay components was as follows: 1) a dilution series of rBPI.sub.23 or BPI functional domain peptides or thaumatin as a control protein, with final concentrations of 100, 50, 25, 10 and 1 .mu.g/well, diluted in PBS in a final volume of 50 .mu.L; 2) 50 .mu.L plasma diluted 1:100 in a buffer supplied by the manufacturer; 3) 50 .mu.l thrombin at 1 nKat/mL in a buffer supplied by the manufacturer; and 4) 50 .mu.L chromogenic substrate at a concentration of 1 .mu.mol/mL in water. The reaction was allowed to proceed for 10 minutes at 37.degree. C. and stopped with the addition of 50 .mu.L 0.1 M citric acid. The colorimetric reaction was quantitated on a microplate reader as described in Example 3.

The results of these assays are shown in FIGS. 7a and 7b, which depict the sample concentrations as weight or molar concentrations respectively. BPI functional domain peptides BPI.3 and BPI.5 each had the most significant heparin neutralization effects. In these assays, the control protein, thaumatin, showed no neutralizing effect and was essentially equivalent to the buffer control at all protein concentrations.

EXAMPLE 12

LPS Neutralization Activity by LAL Assay of BPI Individual Functional Domain Peptides

BPI functional domain peptides BPI.2, BPI.3, and BPI.8, along with rBPI.sub.23 as a positive control, were evaluated in the LAL assay according to the method of Example 4 herein to determine LPS binding and inhibition properties of these peptides. The experiments were performed essentially as described in Example 3 and the results are shown in FIGS. 8a and 8b, which depict the sample concentrations as weight or molar concentrations respectively. The results showed that BPI.3 had moderate LPS inhibition activity and that BPI.2 and BPI.8 had no significant LPS inhibition activity.

EXAMPLE 13

Bactericidal Activity Assay of BPI Individual

Functional Domain Peptides

BPI functional domain peptides BPI.2, BPI.3, and BPI.8, along with rBPI.sub.23 as a positive control, were tested for bactericidal effects against E. coli J5 (rough) and E. coli 0111:B4 (smooth) bacteria in a radial diffusion assay according to the methods of Example 2. The results of these assays are depicted in FIGS. 9a-9d. These results demonstrated that each of the BPI functional domain peptides BPI.2 and BPI.3 exhibited bactericidal activity while BPI.8 had little to no bactericidal activity. Each of the bactericidal peptides showing bactericidal activity tended to be more effective against the rough than the smooth E. coli strain.

In additional experiments, broth antibacterial assays were conducted to further determine the bactericidal activity of certain of the BPI peptides. Specifically, either E. coli J5 (rough) or E. coli 0111:B4 (smooth) bacteria were selected from single colonies on agar plates and used to inoculate 5 mL of Mueller Hinton broth and incubated overnight at 37.degree. C. with shaking. The overnight culture was diluted (.about.1:50) into 5 mL fresh broth and incubated at 37.degree. C. to log phase (.about.3 hours). Bacteria were pelleted for 5 minutes at 3000 rpm (1500.times.g). Bacterial pellets were resuspended in 5 mL PBS and diluted to 2.times.10.sup.6 cells/mL in the Mueller Hinton broth (wherein 1 OD.sub.570 unit equals 1.25.times.10.sup.9 CFU/mL). The BPI functional domain peptides to be tested were diluted to 200 .mu.g/mL in broth and serially diluted 2-fold in 96 well culture plates (100 .mu.L volume). All items were at 2-fold final concentration and experiments were conducted in triplicate. Bacteria were added at 100 .mu.L/well and the plates were incubated on a shaker at 37.degree. C. for a 20 hour period. The plates were then read on an ELISA plate multiple reader at 590 nm. One of the triplicate wells from each peptide concentration was selected for colony forming unit (CFU) determination. A 30 .mu.L aliquot was added to 270 .mu.L of PBS and further ten-fold serial dilutions were performed. Then a 50 .mu.L aliquot was plated on tryptic soy agar and incubated overnight. Colonies were counted and final bacterial concentrations determined. The results of these assays are depicted in FIGS. 9e (for E. coli J5) and 9f (for E. coli 0111:B4). As shown in these Figures, BPI functional domain peptide BPI.3 had significant anti-bacterial activity against E. coli J5 bacteria and less activity against E. coli 0111:B4 bacteria.

EXAMPLE 14

Preparation of BPI Combination Functional Domain Peptides

Combination peptides were prepared using solid-phase chemistry as described in Example 9. The sequences of these peptides are shown in Table IV. It will be noted that the peptides designated BPI.7, BPI.9 and BPI.10 represent partial or even multiple repeats of certain BPI sequences. Specifically, BPI.7 comprises a 20-mer consisting of amino acid residues 90-99 repeated twice in a single linear peptide chain. BPI.10 comprises an approximately 50:50 admixture of a 25-mer (designated BPI.10.1; SEQ ID NO:55) and a 26-mer (designated BPI.10.2; SEQ ID NO:65) consisting of amino acid residues 94-99, 90-99, 90-99 and 93-99, 90-99, 90-99, respectively, in a single linear peptide chain. BPI.9 comprises a 16-mer comprising amino acid residues 94-99 followed by residues 90-99 in a single linear peptide chain.

These peptides were used in each of the BPI activity assays described in Examples 10-13 above. In the heparin binding assay described in Example 10 and shown in FIG. 6, BPI.7 had extremely high heparin binding capacity. In the heparin neutralization assay described in Example 11 and shown in FIGS. 7a and 7b, BPI.7 had significant heparin neutralization effects compared with rBPI.sub.23. In the LAL assay described in Example 12 and shown in FIGS. 8a and 8b, BPI.7 had significant LPS inhibition properties. In bactericidal assays using radial diffusion plates as described in Example 13 and shown in FIGS. 9a-9d, each of the BPI functional domain peptides BPI.7, BPI.9 and BPI.10.1 and BPI.10.2 exhibited bactericidal activity, and significant bactericidal activity was also found for BPI.7, BPI.9 and BPI.10.1 and BPI.10.2 against both rough and smooth variant strains of E. coli in broth assays. The BPI.10 peptides exhibited the highest bactericidal activity observed against either bacterial strain.

These bactericidal activity results obtained with peptides BPI.7 and BPI.10 showed that a linear dimer (BPI.7) and a mixture of linear multimers (BPI.10.1 and BPI.10.2) of the BPI domain II peptide KWKAQKRFLK (i.e., BPI.8, SEQ ID NO:8) had bactericidal activity against E. coli strain J5, and that the monomer (BPI.8) showed essentially no bactericidal activity. Moreover, both the dimer and the multimer peptides had higher bactericidal activity that of BPI.9, comprising amino acids 94-99, 90-99. On the basis of these results, the additional peptides shown in Table IV were synthesized using the methods described in Example

EXAMPLE 15

Bactericidal Activity of Combination Functional Domain Peptide

The BPI combination functional domain peptides described in Example 14 were used in radial diffusion bactericidal assays essentially as described in Examples 2 and 13 above. These results are shown in FIGS. 10a-10e. The results shown in FIG. 10a demonstrate that BPI.8, comprising one copy of a domain II peptide (amino acids 90-99), had no detectable bactericidal activity against E. coli J5 cells at concentrations of 1000 .mu.g/mL. In contrast, BPI.13, comprising one copy of a domain III monomer (amino acids 148-161) showed appreciable bactericidal activity at concentrations greater than 30 .mu.g/mL. BPI.29, comprising two copies of a domain III monomer BPI.13, had greater bactericidal activity, and BPI.30, comprising a linear combination of the domain II peptide BPI.8 and the domain III peptide BPI.13, showed the highest bactericidal activity against J5 cells, approximating that of BPI.

FIG. 10b shows the results of experiments with domain II peptides comprising BPI.8, BPI.7 and BPI.10. (See also summary Table VIII.) Although BPI.8 showed no bactericidal activity against E. coli J5 cells at concentrations of 1000 .mu.g/mL, the combination peptides BPI.7 and BPI.10 showed high levels of bactericidal activity.

Additional experiments were performed using various other bacteria as target cells to examine the range of bactericidal killing of these BPI functional domain peptides. FIG. 10c shows the results of radial diffusion experiments using E. coli strain O7-K1. In these experiments, rBPI.sub.23 showed no bactericidal activity at concentrations of 100 .mu.g/mL, and low bactericidal activity even at concentrations of 1000 .mu.g/mL. Similarly low levels of bactericidal activity were found with the peptides BPI.8 comprising the domain II (DII) monomer and BPI.13 comprising the domain III (DIII) monomer, although the amount of activity of BPI.13 was found to be higher than that of rBPI.sub.23. Surprisingly, the domain II dimer BPI.7 and the domain II-domain III (DII-DIII) heterodimer BPI.30 showed high levels of bactericidal activity, and the domain III dimer BPI.29 showed moderate bactericidal activity. These results demonstrated that peptides of the functional BPI functional domain identified herein possess bactericidal activity qualitatively different from the bactericidal activity of the BPI molecule itself.

FIGS. 10d and 10e show results that further demonstrate that the homo- and heterodimers described herein have qualitatively and quantitatively different bactericidal activity spectra of susceptible bacteria. FIG. 10d shows the results of radial diffusion assays using Klebsiella pneurnoniae bacteria. The DII-DIII heterodimer BPI.30 showed the highest amount of bactericidal activity against this bacteria, the DIII homodimer BPI.29 showed moderate levels of activity, and the DII dimer (BPI.7) and DIII monomer (BPI.13) showed low levels of activity. BPI.8, comprising the DII monomer, showed no bactericidal activity at concentrations of 800 .mu.g/mL, consistent with the lack of bactericidal activity of this peptide seen with the E. coli swains tested.

FIG. 10e shows the levels of bactericidal activity found in radial diffusion experiments using the Gram-positive bacterium Staphylococcus aureus. The DII-DIII heterodimer BPI.30 showed the highest amount of bactericidal activity against this bacteria, the DIII homodimer BPI.29 showed moderate levels of activity, and DII dimer (BPI.7) and the DIII monomer (BPI.13) showed low levels of activity. BPI.8, comprising the DII monomer, showed no bactericidal activity at concentrations of 800 .mu.g/mL, consistent with the lack of bactericidal activity of this peptide seen with the other bacteria.

These results showed that the homo- and heterodimers disclosed herein possessed varying amounts of bactericidal activity, which varied both with regard to the amount of such activity and the minimum effective concentration of the peptide necessary for bactericidal activity to be detected. These results also showed that these peptides possessed quantitatively and, more surprisingly, qualitatively different bactericidal activity than the BPI itself.

EXAMPLE 16

Additional Bactericidal Activity of BPI Combination Functional Domain Peptides

In light of the results of the experiments disclosed in Example 15, the bactericidal activity of domain II-domain III combination peptides were compared with the bactericidal activity of each of the component BPI domain II and domain Ill peptides, against a number of different bacteria and other microorganisms. The following BPI functional domain peptides as described above were used in radial diffusion bactericidal assays (Example 2) and broth bactericidal assays (Example 13) essentially as described in Example 15 above. These results are shown in FIGS. 11a-11q. These Figures show results of bactericidal assays using the following bacterial strains:

The results of these experiments are summarized as follows. None of the BPI peptides tested showed any bactericidal activity against S. marcescens (FIG. 11f) or P. mirabilis (FIG. 11g). BPI.8 showed no bactericidal activity against any organism tested at concentrations up to about 2000 pmol. BPI.13 and BPI.30 showed bactericidal activity against P. aeruginosa (FIG. 11a), E. coli O18:K1:H7 (FIG. 11b), K. pneumoniae (FIG. 11c), and E. coli O75 (FIG. 11d). Additionally, BPI.30 showed bactericidal activity against S. typhurium (FIG. 11h), and, in broth assays, E. coli O86a:K51 (FIG. 11j) and E. coli O4:K12 (FIG. 11k). BPI.23 showed bactericidal activity in a radial diffusion assay against E. coli O86a:K61 (FIG. 11i). Additionally, BPI.30 showed bactericidal activity against E. coli O86a:K61 in human serum (FIG. 11l).

The bactericidal capacity of BPI peptides provided by the invention was also tested against Gram-positive bacteria (See Table VIII A). Surprisingly, every BPI peptide tested showed some bactericidal activity in radial diffusion assays using S. aureus (FIG. 11e), S. pneumoniae (FIG. 11m) and B. megaterium (FIG. 11n) at amounts ranging between about 20 and about 2000 pmol. These results compared favorably with bactericidal activity of the antibiotics gentamicin and vancomycin (FIG. 11o). In addition, peptides were tested for their activity against L-forms of gram-positive bacteria, such as the L-form of S. aureus ATCC 19640. BPI.13, BPI.10, BPI.48, and BPI.120 are representative compounds active against these L-form bacteria.

Most surprisingly, one peptide, BPI.13, was found to have fungicidal activity in a broth assay using C. albicans (FIGS. 11p and 11q). As shown in these Figures, the activity of BPI.13 is clearly distinguishable from the much lower activity levels of BPI.2, BPI.29, BPI.30, and BPI.48. As shown in Table VIII B other representative compounds were shown to be active against C. albicans. These results demonstrate that the BPI functional domain peptides of the invention have antimicrobial activity qualitatively distinct from the activity previously reported for native BPI.

EXAMPLE 17

Heparin Neutralization Activity of BPI Combination

Functional Domain Peptides

The in vitro and in vivo heparin neutralization capacity of the BPI combination functional domain peptides prepared in Example 14 was determined by assaying the ability of these peptides to counteract the inhibitory effect of heparin on clotting time of heparinized blood and plasma.

In vitro, the effect of BPI combination functional domain peptides was determined on heparin-mediated lengthening of activated partial thrombin time (APTT). The APTT is lengthened by the presence of endogenous or exogenous inhibitors of thrombin formation, such as therapeutically administered heparin. Thus, agents which neutralize the anti-coagulant effects of heparin will reduce the APTT measured by the test. Citrated human plasma (200 .mu.L) was incubated for 1 minute at 37.degree. C. with either 15 .mu.L of diluent (0.15 M NaCl, 0.1 M Tris-HCl, pH 7.4) or 15 .mu.L of the diluent also containing 25 .mu.g/mL heparin (187 units/mg). Various concentrations (from 0.0 to 56 .mu.g/mL) of rBPI.sub.23, rBPI.sub.21 .DELTA.cys, or BPI combination peptides BPI.29 (the DIII homodimer) and BPI.30 (heterodimer DII+DIII) in a volume of 15 .mu.L were added, followed immediately by 100 .mu.L of thrombin reagent (Catalog No. 845-4, Sigma Chemical Co., St. Louis, Mo.). Clotting time (thrombin time) was measured using a BBL Fibrometer (Becton Dickenson Microbiology Systems, Cockeysville, Md.). The results are shown in FIGS. 12a, 12b and 12e. FIG. 12a shows the relative decrease caused by addition of varying amounts of rBPI.sub.23 or rBPI.sub.21 .DELTA.cys to the heparin-prolonged APTT. These results establish that each of these BPI-related proteins inhibits the heparin-mediated lengthening of APTT. FIG. 12b shows that the BPI combination peptides BPI.29 and BPI.30 also inhibit the heparin-mediated lengthening of APTT. FIG. 12e illustrates the results obtained with BPI.30 on a non-log scale. FIG. 12g shows that BPI.29, BPI.30, and BPI.7 have the greatest effect on the clotting time of heparinized blood in the assay. BPI.3 and rBPI.sub.23 show a smaller effect, and BPI.14, BPI.2, BPI.4, BPI.5, BPI.7, and rLBP.sub.25, rBPI and rBPI.sub.21 .DELTA.cys all show less of a decrease in clotting times of heparinized blood in this assay.

The in vivo effect of exemplary BPI combination peptides on APTT in heparinized rats was determined and compared with the in vivo effect of rBPI.sub.23. APTr is lengthened by the presence of endogenous or exogenous inhibitors of thrombin formation, such as therapeutically administered heparin. Agents which neutralize the anti-coagulant effects of heparin will reduce the APTT as measured by this test. Sprague-Dawley rats housed under NIH guidelines were administered with 100 U/kg heparin by bolus intravenous injections via the animals' tail vein followed 5 minutes later by administration of varying amounts of test or control protein as compared with rBPI.sub.23. The APTT was then determined from blood samples collected from the abdominal aorta 2 minutes after the administration of the test or control protein. The APTT of untreated animals, as well as animals treated only with a BPI peptide, was also determined. FIG. 12c shows the dose dependence of rBPI.sub.23 inhibition of heparin-mediated lengthening of partial thromboplastin time, and that administration of about 5 mg/kg results in a APTT of the heparinized and BPI-treated animals that is almost the same as the untreated control animals. The results of similar experiments shown in FIG. 12d demonstrate that the unrelated protein thaumatin has no effect on APTT times in heparinized animals. The administration of BPI.10 peptide results in a APTT in heparinized animals that is essentially the same as the APTT in control animals treated with BPI.10 alone. Similar results using BPI.30 were also obtained (FIG. 12f).

These results show that BPI functional domain combination peptides (e.g., BPI.10 and BPI.30) and rBPI.sub.23 effectively neutralize heparin inhibition of coagulation proteases. Based on these characteristics, BPI combination functional domain peptides of the invention are projected to be useful in the clinical neutralization of heparin anti-coagulant effects in dosages generally corresponding functionally to those recommended for protamine sulfate, but are not expected to possess the severe hypotensive and anaphylactoid effects of that material.

EXAMPLE 18

Preparation and Functional Activity Analysis of BPI Substitution Variant Functional Domain Peptides

The results obtained above with peptides from functional domains II and III prompted a further effort to determine the functionally-important amino acid residues within these peptides. Accordingly, a series of peptides comprising the amino acid sequences of domains II and III were prepared in which one of the amino acids in the sequence was substituted with an alanine residue. Diagrams of the domain peptides used in the substitution experiments are shown in FIG. 13 (domain II; IKISGKWKAQKRFLK, SEQ ID No.:7) and FIG. 14 (domain III; KSKVGWLIQLFHKK, SEQ ID No.:13). These peptide series were then tested for heparin binding affinity (K.sub.d), heparin binding capacity (Hep-CAP), LPS neutralization as determined using the Limulus Ameboctye Lysate assay (LAL), and bactericidal activity against E. coli J5 using the radial diffusion assay (RDA), each assay as performed as described in the Examples above.

The results, shown in Table V (domain II) and Table VI (domain III), are expressed in terms of the fold difference in activity in each of these assays (except for the LAL assay where relative differences are noted) between the BPI functional domain II and domain III peptides and each alanine substituted variant peptide thereof.

For domain II peptides, most alanine-substituted peptides showed an approximately 2- to 10-fold reduction in bactericidal activity in the radial diffusion assay. Exceptions to this overall pattern include BPI.19 (Gly.sub.89 .fwdarw.Ala.sub.89), BPI.22 (Lys.sub.92 .fwdarw.Ala.sub.92), BPI.23 (Gln.sub.94 .fwdarw.Ala.sub.94) and BPI.24 (Lys.sub.95 .fwdarw.Ala.sub.95). In contrast, most alanine-substituted peptides showed no difference in the LAL assay; BPI.17 (Ile.sub.87 .fwdarw.Ala.sub.87) and BPI.21 (Trp.sub.91 .fwdarw.Ala.sub.91) showed a moderate and large decrease in activity, respectively, in this assay. For BPI.21, these results were consistent with the more than 10-fold reduction in bactericidal activity found for this peptide, indicating that amino acid 91 (a tryptophan residue in the native sequence) may be particularly important in conferring biological activity on the peptide.

The effect of alanine substitution on heparin binding and capacity was, in almost all cases, no more than 2-fold more or less than the unsubstituted peptide. One exception was the heparin binding capacity of BPI.21, which was 4-fold lower than the unsubstituted peptide. This further supports the earlier results on the particular sensitivity of the various activities of these peptides to substitution at Trp.sub.91. In most cases, the effect on both the K.sub.d of heparin binding and heparin binding capacity was consistent and of about the same magnitude. In some instances, the heparin binding capacity of the substituted peptide decreased, although the K.sub.d increased slightly (BPI.18; Ser.sub.88 .fwdarw.Ala.sub.88), or decreased slightly (BPI.24). There were also instances where capacity was unchanged even though the K.sub.d increased (BPI.20; Lys.sub.90 .fwdarw.Ala.sub.90) or decreased (BPI. 19). In one instance the affinity remained unaffected and the capacity decreased almost 2-fold (BPI.25; Arg.sub.96 .fwdarw.Ala.sub.96).

These results indicated the existence of at least one critical residue in the domain II sequence (Trp.sub.91), and that the activities of the domain II peptides were for the most part only minimally affected by alanine substitution of the other domain II amino acid residues.

For domain III peptides, most alanine-substituted peptides showed an approximately 2- to 5-fold reduction in bactericidal activity in the radial diffusion assay. Exceptions to this overall pattern include BPI.35 (Gly.sub.152 .fwdarw.Ala.sub.152), BPI.39 (Gln.sub.156.fwdarw.Ala.sub.156), BPI.42 (His.sub.159 .fwdarw.Ala.sub.159) and BPI.44 (Lys.sub.161 .fwdarw.Ala.sub.161). Most alanine-substituted peptides showed no difference in the LAL assay; BPI.31 (Lys.sub.148 .fwdarw.Ala.sub.148), BPI.32 (Ser.sub.149 .fwdarw.Ala.sub.149), BPI.33 (Lys.sub.150 .fwdarw.Ala.sub.150), and BPI.34 (Val.sub.151 .fwdarw.Ala.sub.151) showed a moderate decrease in LPS-binding activity, and BPI.36 (Trp.sub.153 .fwdarw.Ala.sub.153) and BPI.40 (Leu.sub.157 .fwdarw.Ala.sub.157) showed a large decrease in LPS-binding activity in this assay. For both BPI.36 and BPI.40, these results were consistent with the approximately 5-fold reduction in bactericidal activity found for these peptides, indicating that the hydrophobic amino acids Trp .sub.153 and Leu.sub.157 in the native sequence may be particularly important in conferring biological activity on the peptide.

Effects of alanine substitution on heparin binding and capacity were of similar magnitude, being no more than about 5-fold more or less than the unsubstituted peptide. In almost every case, the type of effect of alanine substitutions on both the K.sub.d of heparin binding and heparin binding capacity was consistent and of about the same magnitude, unlike the findings with the domain II alanine substitution peptides. In one instance (BPI.42; His.sub.159 .fwdarw.Ala.sub.159), the heparin binding capacity was unaffected although the K.sub.d declined slightly (1.2-fold). In only one instance was the K.sub.d of heparin binding and heparin capacity increased slightly (BPI.35; Gly.sub.152 .fwdarw.Ala.sub.153); an increase of only 10% was found.

Like the results found with the domain II alanine-substitution peptides, these results indicated the existence of at least one critical residue in the domain III sequence (Trp.sub.153), and possibly at least one other (Leu.sub.157). The results also showed that, unlike the domain II alanine-substituted peptides, almost one-half of the substitutions resulted in at least a 2-fold difference in the activities tested. In 6 cases, all four of the tested activities decreased, and in 10 instances bactericidal activity, the K.sub.d of heparin binding and heparin capacity decreased. In only one instance (BPI.35, Gly.sub.152 .fwdarw.Ala.sub.152) was the activity in the bactericidal, heparin binding K.sub.d and heparin capacity assays found to have increased, albeit slightly.

These results indicate that alanine replacement of the hydrophobic amino acid residues Trp.sub.91, and Leu.sub.157 have the greatest effect on the activities of these BPI functional domain substitution peptides. This result is unexpected in light of the cationic nature of rBPI.sub.23. In fact, domain II alanine substitution peptides in which lysine is replaced either by alanine or phenylalanine showed dramatic increases in activity (e.g., BPI.24, BPI.73).

As Table VI B illustrates, substitution of the tryptophan (Trp.sub.153) did not affect fungicidal activity, although it appeared to be crucial for bactericidal activity. Glutamine appears to play a critical role in fungicidal activity as demonstrated by a greater than 8 fold decrease in the radial diffusion assay test upon replacement (BPI.39, Gln.sub.156 .fwdarw.Ala.sub.156).

EXAMPLE 19

Summary of Biological Activity of BPI Functional Domain Peptides

The distribution of the peptides into construct categories is presented in Table VII below.

The BPI functional domain peptides of this invention, or representative subsets thereof, have been assayed for the following biological activities: bactericidal activity against Gram-negative and Gram-positive bacteria, and against certain other microorganisms; LPS binding and neutralization activities; and heparin binding and heparin neutralization activities.

BPI functional domain peptides were assayed for bactericidal activity on E. coli J5 bacteria and for heparin binding as described in Examples 8 and 6, respectively. The assay results for exemplary peptides of the present invention are summarized in Table VIII A for the Gram-negative bacteria E. coli J5 (rough) and E. coli O113 (smooth) and the Gram-positive bacteria S. aureus. The bactericidal activities are expressed as the amount of peptide (pmol/well and .mu.g/well) required to generate a 30 mm.sup.2 bactericidal zone.

An intriguing relationship was observed among representative BPI functional domain peptides when a multiple regression analysis was done using bactericidal activity as the predicted variable and heparin binding capacity and affinity (K.sub.d) as the predictor variables. This analysis revealed that only heparin binding capacity was significantly related to bactericidal activity (heparin capacity, p=0.0001 and heparin affinity, p=0.6007). In other words, the amount of heparin that a given peptide embodiment can bind at saturation (i.e. capacity) has a significant relationship with bactericidal activity and not how soon a given peptide reaches 50% saturation in the heparin titration (i.e. affinity). From the data on LPS binding competition and neutralization, it also appears that capacity is most predictive of bactericidal activity. For examples, the results demonstrate that BPI.7, BPI.29, BPI.30, BPI.46, BPI.47, BPI.48, BPI.63, BPI.65 (reduced), BPI.69, BPI.73, BPI.58, MAP.1 and MAP.2 have extremely high heparin capacity and also are highly bactericidal. Multiple antigenic peptides (MAP peptides) are multimeric peptides on a branching lysine core as described by Posnett and Tam, 1989, Methods in Enzymology 178:739-746. Conversely, BPI.2, BPI.4, BPI.8, BPI.14, BPI.53 and BPI.54 have low heparin binding capacity and accordingly have little or no bactericidal activity.

BPI interdomain combination peptides BPI.30 (comprising domain II-domain III peptides) and BPI.74 (comprising domain III-domain II peptides) were compared for bactericidal activity against Gram-negative and Gram-positive bacteria, and for heparin binding and capacity. These results surprisingly showed that inverting the order of the peptides in the combination changed the relative activity levels observed. For example, BPI.74 was found to have greatly reduced bactericidal activity compared with BPI.30. Specifically, BPI.74 had 10-fold lower bactericidal activity against E. coli J5 bacteria, 50-fold lower bactericidal activity against E. coli O111:B4 bacteria, and 3.5-fold lower bactericidal activity against S. aureus. A 2-fold reduction in heparin binding capacity and a 2-fold increase in heparin affinity, was also observed.

Other bactericidal and endotoxin binding proteins were examined for heparin binding activity. Cecropin A, magainin II amide, Polymyxin B peptide and Limulus anti-LPS factor (LALF) were assayed in the direct heparin binding assay described in Example 3. The magainin II amide (Sigma, St. Louis, Mo.) exhibited the highest heparin binding capacity (437.7 ng heparin/2 .mu.g peptide, K.sub.d =3.17 .mu.M) relative to cecropin A (Sigma, 242 ng/2 .mu.g, K.sub.d =395 nM), LALF (Assoc. of Cape Cod, Woods Hole, Mass., 195.3 ng/2 .mu.g peptide, K.sub.d =1.29 .mu.M), and PMB peptide (Bathem Biosciences, Philadelphia, Pa, 58.0 ng/2 .mu.g peptide, K.sub.d =309 mM). The magainin II amide is a substitution variant of the natural magainin sequence, where 3 alanines have been substituted at positions 8, 13, 15. The magainin II amide is reported to have less hemolytic activity than the natural magainin sequence.

The above results support the relationship between heparin binding, LPS binding and bactericidal activities demonstrated by the BPI peptide data and suggest that other LPS binding proteins will also bind to heparin. The more active bactericidal proteins, cecropin A and magainin II amide, correspondingly, have the highest heparin binding capacity of this series of other LPS binding proteins.

One type of BPI functional domain peptide addition variant incorporates the addition of D-alanine-D-alanine to either the amino- or carboxyl-terminus of a BPI functional domain peptide. The rational for this approach is to confer greater Gram-positive bactericidal activity with the addition of D-alanine. The cell wall biosynthesis in Gram-positive bacteria involves a transpeptidase reaction that specifically binds and utilizes D-alanine-D-alanine. Beta-lactam antibiotics such as the penicillins effectively inhibit this same reaction. Incorporation of D-alanine-D-alanine onto an active bactericidal peptide should target the peptide to the actively growing cell wall of Gram-positive bacteria.

In the domain II substitution series of BPI functional domain peptides, an unexpected increase was observed when Lys.sub.95 was substituted by alanine (BPI.24). A subsequent phenylalanine substitution at position 95 (BPI.73) resulted in improved activity compared with the alanine substitution species. Surprisingly, substitution at position 95 with D-Phe (BPI.76) resulted in dramatically reduced activity, to levels lower than the original peptide (BPI.2). This isomer effect demonstrates that the interactions of this peptide is stereospecific, and implies that BPI.73 can adopt a more active conformation compared with BPI.76. Such stereospecificity, particularly after the phenomenon has been investigated at other residues, provides an important determinant for pharmacophore development.

Peptides derived from the functional domains of BPI as defined herein have been utilized to determine that the hydrophobic amino acids (especially tryptophan) are most critical for optimal activity. This finding was unexpected due the cationic nature of BPI. In fact, for domain II, when a lysine is replaced by an alanine or phenylalanine, the activity increases dramatically (BPI.24, BPI.73). Combinations of functional domain peptides can also increase the potency of individual peptide constructs, including combinations of the most active substitution peptides from the three domains.

The purity of each newly synthesized peptide was determined by analytical reverse-phase HPLC using a VYDAC C-18 column (25 cm.times.4.6 mm, 5 .mu.m particle size, 30 nm pore size; Separation Group, Hesperia, Calif.). HPLC was performed using 5% acetonitrile/0.1% trifluoroacetic acid (TFA) in water as mobile phase A, and 80% acetonitrile/0.065% TFA as mobile phase B. The eluate was monitored spectrophotometrically at 220 nm. The flow rate was 1.0 mL/min. Gradient elution conditions were selected to give optimum resolution for each peptide. Purity was expressed as the percentage that the main peak area contributed to the total peak area (see Table X). Purity and identity of the new synthesized peptides were also determined by electrospray ionization mass spectrometry using a VG Biotech Bio-Q mass spectrometer. Table X presents a summary of the purity analyses of exemplary peptides of the invention by mass spectroscopy and HPLC.

BPI.13, as well as other selected peptides, were purified using a semi-preparative reverse-phase VYDAC C-18 column (25 cm.times.10 mm, 10 .mu.m particle size, 30 nm pore size). The following gradient was used to purify BPI.13: 26.7%B to 33%B/30 min. at a flow rate of 2.0 mL/min. BPI.13 was dissolved in mobile phase A at a concentration of 8.8 mg/mL and injected in a volume in 0.5 mL. Three separate injections were made and the main peak from each injection was collected. The collected material was combined and evaporated to dryness using a SpeedVac.

The purity of the recovered material (which will be referred to as BPI. 13P, for purified) was determined with the analytical reverse-phase system and gradient elution conditions described above. Based on this analysis, BPI.13P was 98% pure. Purity and identity of BPI.13P was also determined by electrospray ionization mass spectometry using a VG Biotech Bio-Q mass spectrometer. The observed moleuclar mass was 1711.0 (the predicted mass was 1711.1). No impurities were detected by mass spectrometry. Recovery of BPI.13P was 55%, assuming that the desired peptide constituted 69% of the starting material.

When peptides of the invention were further purified, as described above, the magnitude of the tested biological activity of the peptides, e.g., BPI.13P and BPI.30P, were found to increase when chemical purity was increased. This indicated that the observed biological activity was due to the peptide itself. In particular, the completely novel and unexpected antifungal activity of BPI.13 against Candida albicans (see Example 16), with a purity of about 69%, was further increased when the purity of the peptide preparation was increased to 98%.

EXAMPLE 20

Analysis of BPI Functional Domain Peptides using Binding and Neutralization Assays

A. LPS Binding Assays

BPI functional domain peptides were subjected to LPS binding assays.

The first of these assays was performed as described in Gazzano-Santoro et al., supra. Briefly, a suspension of E. coli strain J5 Lipid A was sonicated and diluted in methanol to a concentration of 0.2 .mu.g/mL, and then 50 .mu.L aliquots were adsorbed to wells (Immulon 2 Removawell Strips, Dynatech). Following overnight incubation at 37.degree. C., the wells were blocked with 215 .mu.L of a solution of D-PBS/0.1% BSA for 3 hr at 37.degree. C. Thereafter, the blocking buffer was discarded, the wells were washed with a solution of 0.05% Tween-20 in D-PBS (D-PBS/T) and incubated overnight at 4.degree. C. with 50 .mu.L of a solution of [.sup.125 I]-rBPI.sub.23 in D-PBS/T (a total of 234,000 cpm at a specific activity of 9.9 .mu.Ci/.mu.g) and increasing concentrations of BPI functional domain peptides. After this incubation, the wells were washed three times with D-PBS/T and the bound radioactivity counted using a gamma counter. Binding to wells treated with D-PBS/BSA was considered non-specific background binding and was subtracted from the total radioactivity bound in each well to yield the amount of specifically-bound radioactivity.

The results of these experiments are shown in FIGS. 17a (where the concentration of each peptide is given in nM) and 17b (the identical results, with the concentration of peptide given in .mu.g/mL). Competition experiments using unlabeled rBPI.sub.23 are shown for comparison. These results demonstrate that all of the tested peptides have some capacity to compete with rBPI.sub.23 for LPS binding, to differing degrees.

This experiment was repeated, comparing the LPS binding affinity of BPI.10 with rBPI.sub.23, using twice the amount of [.sup.125 I]-rBPI.sub.23 (a total of 454,000 cpm, specific activity 10 .mu.Ci/.mu.g) and in the presence or absence of whole blood. These results are shown in FIG. 18, and demonstrate that, on a molar basis, BPI.10 is within a factor of 2 as potent as rBPI.sub.23 in competing with radiolabeled rBPI.sub.23 in this assay.

The experiment was repeated using peptides BPI.7, BPI.29 and BPI.30, as in the first experiment described above except that a total of 225,000 cpm of [.sup.125 I]-rBPI.sub.23 was used and Lipid A was plated at a concentration of 0.5 mg/mL. The results of this experiment are shown in FIG. 19, and show that, on a molar basis, these peptides are 6 to 10-fold less potent that unlabeled rBPI.sub.23 in binding Lipid A.

A second binding assay was developed, wherein radiolabeled recombinant LPS binding protein ([.sup.125 I]-rLBP) was used instead of radiolabeled rBPI.sub.23 in competition experiments with BPI functional domain peptides BPI.2, BPI.3, BPI.4, BPI.5, BPI.7, BPI.13, BPI.14, BPI.29, BPI.30 and BPI.48. rBPI, rBPI.sub.21 .DELTA.cys, and rLBP.sub.25 were included in these assays as controls. In these experiments, Lipid A was adsorbed to the wells at a concentration of 0.7 .mu.g/mL in methanol. Incubation of radiolabeled rLBP (a total of 650,000 cpm and a specific activity of 3.45 .mu.Ci/.mu.g) was performed for 2.5 hr at 37.degree. C. in the presence of BPI peptides in a series of increasing concentrations. These results are shown in FIGS. 20a and 20b. IC.sub.50 values (i.e., the concentration at which Lipid A binding of radiolabeled rLBP.sub.25 is inhibited to one half the value achieved in the absence of the peptide) are shown in accompanying Table XI.

In a third binding assay, a number of BPI functional domain peptides were tested for their ability to bind to radiolabeled LPS following incubation with human endothelial cells (HUVEC). This assay measures the ability to bind LPS once the BPI peptides are bound to HUVEC cells. HUVEC cells were incubated in the presence of various BPI peptides at a concentration of either 1 .mu.g/mL or 3 .mu.g/mL for 3 hr at 4.degree. C. in 500 .mu.L of a solution of D-PBS/BSA. Following this incubation, the cells were washed twice with ice-cold D-PBS/BSA and then incubated for an additional 2.5 hr at 4.degree. C. in 500 .mu.L of a solution of [.sup.125 I]-RaLPS (a total of 340,000 cpm at a specific activity of 4.6.times.10.sup.6 cpm/.mu.g) in DPBS/BSA. The wells were washed three times with D-PBS/BSA, solubilized in 500 .mu.L of 1 M NaOH and the lysates counted using a gamma counter. These results, shown in FIG. 21, indicate that BPI.29 and BPI.30 retain the capacity to bind LPS while bound to HUVEC cells.

B. LPS Neutralization Screening Assay of BPI Functional Domain Peptides using TNF Cellular Toxicity

A screening assay for LPS neutralization was developed using a tumor necrosis factor (TNF) cellular toxicity assay. A human monocytic cell line (THP-1; accession number TIB202, American Type Culture Collection, Rockville, Md.) grown in media supplemented with Vitamin D produce TNF upon stimulation with LPS in a dosedependent fashion. Mouse fibroblasts (L929 cells; ATCC No.: CCL1) are sensitive to TNF-mediated cell killing, and this cell killing is also dose-dependent. Thus, the extent of cell killing of L929 cells provides a sensitive assay for the degree of TNF induction in THP-1 cells, which in turn is a sensitive indicator of the amount of free LPS in contact with the THP-1 cells. LPS binding and neutralization by BPI functional domain peptides or rBPI.sub.23 reduces the amount of free LPS in contact with THP-1 cells, which reduces the amount of TNF produced, which in turn reduces the amount of L929 cell killing in a standardized assay. Thus, the following assay provides a sensitive method for assessing the LPS binding and neutralization capacity of the BPI functional domain peptides of this invention.

THP-1 cells were grown in RPMI media (GIBCO, Long Island, N.Y.) supplemented with 10% FCS and Vitamin D in spinner culture for 3 days to a density of about 150,000 cells/mL. Cells were then plated in a round-bottomed 96-well culture plate at a density of 100,000 cells/well and incubated in RPMI media without Vitamin D or FCS in the presence of 5 ng/mL E. coli 01113 LPS for 6 hr at 37.degree. C. Experimental control wells also contained varying amounts of rBPI.sub.23 or BPI functional domain peptides, in concentrations varying from about 0.1 .mu.g/mL to about 100 .mu.g/mL. After this incubation, the plates were centrifuged at about 600.times.g to pellet the cells, and 50 gL of the supernatant were added to a 96-well flat bottomed culture dish prepared in parallel with 50,000 L929 cells per well in 50 .mu.L RPMI/10% FCS.

L929 cells were prepared by monolayer growth in RPMI/10% FCS media to a density of about 1 million cells per dish, then split 1:2 on the day before the experiment and allowed to grow overnight to about 70% confluence on the day of the experiment. Actinomycin D was added to the 70% confluent culture to a final concentration of 1 .mu.g/mL 20 min prior to plating in 96-well plates. L929 cell plates were incubated in the presence of the THP-1 supernatant for about 16 hr (overnight) at 37.degree. C. under standard conditions of mammalian cell growth. To each well was then added 20 .mu.L of a solution prepared by diluting 100 .mu.L of phenazine methylsulfonate in 2 mL CellTitre 96.TM.AQ.sub.ueous solution (Promega, Madison, Wis.), containing 3-[(4,5-dimethyl)-thiozol-2-yl]-5-(3-carboxymethoxyphenyl)-2-(4-sulfonyl)- 2H-tetrazolium (inner salt). The cultures were allowed to incubate for 2-4 hr at 37.degree. C. and then analyzed spectrophotometrically to determine the optical absorbance at 490 nm (A490). Experimental results were evaluated relative to a semilog standard curve prepared with known amounts of TNF, varying from about 10 ng/mL to about 10 mg/mL.

The results of these experiments are shown in FIGS. 22a-22h. FIG. 22a shows the relationship between A490 and TNF concentration in cultures of L929 cells in the presence and absence of 5 ng/mL LPS. These results show about the same linear relationship between A490 and concentration of TNF whether or not LPS was present in the assay media. FIG. 22b illustrates an experiment where TNF was incubated with L929 cells in the presence of increasing amounts of heparin. These results show a constant and characteristic A490 for TNF at concentrations of 1 ng/mL and 0.1 ng/mL, indicating that heparin does not affect L929 cell killing by TNF. FIG. 22c illustrates a control experiment, showing that rBPI.sub.21 .DELTA.cys decreased the amount of TNF-mediated L929 cell killing when incubated at the indicated concentrations in cultures of THP-1 cells in the presence of 5 ng/mL LPS. FIG. 22d shows that heparin could compete with LPS for binding with rBPI.sub.21 .DELTA.cys, by inhibiting the BPI-mediated inhibition of LPS-stimulated TNF production by THP-1 cells, as measured by the L929 cell killing assay.

FIG. 22e is a standard curve of A490 versus TNF as a measure of TNF-mediated L929 cell killing; FIG. 22g shows the linearity of the standard curve in a semilog plot over a TNF concentration range of about three logs (about 1000-fold). FIG. 22f shows the THP-1 cell dependence of the assay, wherein detectable amounts of TNF were most readily produced using about 100,000 THP-1 cells and LPS at a concentration of at least 5 ng/mL. Finally, FIG. 22h shows that the assay was found to be dependent on THP-1 cell production of TNF in response to LPS stimulation; human histiocytic lymphoma cells (U937; ATCC No.: CRL1593) produced no detectable TNF when substituted in the assay for THP-1 cells.

This assay was used to analyze LPS binding and neutralization capacity of a number of BPI functional domain peptides of the invention. These results are shown in Table XlI, and indicate that each of the peptides tested had the capacity to inhibit LPS-stimulated TNF production in THP-1 cells, as measured by TNF-mediated L929 cell killing.

C. LPS Neutralization Screening Assay of BPI Functional Domain

Peptides using a Cellular NO Production Assay

An additional LPS neutralization screening assay for BPI functional domain peptides was developed using an assay for NO production in mouse cells treated with LPS (see Lorsbach et al., 1993, J. Biol. Chem. 268:1908-1913). In this assay, mouse RAW 264.7 cells (ATCC Accession No. T1B71) were treated with bacterial LPS. The cells were incubated in 96-well plates and stimulated for 2 hours with E. coli O113 LPS or zymosan, in the presence or absence of .sub.-- -interferon, rLBP, fetal bovine serum (FBS) or normal human serum (NHS), or rBPI.sub.21 .DELTA.cys. After this incubation, the cells were washed with fresh media and incubated overnight in media containing 10% FCS. The NO released from the cells accumulated in the media and spontaneously converted to nitrite. This nitrite was assayed in situ by the Griess reaction, as follows. The nitrite was reacted with the primary amine of an added sulfanilamide and formed a diazonium salt. This salt was then reacted with added naphthylethylenediamine to form a red azo-dye. The Griess reaction was performed at room temperature in about 10 minutes. The amount of produced NO was estimated from a standard curve of Griess reaction products determined spectrophotometrically as Absorbance at a wavelength of 550 nm.

The results of this assay are shown in FIGS. 23a to 23c. FIG. 23a shows the dependence of NO production on the presence of .gamma.-interferon. This interferon effect was found to saturate at a concentration of 100 U/mL. FIG. 23b shows the dependence of LPS-stimulated NO production on the presence of LBP, either added as purified recombinant protein or as a component of FBS or NHS supplements of the cell incubation media. FIG. 23c shows rBPI.sub.23 -mediated inhibition of LPS-stimulated NO production, having an IC.sub.50 of 30-100 ng/mL. These results demonstrated that this assay is a simple, inexpensive and physiologically-relevant assay system for assessing the LPS-neutralizing activity of BPI and BPI functional domain peptides disclosed herein.

The results of such assays performed with BPI functional domain peptides are shown in FIGS. 24a-24g wherein the background production of NO by unstimulated cells is designated as "NO LPS". FIGS. 24a and 24b show inhibition of NO production stimulated by zymosan and LPS, respectively, by rBPI, rBPI.sub.21 .DELTA.cys and rLBP.sub.25. No inhibition of zymosan-stimulated NO production was seen at any concentration of BPI protein (FIG. 24a). In contrast, LPS-stimulated NO production was inhibited in a concentration-dependent manner by incubation with these rBPI-related proteins (FIG. 24b). FIG. 24c (zymosan) and FIG. 24d (LPS) shows the effects on NO production by RAW 264.7 cells of incubation with BPI.2, BPI.3, BPI.4, BPI.7 and BPI. 14; rBPI.sub.21 .DELTA.cys is also shown for comparison. As shown with native BPI, zymosan-stimulated NO production was not inhibited by incubation with any of the BPI functional domain peptides (with the possible exception of a small amount of inhibition by BPI.7 at high concentrations; FIG. 24c). LPS-stimulated NO production, on the other hand, was inhibited efficiently by rBPI.sub.21 .DELTA.cys, and to a lesser degree by BPI.3 and BPI.7 (FIG. 24d).

This experiment was repeated using BPI.5, BPI.13, BPI.29 and BPI.30, with rBPI.sub.21 .DELTA.cys analyzed in parallel for comparison. Zymosan-stimulated NO production by RAW 264.7 cells was found to be inhibited by BPI.30 at high (.sub.-- 100 .mu.g/mL) concentrations; neither any of the other BPI functional domain peptides nor rBPI.sub.21 .DELTA.cys showed any inhibition of zymosan-stimulated NO production (FIG. 24e). LPS-stimulated NO production was inhibited efficiently by rBPI.sub.21 .DELTA.cys, and to varying and lesser degrees by all of the BPI functional domain peptides tested in this experiment (FIG. 24f).

The IC.sub.50 values (i.e., the concentration of inhibitor at which zymosan or LPS-stimulated NO production by RAW 264.7 cells is reduced to one-half its value in the absence of the inhibitor) for the BPI proteins and peptides were calculated from these experiments and are showed in FIG. 24g. With the exception of BPI.30, no significant inhibition of zymosan-mediated NO production was found for either the BPI functional domain peptides or rBPI.sub.21 .DELTA.cys, rBPI or rLBP in these experiments; the IC.sub.50 of BPI.30 for inhibition of zymosan-stimulated NO production was found to be between 10 and 100 .mu.g/mL. BPI.3, BPI.5, BPI.13, BPI.29 and BPI.30 were found to have detectable levels of LPS neutralization in this assay, and the relative IC.sub.50 values for these peptides are shown in FIG. 24g.

D. LPS Neutralization Screening Assay of BPI Functional Domain

Peptides using a Cellular Proliferation Assay

An additional LPS neutralization screening assay for evaluation of BPI functional domain peptides was developed. This sensitive assay for inhibition of cellular proliferation in mouse cells treated with LPS can also be utilized for quantitation of LPS levels in human plasma upon development of a standard curve.

In this assay, mouse RAW 264.7 cells (ATCC Accession No. T1B71), maintained in RPMI 1640 media (GIBCO), supplemented with 10 mM HEPES buffer (pH 7.4), 2 mM L-glutamine, penicillin (100 U/mL), streptomcin (100 .mu.g/mL), 0.075% sodium bicarbonate, 0.15 M 2-mercaptoethanol and 10% fetal bovine serum (Hyclone, Inc., Logan, Utah), were first induced by incubation in the presence of 50 U/mL recombinant mouse .gamma.-interferon (Genzyme, Cambridge, Mass.) for 24 h prior to assay. Induced cells were then mechanically collected and centrifuged at 500.times.g at 4.degree. C. and then resuspended in 50 mL RPMI 1640 media (without supplements), re-centrifuged and again resuspended in RPMI 1640 media (without supplements). The cells were counted and their concentration adjusted to 2.times.10.sup.5 cells/mL and 100 .mu.L aliquots were added to each well of a 96-well microtitre plate. The cells were then incubated for about 15 hours with E. coli O113 LPS (Control Standard, Assoc. of Cape Cod, Woods Hole, Mass.), which was added in 100 .mu.L/well aliquots at a concentration of 1 ng/mL in serum-free RPMI 1640 media (this concentration being the result of titration experiments in which LPS concentration was varied between 50 pg/mL and 100 ng/mL). This incubation was performed in the absence or presence of BPI functional domain peptides in varying concentrations between 25 ng/mL and 50 .mu.g/mL. Recombinant human BPI was used as a positive control at a concentration of 1 .mu.g/mL. Cell proliferation was quantitatively measured by the addition of 1 .mu.Ci/well [.sup.3 H]-thymidine 5 hours after the time of initiation of the assay. After the 15-hour incubation, labeled cells were harvested onto glass fiber filters with a cell harvester (Inotech Biosystems, INB-384, Sample Processing and Filter Counting System, Lansing, Mich.).

The results of this assay are shown in FIGS. 26a-26c. FIG. 26a shows the dependence of LPS-mediated inhibition of RAW 264.7 cell proliferation of the presence of LBP, added to the reaction mixture either as a component of serum or as recombinant LBP (at a concentration of 1 .mu.g/mL). FIGS. 26b and 26c illustrate patterns of BPI functional domain peptide behavior found in the above assay. BPI.5 displayed an EC.sub.50 (i.e., the peptide concentration at which the growth inhibitory effect of LPS was reversed by 50%) of 5.3.+-.0.6 .mu.g/mL. BPI.81 was unable to reverse the growth inhibitory effect of LPS on RAW 264.7 cells, but showed additional growth inhibition with an IC.sub.50 (i.e., the peptide concentration at which RAW cell growth was inhibited by 50% from the value without added peptide) of 14.+-.0.2 .mu.g/mL. BPI.98 showed an EC.sub.50 of 0.16.+-.0.08 .mu.g/mL and an IC.sub.50 of 16.5.+-.1.9 .mu.g/mL. Finally, BPI.86 showed an EC.sub.50 of 0.13.+-.0.04 .mu.g/mL and an IC.sub.50 of 37.5.+-.12.5 .mu.g/mL. Results from representative peptides tested with this assay are shown in Table XIII. Additional representative peptides, for example, BPI.99 thru BPI.169, showed activity in this assay. One such peptide, BPI.157, showed an EC.sub.50 comparable to BPI.29 but a lower IC.sub.50.

E. LPS Neutralization Assay based on Inhibition of LPS-induced TNF

Production in Whole Blood

LPS neutralization by BPI functional domain peptides of the invention was assayed in whole blood as follows. Freshly drawn blood from healthy human donors was collected into vacutainer tubes (ACD, Rutherford, N.J.) Aliquots of blood (170 .mu.L) were mixed with 10 .mu.L Ca.sup.++ -, Mg.sup.++ -free PBS containing 2 .mu.ng/mL E. coli O113 LPS, and with 20 .mu.L of varying concentrations of the BPI peptides of the invention ranging in concentration from 0.5-50 .mu.g/mL. These mixtures were then incubated for 4 h at 37.degree. C., and then the reaction stopped by the addition of 55 .mu.L ice-cold Ca.sup.++ -, Mg.sup.++ -free PBS, followed by centrifugation at 500.times.g for 7 min. Supernatants were then assayed for TNF levels using a commercial ELISA kit (Biokine.TM. ELISA Test, T-cell Sciences, Cambridge, Mass.).

The results of these experiments with representative peptides using whole blood samples from two different donors are shown in FIG. 27 and Table XIV. FIG. 27 shows a comparison of TNF inhibition by BPI functional domain peptides BPI.7, BPI.13 and BPI.29; results obtained using rBPI.sub.21 .DELTA.cys are shown for comparison. These results are quantitated as IC.sub.50 values in Table XIV, and compared with LPS neutralization as assayed using NO production by RAW 264.7 cells as described in Section C above.

F. LPS and Heparin Binding Assays using

Tryptophan Fluorescence Quenching

The naturally-occurring amino acid tryptophan can emit light (i.e., it fluoresces) having a wavelength between 300 and 400 nm after excitation with light having a wavelength of between about 280 nm and 290 nm, preferably 285 nm. The amount of emitted light produced by such fluorescence is known to be affected by the local environment, including pH and buffer conditions, as well as binding interactions between proteins and other molecules. Some BPI functional domain peptides derived from domains II and III contain tryptophan residues, and tryptophan fluorescence was used to assay binding interactions between the BPI functional domain peptides of the invention and LPS or heparin.

Tryptophan fluorescence of the BPI functional domain peptides of the invention was determined in the presence or absence of LPS or heparin using a SPEX Fluorolog fluorimeter. Samples were excited with 285 nm light using a 0.25 nm slitwidth. Emission wavelengths were scanned between 300-400 nm using a 1.25 nm slitwidth. Data were accumulated as the average of three determinations performed over an approximately 5 min time span. Samples were maintained at 25.degree. C. or 37.degree. C. during the course of the experiments using a circulating water bath. Crab endotoxin binding protein (CEBP), a protein wherein the intrinsic fluorescence of tryptophan residues is affected by binding to LPS, was used as a positive control. (See Wainwright et al., 1990, Cellular and Molecular Aspects of Endotoxin Reactions, Nowotny et al., eds., Elsevier Science Publishing B.V., The Netherlands, pp. 315-325).

The results of these experiments are shown in Table XV. K.sub.d values were determined by Scatchard-type Stern-Volmer plots of the quenching data as the negative inverse of the slope of such plots. Comparing the data for BPI.10, BPI.46 and BPI.47, it is seen that as the K.sub.d decreased (indicating an increase in avidity for LPS), the percent fluorescence quenching increased. The differences between these peptides include replacement of basic and polar amino acid residues with non-polar residues in BPI.48 as compared with BPI.10. In contrast, as the K.sub.d of heparin binding decreased, a corresponding increase in the percentage of fluorescence quenching was not detected. This result may indicate fundamental differences between the site or nature of heparin binding compared with LPS binding.

G. Neutralization Assay Of Heparin-Mediated Lengthening of Thrombin Time

The effect of BPI functional domain peptides on heparin-mediated lengthening of thrombin time, i.e., the time required for clotting of a mixture of thrombin and plasma, was examined. Thrombin time is lengthened by the presence of endogenous or exogenous inhibitors of thrombin formation, such as therapeutically administered heparin. Agents which neutralize the anti-coagulant effects of heparin will reduce the thrombin time measured by the test.

In these experiments, thrombin clotting time was determined using a MLA Electra 800 Coagulation Timer. Reconstituted plasma (200 .mu.L, Sigma Chemical Co., No. 855-10) was incubated at 37.degree. C. for two minutes in a reaction cuvette. Thrombin Clotting Time reagent (100 .mu.L, Baxter Diagnostics Inc., B4233-50) was added to the reaction cuvette after incubation and clotting time was then measured. Heparin sodium (13 .mu.L, 40 .mu.g/mL in PBS, Sigma Chemical Co., H3393) and exemplary BPI functional domain peptides (10 .mu.L of various dilutions from about 0.05 .mu.g/ml to about 10 .mu.g/ml) were added to the reaction cuvette prior to plasma addition for testing of the effects of these peptides on thrombin clotting time. TCT clotting time (thrombin time) was measured using the BPI peptides indicates and the results are shown in FIG. 28 and Table XVI. These results shown in FIG. 28 and Table XVI below demonstrate that the tested BPI functional domain peptides neutralized heparin, as shown by inhibition of the heparin-mediated lengthening of thrombin time. The IC.sub.50 of this inhibition was quantitated and is shown in Table XVI.

EXAMPLE 21

Heparin Neutralization Assay based on Inhibition of Heparin/FGF-Induced Anglogenesis into Matrigel.RTM. Basement Membrane Matrix

In Vivo

BPI functional domain peptides of the invention are assayed for their ability to inhibit heparin-induced anglogenesis in vivo in mice. Liquid Matrigel.RTM. (Collaborative Biomedical Products, Inc., Bedford, Mass.) is maintained at 4.degree. C. and angiogenic factors are added to the gel in the liquid state as described in Passaniti et al. (1992, Lab. Invest. 67:519-528). Heparin (Sigma, St. Louis, Mo.) is dissolved in sterile PBS to various concentrations ranging from 1,250-10,000 U/mL. Recombinant fibroblast growth factor (bhFGF; BACHEM Bioscience Inc., Philadelphia, Pa.) is diluted to 200 ng/mL with sterile PBS. A volume of 2.5 .mu.L dissolved heparin solution and 2.5 .mu.L recombinant bhFGF is added to 0.5 mL Matrigel.RTM. per mouse injection. BPI functional domain peptides are added to this Matrigel.RTM. mixture at varying concentrations ranging from 0.5 to 50 .mu.g/mL (final concentration) in 10 .mu.L/0.5 mL Matrigel.RTM. aliquot per experimental animal. Ten .mu.L sterile PBS is substituted for BPI functional domain peptides in Mauigel.RTM. aliquots injected into control animals.

Male C57BL/6J mice (Jackson Laboratory, Bar Harbor, Me.) at 6-8 weeks of age are injected subcutaneously down the dorsal midline with 0.5 mL aliquots of Matrigel.RTM. prepared as described above. Seven days after injection, the Matrigel.RTM. gels are excised and placed in 500.+-.L Drabkin's reagent (Sigma). Total protein and hemoglobin content are determined for the gels stored in Drabkin's reagent after mechanical homogenization of the gels. Total protein levels are determined using a microplate assay that is commercially embodied in a kit (DC Protein Assay, Bio-Rad, Richmond, Calif.). Hemoglobin concentration is measured using Sigma Procedure #525 and reagents supplied by Sigma (St. Louis, Mo.) to be used with this procedure. Hemoglobin levels are expressed relative to total protein concentration.

Gels to be used for histological staining are formalin-fixed immediately after excision from the animals rather than being placed in Drabkin's reagent. Formalin-fixed gels are embedded in Tissue-Tek O.C.T. compound (Miles, Inc., Elkhart, Ind.) for frozen sectioning. Slides of frozen sections are stained with hematoxylin and eosin (as described by Humason, 1979, Animal Tissue Techniques, 4th Ed. W.H. feeman & Co., San Fransisco, Calif., Ch.9, pp 111-131).

The effect of the BPI functional domain peptides of the invention are detected by microscopic examination of frozen stained sections for inhibition of anglogenesis relative to Matrigel.RTM. gel slices prepared without added BPI peptides. The extent of angiogenesis inhibition is quantitated using the normalized amounts of hemoglobin found in BPI peptide-containing gel slices.

EXAMPLE 22

Analysis of BPI Functional Domain Peptides in chronic Inflammatory Disease Collagen-Induced or Reactive Arthritis Models

BPI functional domain peptides are administered for their effects in a collagen-induced arthritis model. Specifically, arthritis is induced in mice by intradermal immunization of bovine Type II collagen at the base of the tail according to the method of Smart et al. (1982, J. Clin. Invest. 69:673-683). Generally, mice begin to develop arthritic symptoms at day 21 after collagen immunization. The arthritic scores of the treated mice are then evaluated in a blinded fashion over a period of 120 days for mice treated on each of days 21-25 with doses of either BPI functional domain peptides, control rBPI.sub.23 or rBPI, or buffer which are injected intravenously via the tail vein.

Specifically, bovine Type II collagen (Southern Biotechnology Associates, Inc., Birmingham Ala.) is administered via intradermal injection (0.1 mg/mouse) at the base of the tail on day 0 to groups of male mice (Mouse/DBA/1J), each weighing approximately 20-25 g. BPI functional domain peptides, and rBPI.sub.23 and rBPI are dissolved in a buffer comprised of 0.5 M NaCl, 20 mM sodium acetate (pH 6.0) and diluted with PBS buffer for administration at various concentrations. PBS buffer alone (0.1 mL) is administered as a control.

The collagen-induced arthritis model is also used to evaluate the performance of BPI functional domain peptides in comparison with prommine sulfate. Specifically, BPI peptides are dissolved in PBS as described above and administered at various concentrations. The other test materials are administered at the following dosages: protamine sulfate (Sigma Chemical Co., St. Louis, Mo.) (0.13 mg/mouse), thaumatin (0.12 mg/mouse), and PBS buffer (0.1 mL). Groups of mice receive test or control materials through intravenous injection via the tail vein on each of days 28 through 32 post-injection with collagen.

BPI functional domain peptides are also administered to treat reactive arthritis in a Yersinia enterocolitica reactive arthritis model according to the method of Yong et al. (1988, Microbial Pathogenesis 4:305-310). Specifically, BPI peptides are administered to DBA/2J mice which have previously been injected intravenously with Yersinia enterocolitica cWA 0:8 T2 (i.e., lacking the virulence plasmid according to Yong et al., supra) at a dosage of 4.times.10.sup.8 bacteria calculated to induce a non-septic arthritis in the mice. Groups of mice each receive test or control materials through intravenous injection via the tail vein.

Borrelia burgdorferi is the pathogen responsible for Lyme Disease and associated arthritis and it possesses an LPS-like complex on its cell walls which is different from but structurally related to that of E. coli. The effect of administration of BPI functional domain peptides on inhibition of B. burgdorferi LPS in a Limulus Amoebocyte Lysate (LAL) inhibition assay is determined. Specifically, an LAL assay according to the method of Example 4 is conducted measuring the effect of BPI peptides on B. burgdorferi LPS administered at 2.5 .mu.g/mL and E. coli O113 LPS administered at 2 ng/mL.

EXAMPLE 23

Analysis of BPI Functional Domain Peptides in Mouse Malignant Melanoma Cell Metastasis Model

BPI functional domain peptides, protamine, or buffer controls are administered to test their efficacy in a mouse malignant melanoma metastasis model. Specifically, groups of C57BL/6J mice are inoculated with 10.sup.5 B16.F10 malignant melanoma cells via intravenous injection into the tail vein on day 0. BPI functional domain peptides in various concentrations are administered into the tail vein of test mice on days 1, 3, 6, 8, 10, 13, 15, 17, and 19. Protamine sulfate (0.13 mg/mouse) as a positive control, or PBS buffer (0.1 mL/mouse) as a negative control are similarly administered to additional groups of control mice. The animals are sacrificed via cervical dislocation on day 20 for observation of lung tissues. The lobes of each lung are perfused and inflated by injecting 3 mL water into the lung via the trachea. Superficial tumor nodules are then counted with the aid of a dissecting microscope and the number of tumors found per group analyzed for statistically significant differences.

EXAMPLE 24

Analysis of BPI Functional Domain Peptides in a Mouse Cerebral Capillary Endothelial Cell Proliferation Assay

BPI functional domain peptides are tested for their effects in all endothelial cell proliferation assay. For these experiments, murine cerebral capillary endothelial cells (EC) as described in Bauer (1989, Microvascular Research 37:148-161) are passaged in Medium 199 containing Earle's salts, L-glutamine and 2.2 g/L of sodium bicarbonate (GIBCO, Grand Island, N.Y.), plus 10% heat inactivated fetal calf serum (FCS; Irvine Scientific, Irvine, Calif.) and 1% penicillin/streptomycin (GIBCO). Harvesting of the confluent cells is performed by trypsinization with trypsin-EDTA (GIBCO) for 3 minutes. Trypsinization is stopped by adding 10 mL of the passage medium to the flask. Proliferation assays are performed on freshly harvested EC in standard flat bottom 96-well microtiter plates. A final volume of 200 .mu.L/well is maintained for each well of the assay. A total of 4.times.10.sup.4 EC cells is added to each well with varying concentrations of BPI peptides, or buffer control. After 48 hours of culture in a 5% CO.sub.2 incubator, 1 .mu.Ci of [.sup.3 H] thyroidine in 10 .mu.L of Medium 199 is added to each well. After a 24 hour pulse, the EC cells are harvested by trypsinization onto glass microfiber filters and incorporated [.sup.3 H]thymidine is quantitated with a gas proportional solid phase beta counter.

Direct binding studies of BPI peptides on EC cells are performed by harvesting the 10-times passaged cells from a confluent flask and resuspending the trypsinized cells in 12.5 mL of culture medium. Then, 0.5 mL of the cell suspension is added to each well of a standard 24 well tissue culture plate and incubated overnight. The plate is washed with 0.1% bovine serum albumin in phosphate buffered saline containing calcium and magnesium (GIBCO). After washing, 0.5 mL BSA/PBS is added per well. Concentration dependent inhibition of EC cell proliferation is measured in terms of decreases in [.sup.3 H]-thymidine uptake.

EXAMPLE 25

Analysis of BPI Function Domain Peptides in Animal Models

A. Analysis in a Mouse Endotoxemia Model

BPI functional domain peptides are tested for their efficacy in a mouse experimental endotoxemia model. Groups of at least 15 mice are administered an intravenous injection of endotoxin (e.g., E. coli O111:B4, Sigma Chemical Co., St. Louis, Mo.) at a LD90 dosage (e.g., 40 mg/kg). This is followed by a second intravenous injection of the test peptide in varying concentrations from about 0.1 mg/kg to about 100 mg/kg, preferably in the range of about 1 to 50 mg/kg. Injections of buffer without added peptide are used in negative control mice. The animals are observed for 7 days and mortality recorded. The efficacy of the peptides of this invention is measured by a decrease in endotoxemia-associated mortality in peptide-injected mice as compared with control mice. BPI.102 is a representative compound active in this murine model.

B. Analysis in a Mouse Peritonitis Model

BPI functional domain peptides are tested for their efficacy in a mouse model of acute peritonitis. Groups of at least 15 mice are challenged with 107 live E. coli bacteria strain O7:K1 in 0.5 mL and then treated with 1.0 mL of a solution of BPI functional domain peptides at varying concentrations from about 0.1 mg/kg to about 100 mg/kg. Injections of buffer without added peptide are used in negative control mice. The animals are observed for 7 days and mortality recorded. Effective BPI functional domain peptides show a decrease in mortality of test group mice compared with control group mice.

C. Analysis in Mouse Candida albicans Model

A murine model for systemic Candidiasis has been used to test the in vivo effectiveness of various therapies in treating infection by Candida albicans. TNF-.alpha. was shown to have a protective role in this murine model (Louie, A., et al., 1994, Infection and Immunity, 62(7):2761-2772), and previous to this, recombinant soluble IL-4 receptor was shown to cure the infection (Puccetti, P., et al., 1993, J. Infec. Diseases, 169:1325-1331). Certain mice have been identified to have a genetic susceptibility for C. albicans and the resulting Candidiasis, and are thus suitable for use as a model system to test effectiveness of treatments which will combat such infections (Romani, L., et al., 1993, J. Immunol. 150:925-931).

Peptides of the invention are tested for their efficacy against systemic C. albicans infection in this murine model according to procedures substantially as described in Louie et al., supra.; Puccetti et al., supra. Suitable mice can be developed and identified substantially as described by Hector et al. (1982, Infection and Immunity, 38(3):1020-1028).

Groups of at least 15 mice are challenged with varying doses of from about 10.sup.3 to 10.sup.8, preferably 10.sup.6 to 10.sup.8 live C. albicans (strain CA-6, B311, 88-689-6, the Candida strain used in Example 18, or other suitable isolated strain), in 0.5 mL and then treated with from about 0.1 to 1.0 mL of a solution of BPI functional domain peptides at varying concentrations from about 0.1 mg/kg to about 100 mg/kg. Injections of buffer without added peptide are used in negative control mice. Assay can be performed by quantitative cultures of organs from mice (Louie et al., supra.), or by determination of survival times. Effective peptides of the invention are active in modifying the effects of diseases caused by C. albicans.

EXAMPLE 26

Therapeutic Use of BPI Functional Domain Peptides in a Human In vivo Endotoxin Neutralization Model

A controlled, double-blind crossover study is designed and conducted as in co-owned, for example copending U.S. patent application Ser. No. 08/188,221 filed Jan. 24, 1994, to investigate the effects of BPI functional domain peptides in humans rendered endotoxemic by intravenous infusion of bacterial endotoxin.

EXAMPLE 27

Bactericidal Activity of BPI Peptides Against Antibiotic Resistant Bacteria

These studies were conducted substantially following the procedures used in Example 2.

A. Polymyxin B Resistant Strain of Salmonella Typhimurium

To investigate any interaction between the mechanism for polymyxin B resistance and resistance to bactericidal activity of peptides of the instant invention, a polymyxin B resistant strain of Salmonella typhimurium was isolated by plating 10.sup.8 wildtype Salmonella typhimurium (SL3770, Genetic Stock Center, Calgary, Canada) onto a polymyxin gradient plate (see e.g., Roland et al., 1993, J. Bacteriology, 175:4154-4164). An overnight culture of the isolated polymyxin B resistant Salmonella typhimurium designated LR-1 and the wild type strain were diluted 1:50 into fresh tryptic soy broth and incubated for 3 hours at 37.degree. C. to attain log phase. The culture for S. typhimurium LR-1 was supplemented with 2.5 .mu.g/mL polymyxin B sulfate. Bacteria were resuspended in fresh buffer and concentration determined by absorbance at 570 nm. Bacteria were added to the molten agarose to give a final concentration of 1.times.10.sup.6 /mL. rBPI.sub.21 was serially diluted two-fold starting from 2 mg/mL in D-PBS. Polymyxin B was diluted in the same fashion starting from 20 mg/mL. Test peptides were 2-fold serially diluted in D-PBS starting from approximately 1 mg/mL. As described in Example 2, approximately 5 gL was added per well.

FIGS. 29a and FIG. 29b show the bactericidal activity of rBPI.sub.21 (Closed Circle); Polymyxin B (Open Circle); BPI.30 (Closed Triangle); BPI.48 (Open Triangle); BPI.69 (Closed Square); BPI.105 (Open square); against S. typhimurium; and against a polymyxin B resistant strain of S. typhimurium designated LR-1, respectively.

B. Antibiotic Resistant Strain of E. coli

In order to test the bactericidal activity of BPI peptides against multi-antibiotic resistant E. coli, a multi-resistant strain E. coli 19536 (ampR 16 .mu.g/mL; ceftazidimeR>16 .mu.g/mL; ceftriazoneR>16 .mu.g/mL) was tested. This strain is a clinical isolate obtained from the Baxter Microscan.RTM. library (Sacramento, Calif.). An overnight culture of E. coli 19536 was tested as described in Example 2 and part A above. Test BPI peptides were 2-fold serially diluted in D-PBS starting at about 1 mg/mL. BPI.48, BPI.63, BPI.69, and BPI.88 are representative compounds with activity against multi-antibiotic resistant E. coli.

FIGS. 29c and 29d show the bactericidal activity of rBPI.sub.21 (Closed Square), ceftriaxone (Open Square), BPI.48 (Closed Circle), BPI.63 (Open Circle), BPI.69 (Closed Triangle), and BPI.88 (Open Triangle), against an antibiotic resistant strain of E. coli (19536) and against E. coli O111:B4, respectively.

C. Bactericidal Activity of BPI Peptides on Pneumoniae

Antibiotic Resistant Strain of Klebsiella pneumoniae

In order to test the bactericidal activity of BPI peptides on antibiotic resistant strains of Klebsiella pneumoniae, the multi-resistant clinically isolated strain 19645 was tested as in Example 2 and above. (K. pneumoniae 19645 has MIC (.mu.g/Ml): am>16; to>16; azt>16; pi>64; ak>16; crm>16; caz>16; cax>16; eft>32; a/s>16; grn>6; to>6; cfz>16). This strain was obtained from the Baxter Microscan.RTM. library (Sacramento, Calif.). An overnight culture of Klebsiella pneumoniae 19645 and a non-resistant strain (ATCC 29011) were diluted 1:50 into fresh tryptic soy broth and incubated for 3 hours at 37.degree. C. to attain log phase. The overnight cultures were tested as described in Example 2 and part A above. BPI.7, BPI.48, BPI.63, BPI.69, BPI.103, BPI.105, and BPI.119 are representative of compounds active against antibiotic resistant Klebsiella.

FIGS. 29e and 29f shows the bactericidal activity of representative peptides against an antibiotic resistant strain of K. pneumoniae (19645). In FIG. 29e the results with BPI.7 (Open Circle), BPI.48 (Closed Circle), BPI.63 (Open Triangle), BPI.69 (Closed Triangle), rBPI.sub.21 (Open Square), ceftazidime (Closed Square), and ceftriaxone (Small Triangle), are shown. In FIG. 29f the results of BPI.88 (Open Circle), BPI.103 (Closed Circle), BPI.105 (Open Triangle), BPI.119 (Closed Triangle), rBPI.sub.21 (Open Square), ceftazidime (Closed Square), and ceftriaxone (Small Circle); are shown.

FIGS. 29g and 29h show the bactericidal activity of representative peptides against K. pneumoniae (ATCC 29011). In FIG. 29g the results with BPI.7 (Open Circle), BPI.48 (Closed Circle), BPI.63 (Open Square), BPI.69 (Closed Square), rBPI.sub.21 (Open Triangle), ceftazidime (Closed Triangle), and ceftriaxone (Small Circle), are shown. In FIG. 29h the results with BPI.88 (Open Circle), BPI.103 (Closed Circle), BPI.105 (Open Square), BPI.119 (Closed Square), rBPI.sub.21 (Open Triangle), ceftazidime (Closed Triangle), and ceftriaxone (Small Circle), are shown.

It should be understood that the foregoing disclosure emphasizes certain specific embodiments of the invention and that all modifications or alternatives equivalent thereto are within the spirit and scope of the invention as set forth in the appended claims.

Claims

What is claimed is:

1. A compound which is a peptide selected from the group consisting of: ASQQGTAALQKELKRIKIPDYSDSFKIKH (SEQ ID NO: 1); GTAALQKELKRIKIPDYSDSFKIKHLGKGH (SEQ ID NO: 2); LQKELKRIKIPDYSKSFKIKHL (SEQ ID NO: 3); QQGTAALQKELKRIK (SEQ ID NO: 4); and GTAALQKELKRIKIP (SEQ ID NO: 5). ______________________________________

2. A compound which is a peptide selected from the group consisting of: SSQISMVPNVGLKFSISNANIKISGKWKAQKRFLK (SEQ ID NO: 6); IKISGKWKAQKRFLK (SEQ ID NO: 7); KWKAQKRFLK (SEQ ID NO: 8); CIKISGKWKAQKRFLK (SEQ ID NO: 9); CIKISGKWKAQKRFLKC (SEQ ID NO: 10); NVGLKFSISNANIKISGKWKAQKRFLK (SEQ ID NO: 11); AKISGKWKAQKRFLK (SEQ ID NO: 16); IAISGKWKAQKRFLK (SEQ ID NO: 17); IKASGKWKAQKRFLK (SEQ ID NO: 18); IKIAGKWKAQKRFLK (SEQ ID NO: 19); IKISAKWKAQKRFLK (SEQ ID NO: 20); IKISGAWKAQKRFLK (SEQ ID NO: 21); IKISGKAKAQKRFLK (SEQ ID NO: 22); IKISGKWAAQKRFLK (SEQ ID NO: 23); IKISGKWKAAKRFLK (SEQ ID NO: 24); IKISGKWKAQARFLK (SEQ ID NO: 25); IKISGKWKAQKAFLK (SEQ ID NO: 26); IKISGKWKAQKRALK (SEQ ID NO: 27); IKISGKWKAQKRFAK (SEQ ID NO: 28); IKISGKWKAQKRFLA (SEQ ID NO: 29); IKISGAWAAQKRFLK (SEQ ID NO: 30); IKISGKWKAAARFLK (SEQ ID NO: 31); IAISGKWKAQKRFLA (SEQ ID NO: 32); IKISGKWKAKQRFLK (SEQ ID NO: 47); IKISGKFKAQKRFLK (SEQ ID NO: 48); IKISGKW.sub.D KAQKRFLK (SEQ ID NO: 49); IKISGKA.sub..beta.-(1-naphthyl) KAQKRFLK (SEQ ID NO: 50); KRFLKKWKAQKRFLK (SEQ ID NO: 51); KWKAQKRFLKKWKAQKRFLK (SEQ ID NO: 54); KRFLKKWKAQKRFLKKWKAQKRFLK (SEQ ID NO: 55); KWKAAARFLKKWKAQKRFLK (SEQ ID NO: 57); KWKAQKRFLKKWKAAARFLK (SEQ ID NO: 58); KWKAAARFLKKWKAAARFLK (SEQ ID NO: 59); KWKAAARFLKKWKAAARFLKKWKAAARFLK (SEQ ID NO: 60); IKISGKWKAQFRFLK (SEQ ID NO: 62); IKISGKA.sub..beta.-(1-naphthyl) KAQKRFLK (SEQ ID NO: 63); IIKISGKWKAQKRA.sub..beta.-(3-pyridyl) LK (SEQ ID NO: 64); QKRFLKKWKAQKRFLKKWKAQKRFLK (SEQ ID NO: 65); A.sub.D A.sub.D IKISGKWKAQKRFLK (SEQ ID NO: 66); IKISGKWKAQF.sub.D RFLK (SEQ ID NO: 71); IKISGKWKAQWRFLK (SEQ ID NO: 72); IKISGKWKAKKRFLK (SEQ ID NO: 73); IKISGKWKAQA.sub..beta.-(1-naphthyl) RFLK (SEQ ID NO: 74); IKISGKWKAFKRFLK (SEQ ID NO: 75); IKISGKA.sub..beta.-(1-naphthyl) KAQFRFLK (SEQ ID NO: 78); IKISGKWKAFFRFLK (SEQ ID NO: 82); IKISGKA.sub..beta.-(1-naphthyl) KAFKRFLK (SEQ ID NO: 84); IKISGKA.sub..beta.-(1-naphthyl) KAFFRFLK (SEQ ID NO: 85); KWKAQWRFLKKWKAQWRFLKKWKAQWRFLK (SEQ ID NO: 93); CIKISGKWKAQKRPLC (SEQ ID NO: 99); IKKRAISFLGKKWQK (SEQ ID NO: 100); IKISGKWKAWKRFLKK (SEQ ID NO: 102); IKISGKWKAWKRA.sub..beta.-(1-naphthyl) LKK (SEQ ID NO: 104); IKISGKA.sub..beta.-(1-naphthyl) KAQA.sub..beta.-(1-naphthyl) RFLK (SEQ ID NO: 111); KWQLRSKGKIKIFKA (SEQ ID NO: 113); IKISGKA.sub..beta.-(1-naphthyl) KAA.sub..beta.-(1-naphthyl) KRFLK (SEQ ID NO: 115); IKISGKWKAQKRKLK (SEQ ID NO: 116); IKISGKWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLK (SEQ ID NO: 117); IKISGKA.sub..beta.-(1-naphthyl) KAA.sub..beta.-(1-naphthyl) A.sub..beta.-( 1-naphthyl) RFLK (SEQ ID NO: 118); IKISGKWKAQERFLK (SEQ ID NO: 132); IKISGKWKAQKRWLK (SEQ ID NO: 137); IKISGKWKAEKKFLK (SEQ ID NO: 143); KWAFAKKQKKRLKRQWLKKF (SEQ ID NO: 148); KWKAQKRFLKKWKAQKAFLKKWKAQKRFLK (SEQ ID NO: 149); KWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLKKWKAQKRFLK (SEQ ID NO: 150); KWKAQKRFLKKWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) (SEQ ID NO: 151); KWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLKKWKAA.sub..bet a.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLK (SEQ ID NO: 152); KWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLKKWKAA.sub..bet a.-(1-naphthyl) A.sub..beta.-(1- (SEQ ID NO: 153); naphthyl) RFLKKWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLK KA.sub..beta.-(1-naphthyl) KAQA.sub..beta.-(1-naphthyl) RFLKKA.sub..beta.- (1-naphthyl) KAQRFLK (SEQ ID NO: 156); KWKAQWRFLKKWKAQWRFL (SEQ ID NO: 159); KWKAA.sub..beta.-(1-naphthyl) KRFLKKWKAA.sub..beta.-(1-naphthyl) KRFLK (SEQ ID NO: 160); KA.sub..beta.-(1-naphthyl) KAQFRFLKKA.sub..beta.-(1-naphthyl) KAQFRFLK (SEQ ID NO: 161); KWKAQKRF (SEQ ID NO: 163); CKWKAQKRFLKMSC (SEQ ID NO: 164); CKWKAQKRFC (SEQ ID NO: 165); IKISGKWKAQKRA.sub..beta.-(1-naphthyl) LK (SEQ ID NO: 166); KWKAFFRFLKKWKAFFRFLK (SEQ ID NO: 101); IKISGKWKAAWRFLK (SEQ ID NO: 223); IKISGKWKAA.sub..beta.-(1-naphthyl) FRFLK (SEQ ID NO: 224); IKISGKWKAAFRFLK (SEQ ID NO: 225); and IKISGKWKAA.sub..beta.-(1-naphthyl) ARFLK (SEQ ID NO: 226). __________________________________________________________________________

3. A compound which is a peptide selected from the group consisting of: VHVHISKSKVGWLIQLFHKKIESALRNK (SEQ ID NO: 12); KSKVWLIQLFHKK (SEQ ID NO: 13); SVHVHISKSKVGWLIQLFHKKIESALRNK (SEQ ID NO: 14); KSKVGWLIQLFHKK (SEQ ID NO: 15); ASKVGWLIQLFHKK (SEQ ID NO: 33); KAKVGWLIQLFHKK (SEQ ID NO: 34); KSAVGWLIQLFHKK (SEQ ID NO: 35); KSKAGWLIQLFHKK (SEQ ID NO: 36); KSKVAWLIQLFHKK (SEQ ID NO: 37); KSKVGALIQLFHKK (SEQ ID NO: 38); KSKVGWAIQLFHKK (SEQ ID NO: 39); KSKVGWLAQLFHKK (SEQ ID NO: 40); KSKVGWLIALFHKK (SEQ ID NO: 41); KSKVGWLIQAFHKK (SEQ ID NO: 42); KSKVGWLIQLAHKK (SEQ ID NO: 43); KSKVGWLIQLFAKK (SEQ ID NO: 44); KSKVGWLIQLFHAK (SEQ ID NO: 45); KSKVGWLIQLFHKA (SEQ ID NO: 46); KSKVGWLIQLFHKKKSKVGWLIQLFHKK (SEQ ID NO: 56); GWLIQLFHKKIESALRNKMNS (SEQ ID NO: 61); VHVHISKSKVGWLIQLFHKKIE (SEQ ID NO: 67); KSKVGWLIQLWHKK (SEQ ID NO: 76); KSKVGA.sub..beta.-(1-naphthyl) LIQFHKK (SEQ ID NO: 77); KSKVLWLIQLFHKK (SEQ ID NO: 79); KSKVGWLILLFHKK (SEQ ID NO: 80); KSKVGWLIQLFLKK (SEQ ID NO: 81); KSKVGWLIFLFHKK (SEQ ID NO: 86); KSKVGWLIKLFHKK (SEQ ID NO: 87); KSKVGWLIQLFFKK (SEQ ID NO: 89); KSKVFWLIQLFHKK (SEQ ID NO: 90); KSKVGWLIQLFHKF (SEQ ID NO: 91); KSKVKWLIQLFHKK (SEQ ID NO: 92); KSKVKWLIKLFHKK (SEQ ID NO: 94); KSKVKWLIKLFFKFKSKVKWLIKLFFKF (SEQ ID NO: 95); KSKVGWLISLFHKK (SEQ ID NO: 103); KSKVGWLITLFHKK (SEQ ID NO: 105); KSKVGWLIQLFWKK (SEQ ID NO: 106); KSKVGWLIQLFHKW (SEQ ID NO: 107); KSKVGWLIQLA.sub..beta.-(1-naphthyl) HKK (SEQ ID NO: 108); KSKVGWLIQLFA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 109); KSKVGWLIQLFHKA.sub..beta.-(1-naphthyl) (SEQ ID NO: 110); KSKVGWLIQFFHKK (SEQ ID NO: 112); KSKVKA.sub..beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 114); KSKVGW.sub.(p-amino) LIFLFHKK (SEQ ID NO: 119); KSKVKWLIQLWHKK (SEQ ID NO: 120); KSKVGWLUYLFHKK (SEQ ID NO: 121); KSKVGW.sub.D LIQLFHKK (SEQ ID NO: 122); KSKVGFLIQLFHKK (SEQ ID NO: 123); KSKVGF.sub.D LIQLPHKK (SEQ ID NO: 124); KSKVGA.sub.D-1-.beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 125); KSKVGA.sub.2-.beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 126); KSKVGA.sub.D-2-.beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 127); KSKVGA.sub.(pyridyl) LIQLFHKK (SEQ ID NO: 128); KSKVGF.sub.(p-amino) LIQLFHKK (SEQ ID NO: 129); KSKVF.sub.(p-amino) WLIQLFHKK (SEQ ID NO: 130); KSKVGKLIQLPHKK (SEQ ID NO: 131); CKSKVGWLIQLFHKKC (SEQ ID NO: 133); KSKVKFLIQLFHKK (SEQ ID NO: 134); KSKVGYLIQLFHKK (SEQ ID NO: 135); KSKVGWLIQWFHKK (SEQ ID NO: 138); KSKVGWLIQA.sub..beta.-(1-naphthyl) FHKK (SEQ ID NO: 139); KSKVGA.sub.(cyclohexyl) LIQLFHKK (SEQ ID NO: 140); KSKVGWLIQLFA.sub..beta.-(1-naphthyl) KA.sub..beta.-(1-naphthyl) (SEQ ID NO: 142); KSKVGA.sub..beta.-(1-naphthyl) LIQLFA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 144); KSKVKALIQLFHKK (SEQ ID NO: 157); KSKVGVLIQLFHKK (SEQ ID NO: 162); KSKVGA.sub..beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 167: KSKVGA.sub..beta.-(1-naphthyl) LIQA.sub..beta.-(1-naphthyl) FHKK (SEQ ID NO: 168); KSKVGA.sub..beta.-(1-naphthyl) LIF.sub.(p-amino) LFHKK (SEQ ID NO: 169); KSKVF.sub.(p-amino) A.sub..beta.-(1-naphthyl) LIQLFHKK (SEQ ID NO: 170); KSKVGA.sub..beta.-(1-naphthyl) LIQLWHKK (SEQ ID NO: 171); KSKVGWLIQA.sub..beta.-(1-naphthyl) FHKA.sub..beta.-(1-naphthyl) (SEQ ID NO: 172); KSKVGWLIA.sub..beta.-(1-naphthyl) LFHKA.sub..beta.-(1-naphthyl) (SEQ ID NO: 173); KSKVA.sub..beta.-(1-naphthyl) WLIQLFHKA.sub..beta.-(1-naphthyl) (SEQ ID NO: 174); KSKVGWLIQLWHKA.sub..beta.-(1-naphthyl) (SEQ ID NO: 175); KSKVGWLIQA.sub..beta.-(1-naphthyl) FA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 176); KSKVGWLIF.sub.(p-amino) LFA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 177); KSKVF.sub.(p-amino) WLIQLFA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 178); KSKVGWLIQLWA.sub..beta.-(1-naphthyl) KK (SEQ ID NO: 179); KSKVGWLIFL.sub.(p-amino) A.sub..beta.-(1-naphthyl) HKK (SEQ ID NO: 180); KSKVF.sub.(p-amino) WLIQA.sub..beta.-(1-naphthyl) FHKK (SEQ ID NO: 181); KSKVGWLIQA.sub..beta.-(1-naphthyl) WHKK (SEQ ID NO: 182); KSKVF.sub.(p-amino) WLIF.sub.(p-amino) LFHKK (SEQ ID NO: 183); KSKVGWLIF.sub.(p-amino) LWHKK (SEQ ID NO: 184); KSKVF.sub.(p-amino) WLIQLWHKK (SEQ ID NO: 185); KSKVGA.sub.D-.beta.-(2-naphthyl) LILLFHKK (SEQ ID NO: 190); KSKVGWLILLFHKKKSKVGWLILLFHKK (SEQ ID NO: 191); KSKVGWLIFLFHKKKSKVGWLIFLFHKK (SEQ ID NO: 192); KSKVGWLILLFHKKKSKVGWLIQLFHKK (SEQ ID NO: 193); KSKVGWLIQLFHKKKSKVGWLILLFHKK (SEQ ID NO: 194); KSKVGWLIFLFHKKKSKVGWLIQLFHKK (SEQ ID NO: 195); KSKVGWLIQLFHKKKSKVGWLIFLFHKK (SEQ ID NO: 196); KSKVGWLILLWHKK (SEQ ID NO: 198); KSKVGA.sub.D-.beta.-(2-naphthyl) LIQLWHKK (SEQ ID NO: 199); KSKVGA.sub.D-.beta.-(2-naphthyl) LILLWHKK (SEQ ID NO: 200); KSKVGGLIQLFHKK (SEQ ID NO: 201); KSKVGLLIQLFHKK (SEQ ID NO: 202); KSKVGWLIQLFHKK (SEQ ID NO: 203); KSKVGA.sub.D LIQLFHKK (SEQ ID NO: 204); KSKVGV.sub.D LIQLFHKK (SEQ ID NO: 205); KSKVGA.sub..beta. LIQLFHKK (SEQ ID NO: 206); KSKVG(delta-aminobutyric acid)LIQLFHKK (SEQ ID NO: 207); KSKVG(gamma-aminobutyric acid)LIQLFHKK (SEQ ID NO: 208); KSKVGA.sub.(delta-Methyl) LIQLFHKK (SEQ ID NO: 209); KSKVGG.sub.(t-butyl) LIQLFIJKK (SEQ ID NO: 210); KSKVGG.sub.(N-Methyl) LIQLFHKK (SEQ ID NO: 211); KSKVGV.sub.(N-Methyl) LIQLFHKK (SEQ ID NO: 212); KSKVGL.sub.(N-Methyl) LIQLFHKK (SEQ ID NO: 213); KSKVGWLINLFHKK (SEQ ID NO: 214); KSKVGWLIELFHKK (SEQ ID NO: 215); KSKVGWLIDLFHKK (SEQ ID NO: 216); KSKVGWLIKLFHKK (SEQ ID NO: 217); KSKVKVLIQLFHKK (SEQ ID NO: 218); KSKVKWAIQLFHKK (SEQ ID NO: 219); KSKVGVAIQLFHKK (SEQ ID NO: 220); and KSKVKVAIQLFHKK (SEQ ID NO: 221). ______________________________________

4. A compound which is a peptide selected from the group consisting of: KWKAQKRFLKKSKVGWLIQLFHKK (SEQ ID NO: 52); IKISGKWKAQKRFLKKSKVGWLIQLFHKK (SEQ ID NO: 53); KSKVGWLIQLFHKKKWKAQKRFLK (SEQ ID NO: 70); IKISGKA.sub..beta.-(1-naphthyl) KAQFRFLKKSKVGWLIFLFHKK (SEQ ID NO: 83); IKISGKA.sub..beta.-(1-naphthyl) KAQFRFLKKSKVGWLIQLFHKK (SEQ ID NO: 88); KWKAQFRFLKKSKVGWLILLFHKK (SEQ ID NO: 96); A.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLKF (SEQ ID NO: 136); KWKAAARFLKKSKVGWLIQLFHKK (SEQ ID NO: 141); KWKVFKKIEKKSKVGWLIQLFHKK (SEQ ID NO: 147); IKISGKWKAA.sub..beta.-(1-naphthyl) A.sub..beta.-(1-naphthyl) RFLKKSKVGWLIQ LFHKK (SEQ ID NO: 154); KA.sub..beta.-(1-naphthyl) KAQA.sub..beta.-(1-naphthyl) RFLKKSKVGKWKAFKRFL K (SEQ ID NO: 155); KWKAQWRFLKKSKVGWLIQLFHKK (SEQ ID NO: 158); KA.sub..beta.-(1-naphthyl) KAQA.sub..beta.-(1-naphthyl) RFLKKSKVGWLILLFHKK (SEQ ID NO: 186); KWKAQFRFLKKSKVGWLIQLWHKK (SEQ ID NO: 187); KWKAQFRFLKKSKVGA.sub.D-.beta.-(2-naphthyl) LIQLFHKK (SEQ ID NO: 188); KA.sub..beta.-(1-naphthyl) KAA.sub..beta.-(1-naphthyl) KRFLKKSKVGA.sub.D-. beta.-(2-naphthyl) LIQLFHKK (SEQ ID NO: 189); KWKAQFRFLKKSKVGWLIFLFHKK (SEQ ID NO: 197); and KA.sub..beta.-(1-naphthyl) AQFRFLKKSKVGWLILLFHKK (SEQ ID NO: 222). __________________________________________________________________________

Patent Citations (8)

Patent	Date	Inventor
US5088164	1992-02-01	Maione et al.
US5112946	1992-05-01	Maione
US5348942	1994-09-01	Little, II et al.
WOXWO9201003	1992-01-01
WOXWO9202240	1992-02-01
WOX9203535	1992-03-01
WOXWO9203535	1992-03-01
WOXWO9313794	1993-07-01

Non-Patent Literature (37)

Little et al., J. of Biol. Chem., vol. 269, No. 3, Jan. 21, 1994, pp. 1865-1872.
Ool et al., J. of Biol Chem., vol. 262, No. 31, Nov. 5, 1987, pp. 14891-14894.
Gray et al., J. of Biol Chem., vol. 264, No. 16, Jun. 5, 1989, pp. 9505-9509.
Boman et al., 1993, "Mechanisms of Action on Escherichia coli of Cecropin P1 and PR-39, Two Antibacterial Peptides from Pig Intestine", Infect. Immunol., 61: 2978-2984.
Charles et al., 1989, "The Three-Dimensional Structure of Bovine Platelet factor 4 at 3.0.ANG. Resolution", J. Biol. Chem. 264: 29092-29099.
Cook et al., 1992, "Platelet Factor 4 Efficiently Reverses Heparin Anticoagulation in the rat without Adverse Effects of Heparin-protamine Complexes", Circulation 85: 1102-1109.
Elsbach et al., 1992, "Oxygen-Independent Antimicrobial Systems of Phagocytes", Inflammation: Basic Principles and Correlates, 2d ed., Review Press, Ltd., Chapter 30, pp. 603-636.
Elsbach et al., 1979, "Separation and Purification of a Potent Bactericidal/Permeability-increasing Protein and a Closely Associated Phospholipase A.sub.2 from rabbit Polymorphonuclear Leukocytes", J. Biol. Chem. 254: 11000-11009.
Folkman et al., 1992, "Abiogenesis and Inflammation", Inflammation: Basic Principles and Correlates, 2d ed., Review Press, Ltd., Chapter 40, pp. 821-829.
Gammon et al., 1991, "T Cell Determinant Structure: Cores and Determinant Envelopes in Three Major Histocompatability Complex Haplotypes", J. Exp. Med. 173: 609-617.
Garcia et al., 1983, "Repligen IND for Platelet Factor", BioWorld Today, p. 5, Jan. 11, 1983.
Gazzano-Santoro et al., 1992, "High Affinity Binding of the Bactericidal/Permeability-increasing Protein and a Recombinant Amino-terminal Fragment of the Lipid-A Region of Lipopolysaccharide", Infect. Immunol. 60 4754-4761.
Gray et al., 1989, "Clonming of the cDNA of a Human Neutrophil Bactericidal Protein", J. Biol. Chem. 264: 9505-9509.
Guo et al., 1992, "Heparin-binding Peptides from the Type I Repeats of Thrombospondin", J. Biol. Chem. 267: 19349-19355.
Heyderman et al., 1992, "Reduction of the Anticoagulant Activity of Glycosaminoglycans on the Surface of the Vascular Endfothelium by Endotoxin and Neutrophils: Evaluation of an Amidolytic Assay", Thrombosis Res. 67:pp. 677-685.
Hiti-Harper et al., "Platelet Factor 4: An Inhibitor of Collagenase", Science 199: 991-992.
Maeji et al., 1990, "Multi-pin Peptide Strategy for T Cell Determinant Analysis", J. Immunolog. Methods 134: 23-33.
Maione et al., 1991, "Inhibition of Tumor Growth in Mice by an Analogue of Platelet Factor 4 that Lacks Affinity for Heparin and Retains Potent Angiostatic Activity", Cancer Res. 51: 2077-2083.
Maione et al., 1992, CAS BioTech Updates 117: 226303.
Marra et al., 1992, "The Role of Bactericidal/Permeability-increasing Protein as a natural Inhibitor of Bacterial Endotoxin", J. Immunol. 148: 532-537.
Merrifield et al., 1966, "Instrument for Automated Synthesis of Peptides", Anal. Chem. 38: 1905-1914.
Merrifield et al., 1963, "Solid Phase Peptide Synthesis", J. Amer. Chem. Soc. 85: 2149-2154.
Miles et al., 1991, "Recombinant Platelet Factor 4 (rPF4) and a Non-heparin Binding Derivative Inhibits AIDS-Kaposi's Sarcoma Derived Cell Lines", VII International Conference on AIDS, 41: 108.
Ooi et al., 1991, "Endotoxion-neutralizing Properties of the 25kD N-terminal Fragment and a Newly-isolated 30kD C-terminal Fragment of the 55-60kD Bactericidal/Permeability-increasing Protein of Human Neutrophils", J. Exp. Med. 174: 649.
Peacock et al., 1992, "Angiobiogenesis Inhibition Suppresses Collagen Arthritis", J. Exp. Med. 175: 1135-1138.
Posnet & Tam, 1989, Methods in Enzymology 178: 739-746.
Repligen Corporation, 1992, Research in Review.
Stuart et al., "Nature and the Specificity of the Immune Response to Collagen in Tyoe II Collagen-induced Arthritis in Mice", J. Clin. Invest. 69: 673-683.
Takayama et al., 1987, "Absence of Lipopolysaccharide in the Lyme Disease Spirochete", Infect. and Immun. 55: 2311-2312.
Taylor et al., 1982, "Protamine is an Inhibitor of Angiogenesis", Nature 297: 307-312.
Tontsch et al., 1981, "Isolation, Characterization and Long-term Cultivation of Porcine and Murine Cerebral Capillary Endothelial Cells", Microvascular Res. 37: 148-161.
Wade et al., 1990, "All-D amino acid-containing chennel-forming antibotic peptides", Proc. Natl. Acad. Sci. USA 87: 4761-4765.
Weiss et al., 1987, "Cellular and Subcellular Localization of the Bactericidal/Permeability-increasing Protein of Neutrophils", Blood 69: 652-659.
Weiss et al., 1987, "Parrtial Characterization and Purification of Rabbit Granulocyte Factor that Increases Permeability of Escherichia coli", J. Clin. Invest. 55: 33-42.
Yayon et al., 1991, "Cell-surface, Heparin-like Molecules are required for Binding of Basic Fibroblast Growth to its Highy Affinity Receptor", Cell 64: 641-648.
Yong et al., 1988, "An Experimental Mouse Model of Yersinia-induced Reactive Arthritis", Microbial Pathogenesis 4: 305-310.
Zasloff et al., 1988, "Antimicrobial activity of synthetic magainin peptides and several analogues", Proc. Natl. Acad. Sci. USA 85: 910-913.

Abstract

Patent Drawings (78 sheets)