CA2162149C - Antibodies to p-selectin and their uses - Google Patents

Abstract

The present invention relates to compositions and methods for treating inflammation and other pathological conditions using novel blocking P-selectin antibodies that inhibit adhesion of leukocytes to activated platelets and/or to activated vascular endothelium in vivo.
Both murine and humanized antibodies are provided.

Description

YC:T/U594J04935 ANTIBODIES TO P-SELECTIN AND THEIR USES

TECHNICAL FIELD
This application relates generally to novel immunoglobulins reactive with functional epitopes on a cell surface receptor P-selectin. This application also relates generally to diagnostic and therapeutic methods of using the antibodies.

BACKGROUND OF TfIE INVENTION
The ability of cells to adhere to one another plays a critical role in development, normal physiology, and disease processes such as inflammation. This ability is mediated by adhesion molecules, generally glycoproteins, expressed on cell membranes. Often, an adhesion molecule on one cell type will bind to another adhesion molecule expressed on a different cell type, forming a receptor counter-receptor pair. Three important classes of adhesion molecules are the integrins, selectins, and immunoglobulin (Ig) superfamily members (see Springer, Nature 346:425 (1990); Osborn, Cell 62:3 (1990);
Hynes, Cell 69:11 (1992)) These molecules are especially vital to the interaction of molecules are especially vital to the interaction of leukocytes and platelets with themselves and with the extracellular matrix and vascular endothelium.
P-selectin, also known as CD62, granule membrane protein-140 (GMP-140)jplatelet activation-dependent granule external membrane (PADGEM)JLECCAM-3 is a specialized cell-SUBST(TUME SHEET (RULE 26) surface receptor on vascular endothelial cells and platelets that is involved in the recognition of various circulating cells. P-selectin is a surface glycoprotein with a lectin-like domain, a region with homology to epidermal growth factor, and a region with homology to complement regulatory proteins (see McEver, Blood Cells 16:73-83 (1990)). The structure of P-selectin is similar to that of two other vascular cell surface receptors, endothelial leukocyte adhesion molecule (ELAM-1) and lymphocyte homing receptor (LHR). The term "selectin" has been suggested for this general class of receptors, because of their lectin-like domain and the selective nature of their adhesive functions.
P-selectin is present on the surface of platelets and endothelial cells in response to a variety of stimuli, where it mediates platelet-leukocyte and endothelium-leukocyte interactions. By contrast, ELAM-1 is expressed only on endothelial cells and LHR is expressed on a variety of leukocytes in endothelial venules of peripheral lymph nodes.
Expression of P-selectin is inducible and it is not expressed on unactivated endothelial cells or platelets. P-selectin expression does not require de novo synthesis since it is stored in secretory granules (or Weibel-Palade bodies) in both platelets and endothelial cells. Thus, within minutes of activation of either cell type by thrombin, histamine, or phorbol esters or other humoral factors, P-selectin is rapidly redistributed to the surface of the cell where it is accessible for binding to other cells.
A number of mouse antibodies to P-selectin have been reported (see, e.g., U.S. Patent No. 4,783,330, WO 91/06632, PCT Application No. FR90/00565, McEver et al., J. Biol. Chem.
259:9799-9804 (1984)); Geng et al., Nature 343:757-760;
Parmentier et al., Blood 77:1734-1739 (1991)). An additional antibody to P-selectin has been described in presentations at the Scripps Institute, La Jolla California and at a Keystone Colloquium held January 16, 1992. Certain of these antibodies that block P-selectin mediated adhesion are Ca++ dependent.
In vitro experiments employing antibodies to P-selectin have provided a limited understanding of the ligand SUBSTITUTE. SHEET (RULE 26) specificity of this receptor. It has been reported that P-selectin binds neutrophils, monocytes and certain carcinoma cells (Bevilacqua et al., Proc. Nat1. Acad. Sci. USA 84:9238-9242 (1987); Geng et al., Nature 343:757-760 (1990); Larsen et al., Cell 63:467-474 (1990); and Polley et al. It has been also been reported that P-selectin recognizes Lex as a high-affinity ligand (Larsen et al., Cell 63:467-474 (1990)).
However, subsequent experiments showed that SLe"-containing oligosaccharide is in fact a more potent inhibitor of P-selectin mediated adhesion than Lex (Polley et al., Proc.
Natl. Acad. Sci. USA 88:6224-6228 (1991)).
The participation of P-selectin in intercellular adhesion in vitro suggests that like other adhesion molecules, P-selectin might participate in cellular interactions that contribute to inflammatory diseases in vivo. However, the specific inflammatory diseases in which P-selectin might play a major role have not yet been elucidated. Similarly, the epitope-specificity of agents effective to abort particular P-selectin-mediated inflammatory diseases is also unclear. Of the known antibodies to P-selectin, some bind to P-selectin but do not block P-selectin mediated intercellular interactions. Other known antibodies have the capacity to block P-selectin-mediated intercellular binding only in the presence of high concentrations of Ca2+, a condition that may not always exist in vivo. The absence of high concentrations of CaZ+ in vivo would render these antibodies ineffective as therapeutic agents. Moreover, all antibodies to P-selectin produced to-date have been of murine origin and might therefore induce a human-antimouse antibody response (HAMA) in clinical use.
Based on the foregoing it is apparent that a need exists for novel agents targeted against P-selectin that have the appropriate epitope-specificity to abort intercellular interactions leading to inflammatory diseases in vivo.
Ideally such agents would not induce a HAMA response in treatment of humans. The present invention fulfills this and other needs.

SUBSTfTUTE SHEET (RULE 26) SUMMARY OF THE INVENTION
In one aspect of the invention, blocking antibodies that specifically bind to P-selectin are provided. One such exemplified antibody is designated mu MAb PB1.3 (produced by a cell line having ATCC accession number HB11041). Other antibodies competitively inhibit the binding of mu MAb PB1.3 to P-selectin measured by a competitive inhibition assay.
Preferred antibodies bind to P-selectin in the absence of Ca++. The antibodies of the invention include fragments such as Fab, Fab' F(ab')2, Fabc, and Fv, as well as intact antibodies.
In another aspect of the invention, pharmaceutical compositions are provided. The pharmaceutical compositions comprise a blocking P-selectin antibody as described above and a pharmaceutically acceptable carrier.
The invention also provides therapeutic methods for treating diseases of the immune system. The methods comprise administering to a patient having such a disease a therapeutically effective dose of a pharmaceutical composition as described above. Acute lung injury and ischemic reperfusion injury are examples of diseases treatable by the present invention.
Some antibodies of the invention are humanized immunoglobulins. The humanized immunoglobulins comprise a humanized heavy chain and a humanized light chain. The humanized light chain comprises three complementarity determining regions (CDR1, CDR2 and CDR3) having amino acid sequences from the corresponding complementarity determining regions of a mouse PB1.3 immunoglobulin light chain, and a variable region framework from a human kappa light chain variable region framework sequence. The humanized heavy chain comprises three complementarity determining regions (CDR1, CDR2 and CDR3) having amino acid sequences from the corresponding complementarity determining regions of a mouse PB1.3 immunoglobulin heavy chain, and a variable region framework from a human heavy chain variable region framework sequence. The humanized immunoglobulins specifically bind to P-selectin with a binding affinity havina a lower limit of SUBSTITUTE SHEET (RULE 26) about 107 M-1 and an upper limit of about five-times the binding affinity of the mouse PB1.3 immunoglobulin.
In many humanized immunoglobulins, the human kappa light chain variable region framework sequence is substituted 5 in at least one position selected from a first group consisting of L21, L46, L60 and L70, by an amino acid present in the equivalent position of the mouse PB1.3 immunoglobulin light chain variable region framework sequence. Similarly, the human heavy chain variable region framework sequence is substituted in at least one position selected from a second group consisting of H1, H2, H71 and H73, by an amino acid present in the equivalent position of the mouse PB1.3 heavy chain variable region framework sequence. In some humanized immunoglobulins, the humanized heavy chain variable region framework is a 21/28'CL heavy chain variable region framework sequence. In some humanized immunoglobulins, the humanized light chain variable region framework is a DEN light chain variable region framework sequence. Exemplary mature light chain variable region amino acid sequences are shown in Figures 14-17. Exemplary mature heavy chain variable region amino acid sequences are shown in Figures 11-13. A preferred humanized immunoglobulin comprises the mature light chain variable region shown in Figure 17 and the mature heavy chain variable region shown in Figure 12.
The invention provides intact humanized immunoglobulins and humanized immunoglobulin fragments such as Fab, Fab' F(ab')2, Fabc and Fv. Some humanized immunoglobulins further comprise a constant region, which may or may not have an effector function.
In another aspect of the invention, nucleic acids encoding the humanized immunoglobulins described above are provided. Some nucleic acid encode heavy chains and others encode light chains.
The invention also provides computers programmed to display a three dimensional representation of the immunoglobulins described above on a monitor or printer.

SUBSTIME SHEET (RULE 26) 6 z162149 The invention further provides pharmaceutical compositions comprising the humanized immunoglobulins described above and a pharmaceutically acceptable carrier.
In another aspect, the invention provides methods for detecting P-selectin using the humanized immunoglobulins described above. A humanized immunoglobulin is administered to a patient or a tissue sample therefrom. Complexes formed by specific binding between the immunoglobulin and P-selectin present in the target sample are detected to indicate the presence of P-selectin.
The invention also provides methods of treating a patient suffering from a disease of the immune systems using humanized immunoglobulins. The methods comprise administering to the patient a therapeutically effective dose of a pharmaceutical composition as described above. Diseases amenable to treatment include ischemia-reperfusion injury and acute lung injury. For example, the compositions are useful for treating patients suffering from epidermal, myocardial, renal, cerebral, splenic, hepatic, spinal, splanchnic, pulmonary, partial-body, or whole-body ischemia.
In a further aspect of the invention, stable cell lines producing humanized immunoglobulins are provided. The stable cell lines comprise two nucleic acids segments. One nucleic acid segment encodes the heavy chain of a humanized immunoglobulin as described above, the segment operably linked to a promoter to allow expression of the heavy chain. The second nucleic acid segment encodes the light chain of the humanized immunoglobulin, the second segment operably linked to a second promoter to allow expression of the light chain.
Preferred cell lines produce about 30 g of the humanized immunoglobulin/106 cells/day.
In another aspect of the invention, fragments of P-selectin are provided. These fragments have up to 100 amino acids, the up to 100 amino acids comprising at least five contiguous amino acids from between positions 448 and 467 of the amino sequence shown in Fig. 34. In preferred fragments, the up to 100 amino acids comprise the entire contiguous segment of amino acids between positions 448 and 467.

SUBSTITUTE SHEET (RULE 26) ~16 2 14 9 BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 presents data showing that anti-inflammatory immunoglobulins of the invention inhibit binding of thrombin activated platelets to neutrophils.
Figs. 2A and 2B show that anti-inflammatory immunoglobulins of the invention effectively prevent lung injury induced by infusion of cobra venom factor (CVF) as measured by permeability (Fig. 2A) and hemorrhage (Fig 2B).
Fig. 3A through 3H show that P-selectin expression content in lungs of animals is upregulated in response to cobra venom infusion visualized by horseradish proxidase stain of mu MAb PB1.3 panel (a)-(d) show expression of P-selectin in lung venules at various times after infusion of CVF; time 0 (a); 5, 10 and 15 min, panels b, c and d, respectively (Fig.
3A). Fig. 3B are light and transmission electron micrographs of CVF-induced acute lung injury in rats treated with 200 g non-blocking antibody (mu MAb PNB1.6) (frames a.c) or with blocking P-selectin antibody (mu MAb PB1.3) to P-selectin (frames b.d). In animals treated with PNB1.6, vascular injury was indicated by extensive intra-alveolar hemorrhage (frame a), associated with intravascular aggregates of neutrophils (frame c. arrows) in close contrast with endothelial cells.
In contrast, in animals treated with mu MAb PB1.3 intra-alveolar hemorrhage was absent (frame b) and intravascular neutrophils showed little contact with the endothelium (frame d. arrows). (Frames a and b, plastic embedded sections stained with toluidine blue, X160: frames c and d, uranyl acetate, lead citrate stained X2250.) Fig. 4 shows the binding of the monoclonal antibody mu MAb PB1.3 to P-selectin: effect of chelating divalent cations with EDTA. In the plot the open symbols (lower trace) is binding in the presence of Ca++ and Mg++, while the filled symbols (upper trace) is binding in the presence of EDTA.
Antibody was diluted shown at 1.6 mg/ml.
Fig. 5 shows the effect of the peptide CQNRYTDLVAIQNKNE on the binding of mu MAb PB1.3 to P-selectin.
Filled symbols are in the presence of 0.35 mg/ml peptide, open SUBSTIME SHEET (RULE 26) symbols are in the absence of peptides. Indicated dilutions of mu MAb PB1.3 were from a 1.6 mg/mi solution.
Figs. 6A, 6B and 6C show the ability of monoclonal antibodies to cross-block binding to P-selectin. Blocking antibodies were present at 50 g/ml. Dilutions of biotinylated antibodies shown were from 1.6 mg/ml mu MAb PB1.3, 0.87 mu MAb mg/ml PNB1.6, and 0.59 mg/ml mu MAb 84/26.
Fig. 7 shows that the extent of platelet aggregation induced by thrombin is equivalent under control conditions in the absence of antibody and in the presence of increasing concentrations (1-100 g/ml) of purified mu MAb PB1.3 antibody.
Fig. 8 shows that the extent of myocardial ischemia, expressed as a percentage of the area at risk, is significantly less in animals treated with the blocking anti-P-selectin antibody (mu MAb PB1.3) than in animals treated with the non-blocking anti-P-selectin antibody. Furthermore, the ability of coronary artery rings to relax in response to acetylcholine is significantly greater in mu MAb PB1.3 compared with mu MAb PNB1.6 treated animals. That these phenomena are attributable to a reduction in myocardial leukocyte infiltration is suggested by the reduced number of PMNs in myocardial tissue from mu MAb PB1.3 treated animals compared with PNB1.6 treated animals.
Fig. 9 shows the sequence and translation of PB1.3 heavy chain signal peptide and variable region.
Fig. 10 shows the sequence and translation of PB1.3 light chain signal peptide and variable region.
Fig. 11 shows the sequence and translation of CY1748 heavy chain-A signal peptide and variable region.
Fig. 12 shows the sequence and translation of CY1748 heavy chain-B signal peptide and variable region.
Fig. 13 shows the sequence and translation of CY1748 heavy chain-C signal peptide and variable region.
Fig. 14 shows the sequence and translation of CY1748 light chain-A signal peptide and variable region.
Fig. 15 shows the sequence and translation of CY1748 light chain-B signal peptide and variable region.

SUBSTITUTE SHEET (RULE 26) Fig. 16 shows the sequence and translation of CY1748 light chain-C signal peptide and variable region.
Fig. 17 shows the sequence and translation of CY1748 light chain-D signal peptide and variable region.
Fig. 18 shows physical,map of the neo-plasmid expressing the human heavy chain (gamma-1 constant region).
Fig. 19 shows the physical map of the neo-plasmid expressing the human light chain (kappa constant region).
Fig. 20 shows the binding curves of the chimeric PB1.3 and reshaped versions of CY1748 to rsP-Selectin.
Fig. 21 shows the binding curves of the chimeric PB1.3 and reshaped versions of CY1748 to rsP-Selectin.
Fig. 22 shows the binding curves of the chimeric PB1.3 and reshaped versions of CY1748 to rsP-Selectin.
Fig. 23 shows the physical map of the dhfr-plasmid expressing the human heavy chain (gamma-1 constant region).
Fig. 24 shows the physical map of the dhfr-plasmid expressing the human heavy chain (gamma-4 constant region).
Fig. 25 shows the inhibition of biotinylated-PB1.3 binding to rsP-Selectin by PB1.3 and CY1748 R$A/RLA-IgGl.
Fig. 26 shows the inhibition of biotinylated-PB1.3 binding to rsP-Selectin by PB1.3 and CY1748 RHB/Ri,D-IgG4.
Fig. 27 shows that antibody PB1.3 reduces swelling associated with ischemia and reperfusion in the rabbit ear.
Fig. 28 shows that antibody PB1.3 prevents the development of necrosis following ischemia and reperfusion of the rabbit ear.
Fig. 29 shows that antibody PB1.3 inhibits leukocyte-endothelial cell interactions induced by the mast cell degranulating agent compound 48/80.
Fig. 30 shows the proposed folding pattern of the domains of P-selectin.
Fig. 31 shows a schematic of the deletion mutants of P-selectin used to map the epitope that PB1.3 recognizes.
Fig. 32 shows the binding of PB1.3 and rabbit polyclonal antibodies to P-selectin deletion mutants.
SUBSTITUTE SHEET (RULE 26) Fig. 33 shows the PB1.3 and P6H6 binding to P-selectin and to P-selectin 05CRP which is a deletion mutant of P-selectin that lacks the fifth CRP.
Fig. 34 shows the sequences of synthetic peptides 5 derived from the amino acid sequence of the human P-selectin 5th CRP and the PB1.3 reactivity to these immobilized peptides. Amino acids are numbered according to the convention of Johnston et al., Cell, 56:1033-1044 (1989) for a full-length P-selectin polypeptide.
DEFINITIONS
Abbreviations for the twenty naturally occurring amino acids follow conventional usage (Immunology - A
Synthesis (2nd ed., E.S. Golub & D.R. Gren, eds., Sinauer Associates, Sunderland, MA, 1991)). Stereoisomers (e.g., D-amino acids) of the twenty conventional amino acids, unnatural amino acids such as a,a-disubstituted amino acids, N-alkyl amino acids, lactic acid, and other unconventional amino acids may also be suitable components for polypeptides of the present invention. Examples of unconventional amino acids include: 4-hydroxyproline, 7-carboxyglutamate, e-N,N,N-trimethyllysine, e-N-acetyllysine, 0-phosphoserine, N-acetylserine, N-formylmethionine, 3-methylhistidine, 5-hydroxylysine, w-N-methylarginine, and other similar amino acids and imino acids (e.g., 4-hydroxyproline). Moreover, amino acids may be modified by glycosylation, phosphorylation and the like.
In the polypeptide notation used herein, the lefthand direction is the amino terminal direction and the righthand direction is the carboxy-terminal direction, in accordance with standard usage and convention. Similarly, unless specified otherwise, the lefthand end of single-stranded polynucleotide sequences is the 5' end; the lefthand direction of double-stranded polynucleotide sequences is referred to as the 5' direction. The direction of 5' to 3' addition of nascent RNA transcripts is referred to as the transcription direction; sequence regions on the DNA strand having the same sequence as the RNA and which are 5' to the 51 SUBSTITUTE SHEET (RULE 26) Zi6%149 end of the RNA transcript are referred to as "upstream sequences"; sequence regions on the DNA strand having the same sequence as the RNA and which are 3' to the 3' end of the RNA
transcript are referred to as "downstream sequences".
The phrase "polynucleotide sequence" refers to a single or double-stranded polymer of deoxyribonucleotide or ribonucleotide bases read from the 5' to the 3' end. It includes self-replicating plasmids, infectious polymers of DNA
or RNA and non-functional DNA or RNA.
The following terms are used to describe the sequence relationships between two or more polynucleotides:
"reference sequence", "comparison window", "sequence identity", "percentage of sequence identity", and "substantial identity". A "reference sequence" is a defined sequence used as a basis for a sequence comparison; a reference sequence may be a subset of a larger sequence, for example, as a segment of a full-length cDNA or gene sequence given in a sequence listing, such as a polynucleotide sequence of Figs. 9-17, or may comprise a complete DNA or gene sequence. Generally, a reference sequence is at least 20 nucleotides in length, frequently at least 25 nucleotides in length, and often at least 50 nucleotides in length. Since two polynucleotides may each (1) comprise a sequence (i.e., a portion of the complete polynucleotide sequence) that is similar between the two polynucleotides, and (2) may further comprise a sequence that is divergent between the two polynucleotides, sequence comparisons between two (or more) polynucleotides are typically performed by comparing sequences of the two polynucleotides over a "comparison window" to identify and compare local regions of sequence similarity. A "comparison window", as used herein, refers to a conceptual segment of at least 20 contiguous nucleotide positions wherein a polynucleotide sequence may be compared to a reference sequence of at least 20 contiguous nucleotides and wherein the portion of the polynucleotide sequence in the comparison window may comprise additions or deletions (i.e., gaps) of 20 percent or less as compared to the reference sequence (which does not comprise additions or deletions) for optimal SUBSTITUTE=SNEET (RULE 26) r~ .

alignment of the two sequences. Optimal alignment of sequences for aligning a comparison window may be conducted by the local homology algorithm of Smith & Waterman, Adv. Appl.
Math. 2:482 (1981), by the homology alignment algorithm of Needleman & Wunsch, J. Mol. Biol. 48:443 (1970), by the search for similarity method of Pearson & Lipman, Proc. Natl. Acad.
Sci. (USA) 85:2444 (1988), by computerized implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package Release 7.0, Genetics Computer Group, 575 Science Dr., Madison, WI), or by inspection, and the best alignment (i.e., resulting in the highest percentage of sequence similarity over the comparison window) generated by the various methods is selected. The term "sequence identity" means that two polynucleotide sequences are identical (i.e., on a nucleotide-by-nucleotide basis) over the window of comparison. The term "percentage of sequence identity" is calculated by comparing two optimally aligned sequences over the window of comparison, determining the number of positions at which the identical nucleic acid base (e.g., A, T, C, G, U, or I) occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the window of comparison (i.e., the window size), and multiplying the result by 100 to yield the percentage of sequence identity. The terms "substantial identity" as used herein denotes a characteristic of a polynucleotide sequence, wherein the polynucleotide comprises a sequence that has at least 85 percent sequence identity, preferably at least 90 to 95 percent sequence identity, more usually at least 99 percent sequence identity as compared to a reference sequence over a comparison window of at least 20 nucleotide positions, frequently over a window of at least 25-50 nucleotides, wherein the percentage of sequence identity is calculated by comparing the reference sequence to the polynucleotide sequence which may include deletions or additions which total 20 percent or less of the reference sequence over the window of comparison. The reference sequence may be a subset of a larger sequence, for example, the sequence shown in Figs. 10-17 .
As applied to polypeptides, the term "sequence identity" means peptides share identical amino acids at corresponding positions. The term "sequence similarity" means peptides have identical or similar amino acids (i.e., conservative substitutions) at corresponding positions. The term "substantial identity" means that two peptide sequences, when optimally aligned, such as by the programs GAP or BESTFIT
using default gap weights, share at least 80 percent sequence identity, preferably at least 90 percent sequence identity, more preferably at least 95 percent sequence identity or more (e.g., 99 percent sequence identity). Preferably, residue positions which are not identical differ by conservative amino acid substitutions. The term "substantial similarity" means that two peptide sequences share corresponding percentages of sequence similarity.
The term "substantially pure" means an object species is the predominant species present (i.e., on a molar basis it is more abundant than any other individual species in the composition), and preferably a substantially purified fraction is a composition wherein the object species comprises at least about 50 percent (on a molar basis) of all macromolecular species present. Generally, a substantially pure composition will comprise more than about 80 to 90 percent of all macromolecular species present in the composition. Most preferably, the object species is purified to essential homogeneity (contaminant species cannot be detected in the composition by conventional detection methods) wherein the composition consists essentially of a single macromolecular species.
For purposes of classifying amino acids substitutions as conservative or nonconservative, amino acids are grouped as follows: Group I (hydrophobic sidechains):
norleucine, met, ala, val, leu, ile; Group II (neutral hydrophilic side chains): cys, ser, thr; Group III (acidic side chains): asp, glu; Group IV (basic side chains): asn, gln, his, lys, arg; Group V (residues influencing chain SUBSTITUTE SHEET (RULE 26) oriantaticn): gly, pro; and Group VI (aromatic side chains):
trY, tyr, phe. Conservative substitutions involve substitutions between amino acids in the same class. Non-conservative substitutions constitute exchanging a member of one of these classes for another.
Amino acids from the variable regions of the mature heavy and light chains of immunoglobulins are designated Hx and Lxx respectively, where x is a number designating the position of an amino acids according to the scheme of Kabat et al., Sequences of Proteins of Immunological Interest (National Institutes of Health, Bethesda, MD (1987) and (1991)) (hereinafter collectively referred to as "Kabat et al.,").
Kabat et al. list many amino acid sequences for antibodies for each subclass, and list the most commonly occurring amino acid for each residue position in that subclass. Kabat et al. use a method for assigning a residue number to each amino acid in a listed sequence, and this method for assigning residue numbers has become standard in the field. ICabat et al.'s scheme is extendible to other antibodies not included in the compendium by aligning the antibody in question with one of the consensus sequences in Kabat et al. The use of the Kabat et al. numbering system readily identifies amino acids at equivalent positions in different antibodies. For example, an amino acid at the L50 position of a human antibody occupies the equivalence position to an amino acid position L50 of a mouse antibody.
"Immunoglobulin," "antibody" or "antibody peptide(s)" refers to an intact antibody or a binding fragment thereof that competes with the intact antibody for specific binding to P-selectin.
"Substantial inhibition" means at least about 50%
inhibition, preferably about 60% to about 80%, and more usually about greater than 85% or more (as measured in an in vitro competitive binding assay).
Chimeric antibodies are antibodies whose light and heavy chain genes have been constructed, typically by genetic engineering, from immunoglobulin gene segments belonging to SUBSTIME SHEET (RULE 26) ciifferent species. For example, the variable (V) segments of the genes from a mouse monoclonal antibody may be joined to human constant (C) segments, such as 71 and 74. A typical therapeutic chimeric antibody is thus a hybrid protein 5 consisting of the V or antigen-binding domain from a mouse antibody and the C or effector domain from a human antibody, although other mammalian species may be used.

DESCRIPTION OF THE PREFERRED EMBODIMENT

10 This invention provides compositions and methods for inhibiting diseases and conditions of the immune system mediated by P-selectin. Specifically, the invention provides immunoglobulins which have the ability to inhibit P-selectin-mediated adhesion of cells in vivo.
I. Characteristics of Immunoclobulins A. General The basic antibody structural unit is known to comprise a tetramer. Each tetramer is composed of two iden-tical pairs of polypeptide chains, each pair having one "light" (about 25 kDa) and one "heavy" chain (about 50-70 kDa). The amino-terminal portion of each chain includes a variable region of about 100 to 110 or more amino acids primarily responsible for antigen recognition. The carboxy-terminal portion of each chain defines a constant region primarily responsible for effector function.
Light chains are classified as either kappa or lambda. Heavy chains are classified as gamma, mu, alpha, delta, or epsilon, and define the antibody's isotype as IgG, IqM, IgA, IgD and IgE, respectively. Within light and heavy chains, the variable and constant regions are joined by a",7"
region of about 12 or more amino acids, with the heavy chain also including a"D" region of about 10 more amino acids.
(See generally, Fundamental Immunology (Paul, W., ed., 2nd ed.
Raven Press, N.Y., 1989), Ch. 7 The variable regions of each light/heavy chain pair form the antibody binding site. The chains all exhibit the SUBSTiTUTE SHEET (RULE 26) 'O 94/2-506 7 PCTIUS94ru4935 *6 same general structure of relatively conserved framework regions (FR) joined by three hypervariable regions, also called complementarity determining regions or CDRs. The CDRs from the two chains of each pair are aligned by the framework regions, enabling binding to a specific epitope. CDR and FR
residues are delineated according to the standard sequence definition of Kabat et al., supra. An alternative structural definition has been proposed by Chothia et al., J. Mol. Biol.
196:901-917 (1987); Nature 342:878-883 (1989); and J. Mol.
Bio1. 186:651-663 (1989) (hereinafter collectively referred to as "Chothia et al.").

B. Binding Specificity and Affinity The immunoglobulins (or.antibodies) of the invention selectively bind a functional epitope on P-selectin associated with a response to tissue injury and inflammation. Usually, binding of the antibodies to a functional epitope on P-selectin effectively inhibits adhesion of leukocytes to activated platelets and/or to the activated vascular endothelium in vivo. Antibodies showing this property are referred to as "blocking" antibodies. Preferred blocking antibodies impair the adhesion of leukocytes to the activated vascular endothelium to prevent or inhibit an inflammatory and/or thrombotic condition.
Many of the blocking antibodies of the invention compete with an exemplary antibody designated mu MAb PB1.3 for specific binding to P-selectin. A hybridoma producing the mu MAb PB1.3 antibody has been deposited with the American Type Culture Collection, Rockville, Maryland under the Budapest Treaty on May 14,1992 and given the Accession No. HB11041. The _ production of this monoclonal antibody is described in Example 1. Other blocking antibodies of the invention compete with a second exemplified antibody designated 84/26.
Competition is determined by an assay in which the immunoglobulin under test inhibits specific binding of a reference antibody (e.g., mu MAb PB1.3) to an antigenic determinant on a P-selectin molecule. Numerous types of SUBSTITUTE SHEET (RULE 26) competitive binding assays are known for example: solid phase direct or indirect radioimmunoassay (RIA), solid phase direct or indirect enzyme immunoassay (EIA), sandwich competition assay (see Stahli et al., Methods in EnzymoZogy 9:242-253 (1983)); solid phase direct biotin-avidin EIA (see Kirkland et al., J. Immunol. 137:3614-3619 (1986)); solid phase direct labeled assay, solid phase direct labeled sandwich assay (see Harlow and Lane, "Antibodies, A Laboratory Manual," Cold Spring Harbor Press (1988)); solid phase direct label RIA
using 1-125 label (see Morel et al., Molec. Immunol. 25(l):7-(1988)); solid phase direct biotin-avidin EIA (Cheung et al., Virology 176:546-552 (1990)); and direct labeled RIA
(Moldenhauer et al., Scand. J. Immunol. 32:77-82 (1990)).
Typically, such an assay involves the use of purified P-15 selectin or cells bearing P-selectin bound to a solid surface, an unlabelled test immunoglobulin and a labelled reference immunoglobulin. Competitive inhibition is measured by determining the amount of label bound to the solid surface in the presence of the test immunoglobulin. Usually the test immunoglobulin is present in excess. An example of a suitable competitive binding assay is presented in Example S.
Antibodies identified by competition assay (competing antibodies) include antibodies binding to the same epitope as the reference antibody and antibodies binding to an adjacent epitope sufficiently proximal to the epitope bound by the reference antibody for steric hindrance to occur. Usually, when a competing antibody is present in excess, it will inhibit specific binding of a reference antibody to P-selectin by at least 10, 25, 50 or 75%.
The epitope bound by mu MAb PB1.3 has been mapped to an epitope between amino acids 448 and 467 of human P-selectin. See Figure 34 and Johnson et al., Cell 56:1033-1044 (1989) The epitope occurs within the fifth CRP domain of P-selectin. See Figure 30. Antibodies that compete with mu MAb PB1.3 usually bind to a segment of amino acids between 448 and 467 or to another segment within the fifth CRP domain.

SUBSTIME SHEET (RULE 26) The binding specificity of many blocking antibodies of the invention is further defined by their capacity to bind P-selectin in the complete or substantial absence of Ca++
(e.g., in the presence of 25 mM EDTA (a calcium chelator) and the absence of Ca++ in an in vitro assay). Antibodies requiring a Ca++ cofactor for blocking activity may not be effective in in vivo conditions where levels of Ca++ are expected to fluctuate.
Antibodies that compete with the mu MAb PB1.3 antibody are further characterized by the observation that their capacity to specifically bind to P-selectin is not inhibited by a fragment of P-selectin having the amino acid sequence CQNRYTDLVAIQNKNE. A molar excess of this fragment usually inhibits specific binding of such antibodies by less than 5%'. Mu MAb PB1.3 and competing antibodies are thereby distinguished from reported mouse antibodies designated Gi, G2 and G3 (Geng et al., The Journal of Biological Chemistry 266:22313-22318 (1991)) whose binding to P-selectin is partially or completely blocked by the above peptide.
Some antibodies of the invention are further characterized by their capacity to selectively inhibit some P-selectin mediated responses while leaving others unimpaired.
For example, some antibodies are capable of blocking specific binding of neutrophils to activated endothelial cells in vivo, without inhibiting platelet aggregation. Such antibodies are particularly useful for treating inflammatory disorders without impairing a patient's capacity to initiate a response to conditions (e.g., wounding) in which platelet aggregation is desirable.
The antibodies of the invention usually exhibit a specific binding affinity for P-selectin of at least 107, 108, 109, or 1010 M-l. Usually the upper limit of binding affinity of the antibodies for P-selectin is within a factor of about three, five or ten of that of the mu MAb PB 1.3. Often the lower limit of binding affinity is also within a factor of about three, five or ten of that of mu MAb PB1.3. In this context, the term "about" encompasses the small degree of SUBSTITUTE SHEET (RULE 26) WO 94125067 PCT(US94/04935 ~,.

experimental error that may typically occur in the measurement of binding affinities.
II. Production of Iminunoalobulins A. Nonhuman Antibodies The production of non-human monoclonal antibodies, e.g., murine, lagomorpha, equine is well known and can be accomplished by, for example, immunizing an animal with a preparation containing cells bearing P-selectin (e.g., thrombin-activated platelets), isolated P-selectin molecules or fragments thereof, such as extracellular domains.
Antibody-producing cells obtained from the immunized animals are immortalized and screened, or screened first for the production of antibody which binds to P-selectin, and then immortalized. See Harlow and Lane, supra. Alternatively, substantially monospecific antibody populations can be produced by chromatographic purification of polyclonal sera.
B. Human Antibodies Against P-Selectin In another aspect of the invention, human antibodies against P-selectin are provided. Some human antibodies are selected by competitive binding experiments, or otherwise, to have the same epitope specificity as a particular mouse antibody, such as mu MAb PB1.3. Such antibodies are particularly likely to share the useful therapeutic properties demonstrated for mu MAb PB1.3. Human antibodies to P-selectin can be produced by screening a DNA library from human B cells according to the general protocol outlined by Huse et al., Science 246:1275-1281 (1989). Antibodies binding to P-selectin or a fragment thereof are selected. Sequences encoding such antibodies (or a binding fragments) are then cloned and amplified. The protocol described by Huse is rendered more efficient in combination with phage-display technology. See, e.g., Dower et al., WO 91/17271 and McCafferty et al., WO 92/01047.
'In these methods, libraries of phage are produced in which members display different antibodies on their outersurfaces.
Antibodies are usually displayed as Fv or Fab fragments.

SUBSTITUTE SHEET (RULE 26) 20 ~162 i U

Phage displaying antibodies with a desired specificity are selected by affinity enrichment to an P-selectin polypeptide or fragment thereof.
In a variation of the phage-display method, human antibodies having the binding specificity of a selected murine antibody can be produced. See Winter, WO 92/20791. In this method, either the heavy or light chain variable region of the selected murine antibody (e.g., mu MAb PB1.3) is used as a starting material. If, for example, a light chain variable region is selected as the starting material, a phage library is constructed in which members displays the same light chain variable region (i.e., the murine starting material) and a different heavy chain variable region. The heavy chain variable regions are obtained from a library of rearranged human heavy chain variable regions. A phage showing strong specific binding for P-selectin (e.g., at least 108 and preferably at least 109 M-1) is selected. The human heavy chain variable region from this phage then serves as a starting material for constructing a further phage library.
In this library, each phage displays the same heavy chain variable region (i.e., the region identified from the first display library) and a different light chain variable region.
The light chain variable regions are obtained from a library of rearranged human variable light chain regions. Again, phage showing strong specific binding for L-selectin are selected. These phage display the variable regions of completely human anti-P-selectin antibodies. These antibodies usually have the same or similar epitope specificity as the murine starting material (e.g., mu MAb PB1.3).
C. Humanized Antibodies The invention provides humanized immunoglobulins having variable framework regions substantially from a human immunoglobulin (termed an acceptor immunoglobulin) and complementarity determining regions substantially from a mouse immunoglobulin termed mu MAb PB1.3 (referred to as the donor immunoglobulin). The constant region(s), if present, are also substantially from a human immunoglobulin. The humanized SUgST1TUTE SHEET (RULE 26) antibodies of the present invention offer several advantages over the mouse donor antibody, which has already shown to be effective in animals models:
1) The human immune system should not recognize the framework or constant region of the humanized antibody as foreign, and therefore the antibody response against such an injected antibody should be less than against a totally foreign mouse antibody or a partially foreign chimeric antibody.
2) Because the effector portion of the humanized antibody is human, it may interact better with other parts of the human immune system.
3) Injected mouse antibodies have been reported to have a half-life in the human circulation much shorter than the half-life of normal human antibodies (Shaw et al., J.
Immunol. 138:4534-4538 (1987)). Injected humanized antibodies have a half-life essentially equivalent to naturally occurring human antibodies, allowing smaller and less frequent doses.

1. Clonina and Sgauenciaa Variable Domains of mu MAb PB1.3 The cloning and sequencing of cDNA encoding the mu MAb PB1.3 antibody heavy and light chain variable regions is described in Example 10, and the nucleotide and predicted amino acids sequences are shown in Figs. 11 and 12. Tables 1 and 2 illustrate the subdivision of the amino acid coding sequence into framework and complementarity determining domains. From N-terminal to C-terminal, both light and heavy chains comprise the domains FR1, CDR1, FR2, CDR2, FR3, CDR3 and FR4. The assignment of amino acids to each domain is in accordance with the numbering convention of Kabat et al., supra.

(2) Selection of Human Antibodies to Supply Framework Residues The substitution of mouse CDRs into a human variable domain framework is most likely to result in retention of their correct spatial orientation if the human variable domain SUBSTITUTE SHEET (RULE 26) Z 1bZ?49 framework adopts the same or similar conformation to the mouse variable framework from which the CDRs originated. This is achieved by obtaining the human variable domains from human antibodies whose framework sequences exhibit a high degree of sequence identity with the murine variable framework domains from which the CDRs were derived. The heavy and light chain variable framework regions can be derived from the same or different human antibody sequences. The human antibody sequences can be the sequences of naturally occurring human antibodies or can be consensus sequences of several human antibodies. See Kettleborough et al., Protein Engineering 4:773 (1991); Kolbinger et al., Protein Engineering 6:971 (1993).
Suitable human antibody sequences are identified by computer comparisons of the amino acid sequences of the mouse variable regions with the sequences of known human antibodies.
The comparison is performed separately for heavy and light chains but the principles are similar for each. This comparison reveals that the mu MAb PB1.3 light chain shows greatest sequence identity to human light chains of subtype kappa-1 and that the mu PB1.3 heavy chain shows greatest sequence identity to human heavy chains of subtype 1, as defined by Kabat et al., supra. Thus, light and heavy human framework regions are usually derived from human antibodies of these subtypes, or from consensus sequences of such subtypes.
The preferred heavy and light chain human variable regions showing greatest sequence identity to the corresponding regions from mu MAb PB1.3 are from antibodies 21/28'CL and DEN
respectively.
(3) Computer Modelling The unnatural juxtaposition of murine CDR regions with human variable framework region can result in unnatural conformational restraints, which, unless corrected by substitution of certain amino acid residues, lead to loss of binding affinity. The selection of amino acid residues for substitution is determined, in part, by computer modelling.
Computer hardware and software for producina three-dimensional SUBSTITUTE SHEET (RULE 26) WO 94/25067 ""-/US94/04935 images of immunoglobulin molecules are widely available. In general, molecular models are produced starting from solved structures for immunoglobulin chains or domains thereof. The chains to be modelled are compared for amino acid sequence similarity with chains or domains of solved three dimensional structures, and the chains or domains showing the greatest sequence similarity is/are selected as starting points for construction of the molecular model. For example, for the light chain of mu MAb PB1.3, the starting point for modelling was the human light chain DEN. The solved starting structures are modified to allow for differences between the actual amino acids in the immunoglobulin chains or domains being modelled, and those in the starting structure. The modified structures are then assembled into a composite immunoglobulin. Finally, the model is refined by energy minimization and by verifying that all atoms are within appropriate distances from one another and that bond lengths and angles are within chemically acceptable limits. Such a model can in turn serve as a starting point for predicting the three-dimensional structure of an antibody containing the mu MAb PB1.3 complementarity determining regions substituted in human framework structures.
Additional models can be constructed representing the structure when further amino acid substitutions to be discussed infra, are introduced.
(4) Substitution of Amino Acid Residues As noted supra, the humanized antibodies of the invention comprise variable framework regions substantially from a human immunoglobulin and complementarity determining regions substantially from a mouse immunoglobulin termed mu MAb PB1.3. Having identified the complementarity determining regions of mu MAb PB1.3 and appropriate human acceptor immunoglobulins, the next step is to determine which, if any, residues from these components should be substituted to optimize the properties of the resulting humanized antibody.
In general, substitution of human amino acid residues with murine should be minimized, because introduction of murine residues increases the risk of the antibodv elicitina a HAMA

SUBSTITUTE SHEET (RULE 26) IL1 6?149 response in humans. Amino acids are selected for substitution based on their possible influence on CDR conformation and/or binding to antigen. Investigation of such possible influences is by modelling, examination of the characteristics of the amino acids at particular locations, or empirical observation of the effects of substitution or mutagenesis of particular amino acids.
When an amino acid differs between a mu MAb PB1.3 variable framework region and an equivalent human variable framework region, the human framework amino acid should usually be substituted by the equivalent mouse amino acid if it is reasonably expected that the amino acid:
(1) noncovalently binds antigen directly (e.g., amino acids at positions H1, H2, and L60 of mu MAb PB1.3) (2) is adjacent to a CDR region, is part of a CDR
region under the alternative definition proposed by Chothia et al., supra, or otherwise interacts with a CDR region (e.g., is within about 3 A of a CDR region) (e.g., H73 of mu MAb PB1.3, Tables 1 and 3), or (3) participates in the VL-VH interface.
Other candidates for substitution are acceptor human framework amino acids that are unusual for a human immunoglobulin at that position. These amino acids can be substituted with amino acids from the equivalent position of more typical human immunoglobulins. Alternatively, amino acids from equivalent positions in mu MAb PB1.3 can be introduced into the human framework regions when such amino acids are typical of human immunoglobulin at the equivalent positions.
In general, substitution of all or most of the amino acids fulfilling the above criteria is desirable.
occasionally, however, there is some ambiguity about whether a particular amino acid meets the above criteria, and alternative variant immunoglobulins are produced, one of which has that particular substitution, the other of which does not.
surprisingly, an exemplified humanized antibody designated reshaped MAb PB1.3 (HBLD) contains only one substitution of the human heavy-chain variable region framework residues and SgSTITUTE SHEET (RULE 26) only two substitutions of the human light-chain variable region framework residues and yet exhibits substantially similar (within a factor of two) binding affinity to a murine or chimeric antibody having intact murine variable regions fused to a human constant region (designated chi MAb PB1.3).
Some humanized antibodies of the invention contain a substitution of a human light chain framework residue with a corresponding mu MAb PB1.3 residue in at least 1, 2 or 3, and sometimes 4 of the following positions: L21, L46, L60 and L70.
(See Tables 2 and 4). Some humanized antibodies usually contain a substitution of a human heavy chain framework residue in at least 1, 2 or 3 and sometimes 4 of the following positions Hi, H2, and H71 and H73. (See Tables 1 and 3) Many humanized antibodies contain substitutions of both light and heavy chain variable framework regions. Preferred antibodies contain a substitution of a human heavy chain framework residue at least at position H71, and substitutions of human light chain framework residues at least at positions 21 and 46.
Usually the CDR regions in humanized antibodies are substantially identical, and more usually, identical to the corresponding CDR regions in the mu MAb PB1.3 antibody.
Although not usually desirable, it is sometimes possible to make one or more conservative amino acid substitutions of CDR
residues without appreciably affecting the binding affinity of the resulting humanized immunoglobulin. Occasionally, substitution of CDR regions can result in enhanced binding affinity.
Other than for the specific amino acid substitutions discussed above, the framework regions of humanized immunoglobulins are usually substantially identical, and more usually, identical to the framework regions of the human antibodies from which they were derived. Of course, many of the amino acids in the framework region make little or no direct contribution to the specificity or affinity of an antibody. Thus, many individual conservative substitutions of framework residues can be tolerated without appreciable change of the specificity or affinity of the resulting humanized SUBSTITUTE SHEET (RULE 26) WO 94/250~7 PCT/US94104935 imr::noglobulin. However, in general, such substitutions are unaesirable.

(5) Production of Variable Reaions Having conceptually selected the CDR and framework coWponents of humanized immunoglobulins, a variety of methods are available for producing such immunoglobulins. Because of the degeneracy of the code, a variety of nucleic acid sequences will encode each immunoglobulin amino acid sequence.
The desired nucleic acid sequences can be produced by de novo solid-phase DNA synthesis or by PCR mutagenesis of an earlier prepared variant of the desired polynucleotide.
Oligonucleotide-mediated mutagenesis is a preferred method for preparing substitution, deletion and insertion variants of target polypeptide DNA. See Adelman et al., DNA 2:183 (1983).
Briefly, the target polypeptide DNA is altered by hybridizing an oligonucleotide encoding the desired mutation to a single-stranded DNA template. After hybridization, a DNA polymerase is used to synthesize an entire second complementary strand of the template that incorporates the oligonucleotide primer, and encodes the selected alteration in the target polypeptide DNA.
L61 Selection of Constant Region The variable segments of humanized antibodies produced as described supra are typically linked to at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. Human constant region DNA
sequences can be isolated in accordance with well-known procedures from a variety of human cells, but preferably immortalized B-cells (see Kabat et al., supra, and W087/02671).
Ordinarily, the antibody will contain both light chain and neavy chain constant regions. The heavy chain constant region usually includes CHi, hinge, CH2, CH3, and CH4 regions.
The humanized antibodies include antibodies having all types of constant regions, including IgM, IgG, IgD, IgA
and IgE, and any isotype, including IgGl, IgG2, IgG3 and IgG4.

When it is desired that the humanized antibody exhibit cytotoxic activity, the constant domain is usually a complement-fixing constant domain and the class is typically IgGl. When such cytotoxic activity is not desirable, the constant domain may be of the IgG4 class. The humanized antibody may comprise sequences from more than one class or isotype.

(7) Exnression Systems Nucleic acids encoding humanized light and heavy chain variable regions, optionally linked to constant regions, are inserted into expression vectors. The light and heavy chains can be cloned in the same or different expression vectors. The DNA segments encoding immunoglobulin chains are operably linked to control sequences in the expression vector(s) that ensure the expression of immunoglobulin polypeptides. Such control sequences include a signal sequence, a promoter, an enhancer, and a transcription termination sequence. Expression vectors are typically replicable in the host organisms either as episomes or as an integral part of the host chromosomal DNA. Commonly, expression vectors will contain selection markers, e.g., tetracycline or neomycin, to permit detection of those cells transformed with the desired DNA sequences (see, e.g., U.S.
Patent 4,704,362).
E. coli is one prokaryotic host useful particularly for cloning the polynucleotides of the present invention.
Other microbial hosts suitable for use include bacilli, such as Bacillus subtilus, and other enterobacteriaceae, such as Salmonella, Serratia, and various Pseudomonas species. In these prokaryotic hosts, one can also make expression vectors, which will typically contain expression control sequences compatible with the host cell (e.g., an origin of replication). In addition, any number of a variety of well-known promoters will be present, such as the lactose promoter system, a tryptophan (trp) promoter system, a beta-lactamase promoter system, or a promoter system from phage lambda. The promoters will typically control expression, optionally with SUBSTIME SHEET (RULE 26) an operator sequence, and have ribosome binding site sequences and the like, for initiating and completing transcription and translation.
Other microbes, such as yeast, may also be used for expression. Saccharomyces is a preferred host, with suitable vectors having expression control sequences, such as promoters, including 3-phosphoglycerate kinase or other glycolytic enzymes, and an origin of replication, termination sequences and the like as desired.
In addition to microorganisms, mammalian tissue cell culture may also be used to express and produce the polypeptides of the present invention (see Winnacker, From Genes to Clones (VCH Publishers, N.Y., N.Y., 1987).
Eukaryotic cells are actually preferred, because a number of suitable host cell lines capable of secreting intact immunoglobulins have been developed. Preferred suitable host cells for expressing nucleic acids encoding the immunoglobulins of the invention include: monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293) (Graham et al., J. Gen. Virol.
36:59 (1977)); baby hamster kidney cells (BHK, ATCC CCL 10);
Chinese hamster ovary-cells-DHFR (CHO, Urlaub and Chasin, Proc. Natl. Acad. Sci. (USA) 77:4216 (1980)); mouse sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251 (1980)); monkey kidney cells (CVi ATCC CCL 70); african green monkey kidney cells'(VERO-76, ATCC CRL 1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34);
buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, HB
8065); mouse mammary tumor (MMT 060562, ATCC CCL51); and, TRI
cells (Mather, et al., Annals N.Y. Acad. Sci. 383:44-46 (1982)); baculovirus cells.
The vectors containing the polynucleotide sequences of interest (e.g., the heavy and light chain encoding sequences and expression control sequences) can be transferred into the host cell by well-known methods, which vary depending on the type of cellular host. For example, calcium chloride transfection is commonlv utilized for Drokarvotic cells.

SUBSTITUTE SHEET (RULE 26) whereas calcium phosphate treatment or electroporation may be used for other cellular hosts. (See generally Sambrook et al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Press, 2nd ed., 1989).
When heavy and light chains are cloned on separate expression vectors, the vectors are co-transfected to obtain expression and assembly of intact immunoglobulins. After introduction of recombinant DNA, cell lines expressing immunoglobulin products are cell selected.
Cell lines capable of stable expression are preferred (i.e., undiminished levels of expression after fifty passages of the cell line).
Once expressed, the whole antibodies, their dimers, individual light and heavy chains, or other immunoglobulin forms of the present invention can be purified according to standard procedures of the art, including ammonium sulfate precipitation, affinity columns, column chromatography, gel electrophoresis and the like (see generally Scopes, Protein Purification (Springer-Verlag, N.Y., 1982). Substantially pure immunoglobulins of at least about 90 to 95% homogeneity are preferred, and 98 to 99% or more homogeneity most preferred, for pharmaceutical uses.
The recombinant techniques described above can also be used for expression of native sequences encoding human or murine antibodies. This approach is particularly advantageous for expression of human antibodies that are isolated as unstable cell lines.

III. Antibodies Fraatnents In another embodiment of the invention, fragments of the intact antibodies described above are provided.
Typically, these fragments compete with the intact antibody from Which they were derived for specific binding to P-selectin, and bind with an affinity of at least 107, 108, 109 M-1, or 1014 M"'1. Antibody fragments include separate heavy chains, light chains Fab, Fab' F(ab')Z, Fabc, and Fv.
Fragments can be produced by enzymic or chemical separation of intact immunoglobulins. For example, a F(ab'), fraQment can SUBSTIME SHEET (RULE 26) ~16 4 9 be obtained from an IgG molecule by proteolytic digestion with pepsin at pH 3.0-3.5 using standard methods such as those described in Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Pubs., N.Y. (1988). Fab fragments may be 5 obtained from F(ab')2 fragments by limited reduction, or from whole antibody by -digestion with papain in the presence of reducing agents. (See id.) Fragments can also be produced by recombinant DNA techniques. Segments of nucleic acids encoding selected fragments are produced by digestion of full-10 length coding sequences with restriction enzymes, or by de novo synthesis. Often fragments are expressed in the form of phage-coat fusion proteins. This manner of expression is advantageous for affinity-sharpening of antibodies as discussed in Section IV.
IV. Affinity-Sharpened Antibodies Many of the immunoglobulins described above can undergo non-critical amino-acid substitutions, additions or deletions in both the variable and constant regions without loss of binding specificity or effector functions, or intolerable reduction of binding affinity (i.e., below about 107 M-1). Usually, immunoglobulins incorporating such alterations exhibit substantial sequence identity to a reference immunoglobulin from which they were derived.
Occasionally, a mutated immunoglobulin can be selected having the same specificity and increased affinity compared with a reference immunoglobulin from which it was derived. Phage-display technology offers powerful techniques for selecting such immunoglobulins. See, e.g., Dower et al., WO 91/17271 McCafferty et al., WO 92/01047; and Huse, WO 92/06204.

V. Hybrid Antibodies The invention also provides hybrid antibodies that share the specificity of blocking antibodies against P-selectin but are also capable of specific binding to a second moiety. In hybrid antibodies, one heavy and light chain pair is usually from an anti-P-selectin antibody and the other pair from an antibody raised against another epitope. This results SUBSTITUTE , SHEET (RULE 26) in the property of multi-functional valency, i.e., ability to bind at least two different epitopes simultaneously, where at least one epitope is the epitope to which the blocking P-selectin antibody binds. Such hybrids can be formed by fusion of hybridomas producing the respective component antibodies, or by recombinant techniques.
Immunoglobulins can also be fused to functional regions from other genes (e.g., enzymes) to produce fusion proteins (e.g., immunotoxins) having novel properties.
VI, Nucleic Acids The intact antibodies and antibody fragments described above are often produced by expression of nucleic acids. All nucleic acids encoding any antibody or antibody described in this application are expressly included in the invention. Modifications of nucleic acids are readily accomplished by a variety of well-known techniques, such as site-directed mutagenesis (see Giliman and Smith, Gene, 8:81-97 (1979) and Roberts, et al., Nature, 328:731-734 (1987)).
Many of the nucleic acids of the invention show substantial sequence identity to nucleic acids encoding the heavy and light chains of mu MAb PB1.3 or the exemplified humanized derivatives thereof.

VII. Comvuters In another aspect of the invention, computers programmed to display three dimensional images of antibodies on a monitor or printer are provided. For example, a Silicon Graphics IRIS 4D workstation running under the UNIX operating system and using the molecular modelling package QUANTA
(Polygen Corp. USA) is suitable. Computers are useful for visualizing models of variants of humanized antibodies. In general, the antibodies of the invention already provide satisfactory binding affinity. However, it is likely that antibodies with even stronger binding affinity could be identified by further variation of certain amino acid residues. The three dimensional image will also identify many noncritical amino acids, which could be the subject of SUBSTITUTE SHEET (RULE 26) ~16~149 conservative substitutions without appreciable affecting the binding affinity of the antibody.

VIII. Large-Scale Production of Antibodies For large-scale production of antibodies, stable cells are produced containing multiple copies of the genes encoding heavy and light chain immunoglobulins in their chromosomes. Multiple copies are obtained by amplification of isolated region(s) of a cell's chromosomal DNA. Amplification is achieved using a selection agent, e.g., methotrexate (MTX), that inactivates a cellular enzyme (e.g., DHFR) that is essential under certain growth conditions. Amplification, i.e., the accumulation of multiple copies of the DHFR gene, results in greater amounts of DHFR being produced in response to greater amounts of MTX. Amplification pressure is applied notwithstanding the presence of endogenous DHFR, by adding ever greater amounts of MTX to the media. Amplification of desired genes (here genes encoding immunoglobulin heavy and light chains) is achieved by linking each gene to a copy of a DHFR gene on the same or separate plasmid(s), and cotransfecting the plasmid(s). Amplification of the DHFR
genes in response to MTX results in a concomitant increase in copy number of the desired immunoglobulin genes. Increased copy number of the desired genes results in greater expression of the desired heterologous protein. Stable cell lines expressing about 10-100, 20-50 or about 30 g humanized immunoglobulin per 106 cells per day are preferred.

IX. P-Selectin Fragments In another aspect of the invention, fragments of P-selectin are provided. The fragments comprise amino acids from the fifth C3b-C4b regulatory domain of P-selectin (fifth CRP). The fragments contain the epitope bound by mu MAb PB
1.3 and/or proximal epitopes bound by competing antibodies.
The fragments usually contain up to 5, 10, 20, 25, 50, 75, 100 or 200 amino acids in total. Most fragments contain some or all of the amino acids between positions 448 and 467 of the human P-selectin sequence shown in Fig. 34. These amino acids SUBS'TITUTE SHEET (RULE 26) define the epitope specifically bound by mu MAb PB1.3. In many fragments, at least five contiguous amino are from between positions 448 and 467 of the amino acid sequence shown in Fig. 34. Other fragments contain the entire contiguous segment of amino acids between positions 448 and 467. Some fragments consist essentially of this contiguous segment of amino acids. In these fragments, no other amino acids are present that contribute to the binding specificity or affinity of the fragment. Often, no other amino acids of any description are present.
The polypeptide fragments of the invention have a variety of uses. The fragments are useful as immunogens for generating antibodies that compete with mu MAb PB 1.3 for specific binding to P-selectin. The use of defined fragments as immunogens reduces the extent of screening required to isolate an antibody of desired specificity. The polypeptide fragments are also useful in the diagnostic and therapeutic methods described infra. Some fragments compete with full-length P-selectin molecules for binding to activated neutrophils. Thus, these fragments are useful for aborting diseases and conditions of the immune system mediate by interactions of P-selectin with its ligand(s) on activated neutrophils. Some fragments are also useful for monitoring the presence of activated neutrophils by binding to ligands to P-selectin present on the surface of these cells. The small size of the fragments renders them particularly suitable for effective delivery following in vivo administration. The exclusively human origin of the fragments reduces the risk of side-effects that might occur with some other agents.
X. Therapeutic and Diaanostic Methods The therapeutic methods employ the antibodies (whole and binding fragments), and fragments of P-selectin discussed above as therapeutic agents for treatment of various diseases.
The therapeutic agents are useful for prophylactic and therapeutic treatment of a wide variety of diseases and disorders of the immune system. Such diseases and disorders include transplant rejection, graft versus host disease, autoimmune diseases such as insulin-dependent diabetes SUBSTITUTE SHEET (RULE 26)"

~16Z149 mellitus, multiple sclerosis, stiff man syndrome, rheumatoid arthritis, myasthenia gravis and lupus erythematosus, and inflammatory disorders. The agents are also useful for preventing tumor metastasis inhibiting the adhesion of circulating cancer cells, such as carcinomas of the colon and melanoma. Some therapeutic agents function by blocking or otherwise antagonizing the action of a P-selectin molecule with its ligand. Other therapeutic agents function by killing cells bearing a polypeptide against which the agent is targeted.
The therapeutic agents are particularly suitable for treatment of inflammatory and thrombotic conditions including post-ischemic leukocyte-mediated tissue damage (reperfusion injury) arising from traumatic shock, stroke, myocardial infarction, acute transplantation rejection, frost-bite injury, compartment syndrome, and pathophysiologic conditions associated with cardio-pulmonary bypass, acute leukocyte-mediated lung injury (e.g., adult respiratory distress syndrome), septic shock, wound associated sepsis secondary to viral infection by e.g., herpes simplex virus, IgE-mediated allergic reactions such as acute phase asthmatic disease, and chronic inflammatory conditions, including rheumatoid arthritis, atopic dermatitis and psoriasis.
Ischemia/reperfusion injury is an inflammatory condition that occurs on restoring blood flow to organs suffering from an obstructed supply causing ischemia (oxygen deprivation). Unless rapidly relieved by reperfusion, ischemia causes death of surrounding cells, and eventually, death of a whole organ or patient. However, accumulating evidence suggests that reperfusion may itself exert deleterious effects on surrounding tissue. The deleterious effects of reperfusion are believed to result at least in part from an inflammatory response mediated by activated neutrophils in the restored blood flow. Some patients have whole-body ischemia, whereas in other patients ischemia is confined to particular parts or organs of the body. For example, a patient may suffer from epidermal, myocardial, renal, cerebral, splenic, hepatic, spinal, splanchnic, SUBSTITUTE. SHEET (RULE 26) pulmonary, partial-body, or whole-body ischemia. The therapeutic agents of the invention function by antagonizing the interaction of such lymphocytes with P-selectin.
One major component of acute allergic conditions, including asthma, is the degranulation of mast cells following challenge of subjects with antigens to which they are specifically sensitized. The consequences of mast cell degranulation include a bronchoconstrictor response and also an inflammatory response characterized in part by leukocyte accumulation. Histamine, which is contained together with other inflammatory mediators in mast cells, can induce the expression of P-selectin on vascular endothelial cells, indicating that degranulation of mast cells may also result in P-selectin expression and subsequent leukocyte accumulation.
Because mast cell degranulation is a key element in the pathogenesis of allergic conditions, administration of antibodies to P-selectin will be useful for the treatment of human allergic conditions.
The methods are particularly useful for humans, but may also be practiced on veterinary subjects. The therapeutic methods are usually applied to organs present in living subjects. However, some methods, such as ischemia-reperfusion therapy, are equally applicable to dissected organs, particularly when such organs are awaiting transplant to a recipient patient. Therapeutic methods can also be performed ex vivo.
Therapeutic agents and compositions targeted against P-selectin can also be used in combination with agents targeted against other molecules, particularly humanized or human antibodies reactive with different adhesion molecules.
Suitable immunoglobulins include those specific for CD11a, CD11b, CD18, E-selectin, L-selectin and ICAM-1. The immunoglobulins should bind to epitopes of these adhesion molecules so as to inhibit binding of leukocytes, particularly neutrophils, to endothelial cells. Other suitable antibodies for use in combination therapies are those specific for lymphokines, such as IL-1, IL-2 and IFN-7, and their receptors. Such antibodies serve to block activation of SUgS'(ITUTE SHEET (RULE 26) ~-62 149 endothelial cells, and thereby prevent their interaction with neutrophils in an inflammatory response.
The blocking P-selectin antibodies and pharma-ceutical compositions of this invention are particularly useful for parenteral administration, i.e., subcutaneously, intramuscularly or intravenously. A number of new drug delivery approaches are being developed, the pharmaceutical compositions of the present invention are suitable for administration using these new methods, as well. See Langer, Science, 249:1527-1533 (1990).
The blocking P-selectin antibodies can be directly or indirectly coupled to the chemotherapeutic agent. The coupling, which may be performed by means, generally known in the art, should not substantially inhibit the ability of the immunoglobulin to bind the receptor nor should it substantially reduce the activity of the chemotherapeutic agent. A variety of chemotherapeutics can be coupled for targeting. For example, anti-inflammatory agents which may be coupled include immunomodulators, platelet activating factor (PAF) antagonists, cyclooxygenase inhibitors, lipoxygenase inhibitors, and leukotriene antagonists. Some preferred moieties include cyclosporin A, indomethacin, naproxen, FK-506, mycophenolic acid, and the like. Similarly, anti-oxidants, e.g., superoxide dismutase, are useful in treating reperfusion injury. Likewise, anticancer agents, such as daunomycin, doxorubicin, vinblastine, bleomycin, and the like can be targeted.
The P-selectin targeting may also be accomplished via amphipaths, or dual character molecules (polar:nonpolar) which exist as aggregates in aqueous solution. Amphipaths include nonpolar lipids, polar lipids, mono- and diglycerides, sulfatides, lysolecithin, phospholipids, saponin, bile acids and salts. These molecules can exist as emulsions and foams, micelles, insoluble monolayers, liquid crystals, phospholipid dispersions and lamellar layers. These are generically referred to herein as liposomes. In these preparations, the drug to be delivered is incorporated as part of a liposome in which an anti-P-selectin immunoglobulin is embedded. In this SU6STITUTE. SHEET (RULE 26) -~..

embodiment, the immunoglobulin need not bind a functional epitope on the P-selectin molecule, so long as the immunoglobulin effectively targets the liposome to P-selectin molecules. When the liposomes are brought into proximity of the affected cells, they deliver the selected therapeutic compositions.
A variety of methods are available for preparing liposomes, as described in, e.g., Szoka et al., Ann. Rev.
Biophys. Bioeng. 9:467 (1980), U.S. Pat. Nos. 4,235,871, 4,501,728 and 4,837,028, incorporated herein by reference.
Targeting of liposomes using a variety of targeting agents (e.g., ligands, receptors and monoclonal antibodies) is well known in the art. (see, e.g., U.S. Patent Nos. 4,957,773 and 4 , 603 , 044) .
Standard methods for coupling targeting agents to liposomes can be used. Antibody targeted liposomes can be constructed using, for instance, liposomes which incorporate protein A (see Renneisen, et al., J. Biol. Chem., 265:16337-16342 (1990) and Leonetti et al., Proc. Natl. Acad. Sci. (USA) 87:2448-2451 (1990).
Pharmaceutical compositions for parenteral administration usually comprise a solution of a therapeutic agent (e.g., an antibody against P-selectin (intact or binding fragment or a P-selectin fragment) or a cocktail of several such agents dissolved in an acceptable carrier, preferably an aqueous carrier. A variety of aqueous carriers can be used, e.g., water, buffered water, 0.4% saline, 0.3% glycine and the like. These solutions are sterile and generally free of particulate matter. These compositions may be sterilized by conventional, well known sterilization techniques. The compositions may contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions such as pH-adjusting and buffering agents, tonicity adjusting agents and the like, for example sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate, etc. The concentration of antibody in these formulations can vary widely, i.e., from less than about 0.5%, usually at or at least about 0.1% to as much as 1.5% or 2.0%
_. ....._._..__ .., ,~ _.....___.
_.___ _..__.............~.-------=......_.

'C 9412.S06% PCT/US94704935 by weight and will be selected primarily based on fluid vol=es, viscosities, etc., in accordance with the particular mode of administration selected. Actual methods for preparing parenterally administrable compositions will be known or apparent to those skilled in the art and are described in more detail in, for example, Remington's Pharmaceutical Sciences, 17th Ed., Mack Publishing Company, Easton, Pennsylvania (1985).
The antibodies of this invention can be lyophilized for storage and reconstituted in a suitable carrier prior to use. This technique has been shown to be effective with conventional immunoglobulins and art-known lyophilization and reconstitution techniques can be employed. Lyophilization and reconstitution can lead to varying degrees of antibody ac-tivity loss (e.g., with conventional immunoglobulins, IqM
antibodies tend to have greater activity loss than IgG
antibodies) and that use levels may have to be adjusted to compensate.
The compositions containing the present antibodies or a cocktail thereof can be administered for the prophylactic and/or therapeutic treatments. In therapeutic application, compositions are administered to a patient in an amount sufficient to cure or at least partially arrest the infection and its complications. An amount adequate to accomplish this is defined as a "therapeutically effective dose." Amounts effective for this use will depend upon the severity of the disease and the general state of the patient's own immune system, but generally range from about .05 mg/kg body weight to about 5 mg/kg body weight, preferably between about .2 mg/kg body weight to about 1.5 mg/kg body weight.
The materials of this invention may generally be employed in serious disease states, that is, life-threatening or potentially life-threatening situations. In such cases, in view of the minimization of extraneous substances and the lower probability of "foreign substance" rejections (e.g., HAMA) which are achieved by human or humanized antibody forms of this invention, it is possible and may be felt desirable by ~ ~~NTE SNEET (RULE 26) the treating physician to administer substantial excesses of these antibodies.
In prophylactic applications, compositions containing the present antibodies or a cocktail thereof are administered to a patient not already in a disease state to erihance the patient's resistance. Such an amount is defined to be a "prophylactically effective dose." In this use, the precise amounts again depend upon the patient's state of health and general level of immunity, but is generally in the ranges described above.
Single or multiple administrations of the compositions can be carried out with dose levels and pattern being selected by the treating physician. In any event, the pharmaceutical formulations should provide a quantity of the immunoglobulins of this invention sufficient to treat the patient effectively.
The antibodies of invention (whole and binding fragments) and fragments of P-selectin can also be used for diagnostic purposes. An amount sufficient for these purposes is defined to be a "diagnostically effective dose." In diagnostic uses, the precise amounts will depend upon the patient's state of health, mode of administration, and the like. The antibodies may either be labeled or unlabeled.
Unlabeled antibodies can be used in combination with other labeled antibodies (second antibodies) that are reactive with the antibody, such as antibodies specific for the particular immunoglobulin constant region. Alternatively, the antibodies can be directly labeled. A wide variety of labels may be employed, such as radionuclides, fluorescers, enzymes, enzyme substrates, enzyme cofactors, enzyme inhibitors, ligands (particularly haptens).
The antibodies (whole and binding fragments) are useful for detecting the presence of cells bearing the P-selectin receptor. The presence of such cells is diagnostic of an inflammatory condition or disease and may signal the need for commencement of a therapeutic method discussed supra.
Diagnosis can be accomplished by removing a cellular sample from a patient. The amount of expressed P-selectin receptor SUBSTITUTE SHEET (RULE 26) 40 ~C,L1't9 in individual cells of the sample is then determined, e.g., by immunohistochemical staining of fixed cells or by Western blotting of a cell extract with an antibody of the invention.
Fragments of P-selectin are useful for monitoring the presence of activated neutrophils, a presence that can be indicate of an'undesirable immune response.
Diagnosis can also be achieved by in vivo administration of a labelled antibody (preferably a humanized or human antibody) and detection by in' vivo imaging. The concentration of MAb administered should be sufficient that the binding to cells having the target antigen is detectable compared to the background signal. The diagnostic reagent can be labelled with a radioisotope for camera imaging, or a paramagnetic isotope for magnetic resonance or electron spin resonance imaging.
A change (typically an increase) in the level of P-selectin protein in a cellular sample or imaged from an individual, which is outside the range of clinically established normal levels, may indicate the presence of an undesirable inflammatory response reaction in the individual from whom the sample was obtained, and/or indicate a predisposition of the individual for developing (or progressing through) such a reaction. P-selectin can also be employed as a differentiation marker to identify and type cells of certain lineages and developmental origins. Such cell-type specific detection can be used for histopathological diagnosis of undesired immune responses.
Kits can also be supplied for use with the subject antibodies. Thus, the subject antibody composition of the present invention may be provided, usually in a lyophilized form in a container, either alone or in conjunction with additional antibodies specific for the desired cell type. The antibodies, which may be conjugated to a label or toxin, or unconjugated, are included in the kits with buffers, such as Tris, phosphate, carbonate, etc., stabilizers, biocides, inert proteins, e.g., serum albumin, or the like, and a set of instructions for use. Generally, these materials will be present in less than about 5% wt. based on the amount of SUBSTITUTE SHEET (RULE 26) ~SU 9-4,v067 PCTlUS94l04935 active antibody, and usually present in total amount of at least about 0.001% wt. based again on the antibody concentration. Frequently, it will be desirable to include an inert extender or excipient to dilute the active ingredients, where the excipient may be present in from about 1 to 99% wt.
of the total composition. Where a second antibody capable of binding to an anti-P-selectin antibody is employed in an assay, the second antibody will usually be present in a separate vial. The second antibody is typically conjugated to a label and formulated as described above.
The following examples are offered by way of illustration, not by limitation.

Examole 1 Prenaration of Reagents This example describes preparation or isolation of reagents used for the immunizations in Examples 2 and 3 and the ELISAs in Example 4.

A. "OUTDATED PLATELETS"
A preparation of platelets which was purified from "outdated platelets" obtained from the San Diego Blood Bank.
The method used was a modification of the method described in Moore et al., J. Cell. Bfol. 112:491-499 (1991), Briefly, 25 units of outdated platelet-rich plasma were centrifuged twice for 10 min at 1200rpm (300xg) to remove contaminating non-platelet blood cells. The purified platelet preparation was then washed three times with a buffer containing O.1M NaCl, 20mM
Tris and 5mM Benzamidine. It also contained 3.8% sodium citrate. The pH was adjusted to 7.5 with iN HC1. The purified platelets were stored at 70 C and were washed once in PBS before injection.

B. PURIFIED P-SELECTIN
Fractionation of Platelets: Washed outdated platelets were fractionated essentially according to Moore et al., 1991, supra, 25 units washed platelets were made 100 uM

~5 V Y+r~~uu. _ - _ _ _ -Nfs in i-eupeptin and 1 mM in 4-(2aminoethyl)-benzenesulfonyl-fl,.:oride (AEBSF), freeze-thawed three times in dry ice/methanol, and homogenized ten strokes with a Dounce hoLogenizer. The suspension was then centrifuged in a Beckman Ti 70 rotor at 4 C at 35,000 rpm for 60 min. The supernatant from this spin was the starting material for the purification of "soluble" P-selectin. The pellet was resuspended in 30 ml 1 nM MnC121 1 mM CaC12, 5 mM benzamidine-HC1, 0,1M NaCl, 20 mM
Tris, 2% Triton X-100, pH 7.5, homogenized 10 strokes with a Dounce homogenizer, and incubated for 1 h at 4 C. The suspension was then centrifuged for 1 h in a Beckman Ti 70 rotor at 4 C at 35,000 rpm. The supernatant from this spin was the starting material for the isolation of "membrane-bound" P-selectin.
Isolation of P-Selectin from Fractionated Platelets:
"Soluble" and "membrane-bound" P-selectin were then further purified separately by some or all of the following procedures. The non-ionic detergent Rennex 30 (Accurate Chemical and Scientific Corp.) was included at a concentration of 0.1% in all buffers for isolation of the membrane-bound form of P-selectin. All buffers also included 1 mM CaCl2 and 1 mM MnC12. Fractions containing P-selectin were detected by Western blotting with the monoclonal antibody PNB1.6.
Lentil Lectin Chromatoaraphv: "Soluble P-selectin"
supernatant or "membrane-bound P-selectin" detergent extract from 25 units of platelets were passed over a column (1.5 x 5.5 cm) of lectin from Lens culinaris (lentil lectin) coupled to Sepharose 4B (Sigma Chemical Co., L-0511). The column was then washed with 150 ml 20 mM Tris, 0.1M NaCl, pH 7.5. Bound glycoproteins were then eluted from the column with 10 x 3 ml aliquots of 20 mM Tris, 0.5M NaCl, 0.5M alpha-methyl D-mannopyranoside, pH 7.5: Fractions containing P-selectin were concentrated to 1.5-2 ml in an Amicon Centriprep 30.
Gel Filtration Chromatoaranhv: Concentrated fractions from lentil lectin chromatography were injected onto a Toso Haas G3000 SW HPLC column (21.5 mm x 30 cm) equilibrated with 20 mM Tris, 0.1M NaCl, pH 7.5. The column was eluted with the same buffer at a flow rate of 0.75 ml/min.

* Trademark SJBSiiiUi SHEET (RULE 26) %VO 94;2:-067 PCT/US94/04935 fractions (1.5 ml) were collected, and analyzed by Western blotting. Fractions containing P-selectin were pooled and concentrated to 1.5-2 ml in an Amicon Centriprep 30, followed by several buffer exchanges with 20 mM Tris, pH 7.5.
Hevarin-Aaarose Chromatoaraphy: The sample, in 20 ~
mM Tris, pH 7.5, was applied to a 1 ml column of Heparin-Agarose, the column was washed with the same buffer, and eluted with 20 mM Tris, 1.5M NaCl, pH 7.5.
Anion Exchanae Chromatographv: Fractions were dialyzed into 20 mM Tris, pH 7.5, and injected onto a Toso Haas DEAE-5PW column (7.5 mm x 7.5 cm) that had been equilibrated with 20 mM Tris, pH 7.5, and eluted with a salt gradient to 1M NaCl. Fractions containing P-selectin were pooled, and concentrated in an Amicon Centriprep-30, and stored at -60 C.
Cation Exchange Chromatoaraphv: Samples were dialyzed into 10 mM phosphate, pH 7, and injected onto a Poros*
CM/P (4.6/100) column that had been equilibrated in the same buffer, and eluted with a salt gradient to iM NaCi.
C. ISOLATION AND ACTIVATION OF FRESH HUMAN PLATELETS
All chemicals used throughout the platelet isolation procedure were obtained from Sigma Chemical Company, St.
Louis, MO.
1. 42 ml of blood was drawn from a human volunteer blood donor into a syringe containing 7 ml Acid Citrate Dextrose anticoagulant (ACD).
Pregaration of ACD anticoaaulant Dextrose 2.Og Sodium Citrate 2.49g Citric Acid 1.25g Bring to 100 ml with distilled water 2. The needle was removed and the blood was transferred to two sterile 50 ml tubes.
3. The tubes were centrifuged in an IECC-6000 centrifuge equipped with a 921 head (radius=17.2 cm.) at 800 rpm (approx. 90xg) for 15 min at room temperature with the brake off.

* Trademark 2162i49 4. The supernatant was removed using a plastic pipette.
The supernatant was removed as close as possible to the buffy layer.
5. The supernatant was centrifuged at 1200 rpm (approx.
300xg) for 6 min.
6. The supernatant was removed.
7. The supernatant was centrifuged at 2000 rpm (approx.
1200xg) for 10 min. The plateJgts appeared as a cell button at the bottom of the tube. The supernatant was discarded.
8. The platelets were washed as follows: 2 ml Tyrode-Hepes Buffer pH 6.5, containing Prostaglandin, E1(PGE1 at a final concentration of l00nM was added and the platelets were gently resuspended. A
further 10 ml of the same buffer was added and the sample was centrifuged at 3000 rpm for 10 min.
Preparation of Tvrode-Henes Buffer NaCl 8.Og KC1 0.2g NaH2PO4HZO 0.057g MgC126H2O 0.184g NaHCO3 0.1g Dextrose 1.Og HEPES 2.383g Bring to 1 liter with distilled water. Adjust pH to 6.5 with iN NaOH.
9. Step 18 was repeated once more and then the platelets were washed once in the same buffer without PGE1.
10. The platelets were counted in a Coulter Counter and diluted to 2x108/ml.

11. The pH of the platelet suspension was adjusted to 7.2, and the amount of thrombin (Human thrombin-Sigma) required for maximal activation was determined. There is a certain amount of donor variation, but optimal activation is normally obtained in the range of 0.25 - 0.5 Units/ml.
Maximal activation i.e., maximal thrombin-induced SUBSTITUTE SHEET (RULE 26) aggregation appears to correspond to maximal expression of P-selectin.

12. The required amount of thrombin was added to the platelet preparation and the mixture was allowed to stand at room temperature for 20 min without stirring to prevent aggregation.
Example 2 p*"pDaration of Blocking P-selectin Antibody This example describes preparation of mu mAB PB1.3, a monoclonal antibody to P-selectin that inhibits binding of thrombin-activated platelets to neutrophils.
One RBF/DnJ male mouse, received from Jackson Laboratories, was used as a source of antibody producing cells and was immunized according to the following schedule (all injections were performed intraperitoneally):
1. Month 1 - 200 l of packed "outdated platelets"
containing approximately 6x109 platelets 2. Month 2 - 250 l of packed "outdated platelets"
(8x109 platelets) 3. Month 4 - 250 l of packed "outdated platelets"
(8x109 platelets) 4. Month 5-"Soluble" fraction of P-selectin, isolated on lentil lectin.
5. Month 6-"Soluble" P-selectin, purified by lentil lectin, gel filtration, heparin-agarose, and ion-exchange chromatography.
Details of the preparation of all agents used for immunization are presented in Example 1.
A. MYELOMA FUSION CELL LINE
FOX-NY, a mouse myeloma cell line deficient for adeNosine phosphoribosyltransferase (APRT) and hypoxanthine phosphoribosyltransferase (HPRT) was obtained from the American Type Culture Collection (CRL 1732) and maintained in RPMI 1640 containing 10% fetal bovine serum (Hyclone) and 1%
L-glutamine.

SUBSTITUTE SHEET (RULE 26) B. CELL FUSION PROCEDURE
Four days after the final boost the spleen was removed and 1.2x108 splenocytes were recovered. These were fused with FOX-NY myeloma cells using PEG 1500(BMB) using the following protocol: The isolated splenocytes were washed twice in serum-free cell culture medium. Splenocytes and myeloma cells were combined at a ratio of 1:4.8 (myelomas to spleen cells). The combined cell pellet waFwashed twice in serum free medium, then aspirated to drynessA. The cell pellet was resuspended by gentle tapping and he3ted in a waterbath at 37 C for 1 min. The pellet was distributed around the sides and bottom of a 50 ml conical centrifuge tube. One ml of PEG
1500 (50% w/v in 75mM HEPES, BMB Lot #14702800) previously warmed to 37 C was added over 60 seconds while rotating the tube to maintain a thin layer of cells. One ml of serum free medium was added slowly over 60 seconds. An additional lml of medium was added slightly faster. A further 8 ml of medium was added and the tube was allowed to stand undisturbed for 8 min and was then centrifuged for 5 min at 300xg. The final pellet was resuspended in RPMI 1640 containing 10% fetal bovine serum Hyclone), 1% L-glutamine, 1% Sodium Pyruvate, and 1xAAT (Sigma). The cells were plated in 10 flat bottom 96 well microtiter plates (COSTAR). No feeder cells were used.

Example 3 Preparation of Non-blocking Anti-P-selectin antibody This example describes preparation of mAB PNB1.6 a monoclonal antibody to P-selectin that does not inhibit binding of thrombin-activated platelets to neutrophils.
One RBF/DnJ male mouse, received from Jackson Laboratories, was used as a source of antibody producing cells and was immunized according to the following schedule (all injections were performed intraperitoneally):
1. Month 1 - 100 l of thrombin-activated freshly isolated platelets (approximately 3x109 platelets) 2. Month 2 - 200 l of thrombin-activated fresh platelets (6x109 platelets) SUBSTITUTE SHEET (RULE 26) V Jti ~~vu= L. = i VJ/'.i v'./.~ u 3. Month 4 - 100 1 of thrombin-activated fresh platelets (3x109 platelets) Four days post final boost, the spleen was removed and the splenocytes were recovered. They were fused with FOX-NY myeloma cells using PEG 1500(Sigma) generally as described in Oi et al., "Immunoglobulin-Producing Hybrid Cell Lines" in Selected Methods in Cellular Immunology, eds. Mishell and Shiigi, pp 351-372, 1980.

Example 4 Screenina for Antibodv This example describes screening of supernatant media from fused cells produced in Examples 2 and 3.
Supernatants from Example 2 were tested by ELISA assay against (a) thrombin-activated platelets, (b) purified P-selectin and (c) recombinant P-selectin. Supernatants from Example 2 were tested by ELISA assay against thrombin-activated platelets.
Each of these assays is described below. Preparation of the reagents used in each of the assays is described in Example 1.
A. THI2OMBIN-ACTIVATED PLATELETS
1. A 96 well flat bottom COSTAR plate was coated with 0.1% gelatin (2% gelatin-Sigma) by adding 100 l/well and incubating at 37 C.for 15 min.
2. The gelatin was removed and 100 l of thrombin activated platelets (108/ml) were added to each well.
3. The plate was incubated at 37 C for 15 min.
4. The plate was centrifuged at 800 rpm (90xg) for 2 min.
5. The unbound platelets were removed and the plate was washed x2 with PBS.
6. 100 1 of PBS containing 1t BSA were added to each well and the plate was allowed to stand at room temperature for 60 min.

SUBSTfTIJTE SHEET (RULE 26) ~VU Y-a t_vo .--4 7. 10041 of supernatant were added to each well. The plate was placed on a shaker at room temperature for 60 min.
8. The plate was washed x4 with PBS.
9. 10041 of peroxidase-conjugated goat anti-mouse IgG
diluted 1/1000 in PBS containing 1% BSA was added to each well. The plate was allowed to stand at room temperature for 30 min.
10. The plate was washed x7 with PBS.
11. 10041 of ABTS substrate system (Kirkegaard and Perry, Laboratories, Inc.) was added to each well.
12. The plate was allowed to develop at room temperature for up to 30 min.

13. The OD was measured at 414nm in a Titertek Multiskan MCC/340.

B. PURIFIED P-selectin 1. "Soluble" or "membrane-bound" P-selectin, isolated by chromatography on lentil lectin, gel filtration, and DEAE columns, was diluted (1:50 to 1:1000) in DPBS (Dulbecco's Phosphate Buffered Saline is Phosphate-buffered saline (PBS) containing CaC12(1mM) and MgS0a (0.5mM)) and coated overnight at 40C onto Falcon 96 well microtiter plates.
2. The plates were then blocked for 1 h or longer with DPBS+I%BSA.
3. Supernatants to be assayed were then added to the wells, and incubated for 30-60 min at room temp.
4. The plates were washed with DPBS, and then sheep anti-mouse IgG horseradish peroxidase conjugate (Sigma A-6782, 1:1000 in DPBS+I%BSA) was added to the wells, and the plates were incubated at room temp. for 1 hr.
5. The plates were then washed with DPBS, and developed with the TMB Microwell~Substrate System (Kirkegaard and Perry Laboratories, Inc.) * Trademark SUBSTiTUTE SHEET (RULE 26) WO 94/25067 PCTRT~94/04935 C. RECOMBINANT P-selectin 1. Ninety six well flat bottom plates were coated overnight at 4 C with 100 1 of P-selectin which had been shed into the serum free medium of a pool of 293 cell clones transfected with the recombinant P-selectin gene. This supernatant contained at least 8ng/ml of P-selectin and was coated onto COSTAR
plates either undiluted or diluted 1:8 with media.
2. The plate was washed once with Dulbecco's PBS (DPBS) and blocked with 200 1 DPBS containing 1% BSA for 30 min at room temperature.
3. The plate was then processed for the ELISA Assay according to steps 7-13 in Section A, Page 9.

D. ISOTYPING ANALYSIS OF mABs to P-Selectin Isotype analysis of mu MAbs PB1.3 and PNB1.6 indicated that both monoclonal antibodies were of the mouse immunoglobulin subclass, IgGi. Isotyping was performed using the capture method of Boehringer Mannheim Mouse-Hybridoma-Subtyping Kit.

Exam 1~ 5 Detection of P-selectin This example describes the detection of P-selectin by western blotting. Samples containing P-selectin were mixed with an equal volume of SDS-PAGE sample buffer (non-reducing), heated at 100 C, run on a Novex 8-16% gradient gel, and transferred electrophoretically to nitrocellulose. The nitrocellulose membrane was then blocked for at least 1 h in phosphate-buffered saline (PBS) +1% bovine serum albumin (BSA). The membrane was then incubated at room temperature for 1 h or longer with either tissue culture supernatant from hybridomas expressing the appropriate monoclonal antibody or purified antibody (5 g/ml) in PBS+1%BSA. The membrane was then washed several times in PBS+1%BSA, and incubated for 1 h with a 1 to 1000 dilution of sheep anti-mouse IgG horseradish peroxidase conjugate (Sigma A-6782) in PBS+1%BSA. The membrane was then washed, and bands visualized with SUBSTITUTE SHEET (RULE 26) 50 ~1h2 14 9 tetramethylbenzidine/membrane enhancer (Kirkegaard and Perry Laboratories, Inc.) For tissue culture supernatants from both mu MAbs PB1.3 and PNB1.6 a protein was detected having a molecular weight of 14OKd. This was shown in preparations of solubilized platelet membranes as well as in purified P-selectin.

ExamDle 6 In Vivo Inhibition of Acute Inflammatory InZurv by Blocking P-selectin Antibody This example provides data demonstrating the ability of blocking P-selectin antibody of the invention to treat acute lung injury. In this example, systemic complement activation was produced by vascular infusion of the cobra venom factor (CVF) into rats. Injury in this model develops rapidly and is known to be dependent on toxic oxygen products generated from neutrophils.
For these studies, the two monoclonal antibodies described in Examples 2 and 3 were used. Of particular importance for this study, the blocking P-selectin antibody mu MAb PB1.3 also inhibited adhesion'of thrombin-activated rat platelets to human neutrophils (Fig. 1). These data suggest that mu MAb PB1.3 recognizes a conserved functional P-selectin epitope and that a blocking P-selectin antibody might be useful in the treatment of other inflammatory injuries.
For the in vivo experiments 20 units CVF/kg body weight were infused intravenously into 300 gm adult male Long-Evans rats. This results in rapid intrapulmonary, intravascular sequestration of blood neutrophils, with resulting damage of interstitial capillary endothelial cells at points of physical contact of endothelial cells and neutrophils. When employed, mu MAbs PB1.3 and PNB1.6.were infused intravenously in the amounts indicated together with CVF, in total volume of 0.5 ml. Negative control animals received 0.5 ml intravenous infusion of phosphate buffered saline (PBS). In all cases the infusion material also contained trace amounts of (1251) -bovine serum albumin and SUBSTiTUTE SHEET (RULE 26) homologous (51Cr)-red blood cells (RBC). The parameters of lung injury (leakage of albumin and extravasation of RBC) were determined at 30 min. according to established techniques Mulligan, et al., J. C1in Invest. 88:1396 (1991).
The results of these experiments are shown in Figs.
2A and 2B. Coinfusion of 200 g PNB1.6 (the non-blocking anti-P-selectin antibody) together with CVF failed to cause any reduction in lung injury when compared to the values of untreated CVF positive controls. Thus animals receiving CVF
PNB1.6 in PBS served as the reference positive control values.
When 100 g mu MAb PB1.3 antibody was infused with the CVF
permeability was reduced by 19.2% (p=.002) and hemorrhage was reduced by 37.5% (p=.001) (data now shown). When 200 g mu MAb PB1.3 was added, permeability was reduced by 50.8%
(p<.001) (Fig. 2A) and hemorrhage by 70.0% (p<.001) and 70.1%
(p<.001), respectively (Fig. 2B).
Lungs from companion sets of animals were homogenized sonicated and the myeloperoxidase activity (MPO) measured in order to obtain an estimate of the neutrophil content in lung. MPO was determined by the decomposition of H202 in the presence of o-dianisidine according to standard procedures, Warren et al. J. C1in. Invest. 84:1873 (1989). As shown in Fig. 3A, treatment with 100, 200 or 400 g mu MAb PB1.3 reduced MPO content below the values in the reference (untreated) positive control group by 26% (p=.024), 41%
(p<.001) and 50% (p<.001), respectively. In animals injected with 200 g of PNB1.6 there was no reduction in MPO content (Fig. 3A), consistent with the inability of this antibody to protect against CVF-induced lung injury. Thus, the protective effects of the blocking antibody mu MAb PB1.3, correlate with its ability to interfere with neutrophil accumulation in lung tissue. Transmission electron microscopic examination of lung sections confirmed that treatment with mu MAb PB1.3 resulted in reduced accumulation of neutrophils within the pulmonary interstitial capillaries, diminished adherence of neutrophils to endothelial cells, and reduced evidence of damaged endothelial cells (Fig. 3B).

SUBSTITUTE SHEET (RULE 26) 52 ~162 149 In order to investigate lung expression of P-selectin following intravenous injection of CVF, an additional group of rats was infused with CVF and animals sacrificed at times 0, 5, 10, 15, 20 and 60 min later. Lungs were inflated with O.C.T., snap frozen, and sections obtained and examined for presence of P-selectin by immunohistochemical techniques, using mu MAb PB1.3. Little detectable reactivity in the pulmonary vasculature was found at time 0, W,hereas staining was clearly evident at 5 min and was increAsed at 15 and 20 min after infusion of CVF. The pattern of staining involved pulmonary venules and septal areas in a pattern consistent with staining of interstitial capillaries. At 60 min.
staining had largely disappeared (data not shown). Although P-selectin is presumably present in intracellular storage granules prior to CVF infusion, it is apparent that mobilization of P-selectin to the surface of the endothelium dramatically enhanced its reactivity with mu MAb PB1.3 antibody.

Example 7 Competitive Binding Assay Methods of isolation of platelets from human blood, activation of isolated platelets with thrombin and isolation of PMNs (neutrophils) are described in Example 1 above.
INCUBATION OF THROMBIN ACTIVATED.PLATELETS WITH MONOCLONAL
ANTIBODIES TO P-SELECTIN.
a. Place 20 l thrombin-activated platelets (2x108/ml) into each of 24 Eppendorf tubes (1.5 ml) in duplicate.
b. To 8 of these tubes, add 20 l of the IgG fraction purified from PNB1.6 at a concentration of 10 g/ml in Tyrode-Hepes Buffer, pH 7.2 containing 1 mM
CaC12. To a second row of 8 tubes, add 20 l of the same IgG preparation at a concentration of 1 g/ml.
To the third row of 8 tubes add 20 l of the buffer alone.
c. Mix and stand at room temperature for 20 min.
SUBSTITUTE SHEET (RULE 26) V lti.~...v1l.J.l.JJ11 ~.~

d. To each row of 8 tubes add 20 l of the IgG fraction purified from mu MAb PB1.3 at concentrations ranging from 10 g/ml to 0.03 g/ml.
e. Mix and stand at room temperature for 20 min.
f. Add 20 l neutrophils at 3x106/ml.
g. Mix and stand at room temperature for 20 min.
h. Evaluate adhesion microscopically as follows: Count 100 neutrophils in each sample. Score a cell as positive if it has bound 2 or more platelets, and as negative if it has bound less than 2. Calculate the percentage of positive cells.

As shown in Fig. 1, pretreatment of thrombin activated platelets with mu MAb PNB1.6 did not affect the ability of mu MAb PB1.3 to block their P-Selectin mediated adhesion to neutrophils.

Example 8 Characterization of Bindinq of Blockina P-selectin Antibodies I(a) Purification of Monoclanal Antibodies: The monoclonal antibodies PNB1-6 and 84/26 were isolated from tissue culture supernatants by passage over a column of protein G Sepharose 4 Fast Flow (Pharmacia). The column was washed extensively with phosphate-buffered saline (PBS), and bound antibody was eluted with 0.1M glycine-HC1 pH
2.7 into tubes containing 0.2 to 0.5 volume 1M Tris, pH
8.8. The procedure used for isolating mu MAb PB1.3 was identical except that the antibody bound to protein G was eluted with 0.1M acetate-HC1 pH 2.5 into tubes containing 0.3 volumes 2M Tris, pH 10. Antibody containing fractions were pooled, and dialyzed against several changes of PBS at 40C.
(b) Preparation of mu MAb PB1.3 Affinity Column: For use in the purification of P-selectin, mu MAb PB1.3 was coupled to tresyl activated agarose according to the manufacturer's affinity instructions (Schleicher and Schuell).

* Trademark SUBSTTTUTE SHEET (RULE 26) 1NU 'JJILZUO i . . .. . ~.....~. _ (c) Purification of P-Selectin: to washed, outdated human platelets, prepared as described above, was added 1/100th volume 5 mg/mi leupeptin, and 1/100th volume 35 mg/ml AEBSF (Calbiochem). After mixing, the platelets were freeze-thawed three times in either a dry-ice/methanol bath or liquid nitrogen, homogenized ten strokes with a Dounce homogenizer, and centrifuged for 1 h at 35,000 rpm at 4 C in a Beckman 70TI rotor. The pellet was resuspended in Dulbeccols phosphate-buffered saline (DPBS, Whittaker Bioproducts, which contains approximately 0.9 uM CA++ and 0.5 mM Mg++), and made 5 mM
in benzamidine-HC1, 100 uM in leupeptin, and 2% in Triton*
X-100. The suspension was resuspended, and homogenized 10 strokes with a Dounce homogenizer. After a 1 h incubation at 4 C, the solution was again centrifuged in a Beckman 70TI rotor for 1 h at 4 C at 35,000 rpm.. The supernatant was passed over a column of mu MAb PB1.3-agarose that had been equilibrated with DPBS+0.05% Rennex 30 (Accurate Chemical and Scientific Corp.). The column was washed extensively with DPBS+0.05% Rennex 30, and bound P-selectin was eluted with 0.1M triethylamine-HCL, 0.05% Rennex 30, pH 11.5, into tubes containing 0.1 volume 1 M phosphate, pH 6.8. Fractions containing P-selectin were pooled, dialyzed against DPBS+0.05% Rennex 30, concentrated using either a Centricon 30 or Centriprep 30 spin concentrator (Amicon), aliquoted, and stored at -80 C.
(d) Prenaration of the Pevtide CONRYTDLVAIONKNE (Cys-Gln-Asn-Arg-Tyr-Thr-Asn-Leu-Val-Ala-Ile-Gln-Asn-Lys-Asn-Glu-NH2)_ The peptide was synthesized on an Applied Biosystems (Foster City, CA) 430A peptide synthesizer using Fmoc protected amino acids and 2-(1H-Benzotriazol-l-y1)-1,1,3,3-tetramethyluronium hexafluorophosphate (HBATU) esters for amino acid activation. Each amino acid was routinely double coupled. Fmoc protected amino acids and Hydorxybenzotriazole were purchased from Applied Biosystems. All solvents were purchased from Burdick and Jackson. HBTU was Durchased from Richelieu * Trademark SUBSTiNTE SHEET (RULE 26) V J~~NV! =L.LUJJ'.V~/J.
.d._ Biotechnologies (St. Hyacinthe, Canada). Piperidine and trifluoroacetic acid, acetic anhydride, thioanisole, phenol and ethanedithiol were purchased from Sigma Chemical Corporation.
5 F7moc-Amide resin (Bachem Biosciences) was loaded into the peptide synthesis reaction vessel and washed one time with N-methylpyrolidone (NMP). The following operations were then sequentially performed:
1. The Fmoc protecting group was removed by 10 treatment of the resin bound amino acid with 25% piperidine in NMP.
2. The resin was washed 5 times with NMP.
3. A mixture containing N-a-Fmoc-L-Glutamic acid 7-t-butylester, diisopropylethylamine, HBTU and 15 NMP was added to the reaction vessel and allowed to react for 30 min, under vortex agitation.
4. The solvent was drained, and the resin was washed three times with NMP.
20 S. Steps (3) and (4) were repeated two more times.
6. The resin was washed four more times with NMP.
Steps 1-6 were repeated for each amino acid of the peptide.
Following the final coupling cycle, the resin-bound peptide 25 was deprotected by reaction with 25% piperidine in NMP, washed 7 times with NMP, and washed 2 times with dichioromethane.
The resin was dried in vacuo for 24 h. The peptide was cleaved from the resin by treatment with trifluoroacetic acid containing 2.5% ethanedithiol, 5% thioanisole, 7.5% phenol, 30 and 5% water. The polystyrene resin was separated from the trifluoroacetic acid solution by filtration. Trifluoroacetic acid was removed by evaporation in vacuo. The crude peptide was triturated with diethylether and dissolved in water. The water was removed by lyophilization. The peptide was then 35 purified by reverse phase HPLC on a C8 column (vYDAC)* using a gradient of acetonitrile, water, each containing 0.1% TFA as modif ier .

* Trademark SUBSFiNTE SHEET (RULE 26) 11U Y4/L_;;o ik,ai L. +1.v+~J

(e Biotinylation of Monoclonal Antibodies: Antibodies were dialyzed against 0.1M sodium bicarbonate, pH 8.5. NHS-LC
Biotin*(Pierce) was added to 0.3 mg/ml. After 2 hrs at room temperature, antibodies were dialyzed against several changes of PBS.
(f; Coatina of Microtiter Plates for ELISA: 96 well microtiter plates (Falcon Microtest*III) were coated overnight at 4 C with 50 l/well of 2 g/ml affinity purified P-selectin in DPBS.
(g) Western Blotting: mu MAbs PB1.3, PNB1.6, and 84-26 all bind to a single band of 140 KD when platelets are dissolved in SDS-PAGE running buffer (unreduced), subjected to SDS-PAGE, and transferred to nitrocellulose (data not shown). We have also found that mu MAbs PB1.3 and PNB1.6 no longer recognize P-selectin on western blots when the samples have been reduced with ~B-mercaptoethanol.

II. Effect of Chelating Divalent Cations on Binding of Monoclonal Antibodies to P-Selectin: Two microtiter plates were coated with affinity purified P-selectin as above.
Plate 1 was blocked with 200 l/well PBS+1% BSA, while plate 2 was blocked with 200 l/well DPBS+1% BSA (DPBS contains Ca++
and Mg++). After 1 hr, the plates were washed (all washes for plate 1 were PBS, for plate 2, with DPBS. 25 l of 25 mM EDTA
in PBS+1% BSA was added to the wells of plate 1, and 25 l of DPBS+1% BSA was added to plate 2. After 1 hr, 25 l of dilutions of the appropriate antibody were added to the wells of either plate. Dilutions for plate 1 were made in PBS+1%
BSA, for plate 2 in DPBS+1$ BSA. After 1 hr, the plates were washed, and 50 l of a 1 to 1000 dilution of sheep anti-mouse IgG horseradish peroxidase conjugate (dilutions for plate 1 were made in PBS+1% BSA, for plate 2 in DPBS+1% BSA) was added. After 1 hr, the plates were washed, and 50 l of the peroxidase substrate TMB (3,3',5,5'-tetramethylbenzidine, Kirkegaard and Perry Laboratories, Inc.) was added. After the color had developed to the appropriate level, the substrate * Trademark SUBSTTfUTE SHEET (RULE 26) WO 94/25067 Pl'T'/US94/04935 57 ~162149 reaction was quenched by the addition of 25 l of 1M
phosphoric acid and the absorbance at 450 nm was read.
Fig. 4 shows the binding of the anti-P-selectin monoclonal antibodies to P-selectin in the buffer DPBS, which contains CA++ and Mg++, and in PBS + 25 mM EDTA, a chelator of divalent metal cations. The concentration of divalent cations present in DPBS is more than sufficient to support neutrophil adhesion to P-selectin (Geng et al., 1991), while the chelator EDTA, at 25 mM, is more than sufficient to block neutrophil adhesion to P-selectin. The binding of all of the monoclonal antibodies to P-selectin is little affected by the presence of EDTA, demonstrating the Ca++ is not required to be present for binding to occur. Mu MAb PB1.3 is a blocking antibody, that is, it is able to block the binding of neutrophils P-selectin, for example on activated platelets. The PNB1.6 is a non-blocking antibody. The ability of 84/26 to block the binding of neutrophils to P-selectin has not yet been fully characterized. In contrast, of the anti-P-selectin monoclonal antibodies described by Geng et al., only the non-blocking antibody S12 is able to bind in the absence of Ca++.

III. Effect of the Pegtide C NRYTDLVAI NKNE on the binding of mu MAb PB1.3 to P-Selectin: A microtiter plate was coated with affinity purified P-selectin as above, blocked with 200 l/well DPBS+1% BSA for 1 hr, and washed. Dilutions of mu MAb PB1.3 in DPBS+1% BSA were mixed with either 0.35 mg/ml CQNRYTDLVAIQrIIQdE in PBS, or as a control, PBS alone. After 1 hr, 50 l of each antibody dilution was added to the microtiter plate, and incubated for 1 hr. The plate was washed with DPBS, and a 1 to 1000 dilution of sheep anti-mouse IgG horseradish peroxidase conjugate in DPBS+1% BSA was added.
After 1 hr, the plate was washed, the substrate TMB was added, and the color development was stopped and the plates read as above.
Fig. 5 shows that the peptide CQNRYTDLVAIQNKNE, homologous to residues 19-34 of the lectin domain of P-selectin, has no effect on the binding of mu MAb PB1.3 to P-selectin when the peptide is present at a concentration of SUBSTIME SHEET (RULE 26) L1b2149 0.35 mg/ml. This distinguishes this particular monoclonal antibody from the monoclonal antibodies Gi, G2, and G3, whose binding to P-selectin is partially or completely blocked by this peptide.
IV. Ability of Cvtel Mu MAbs PB~ :4,1 PNB1 6 and 84/26 to Block Binding of Each Other to-PtSglectin: Microtiter plates coated with affinity purified P-selectin as above were blocked for 1 h with DPBS+1$ BSA, and washed. 25 l of a 50 g/ml solution of mu MAbs PB1.3, PNB1.6, or 84/26 in DPBS+1$ BSA
were added to appropriate wells, and incubated for 1 hr.
Then, 25 l of a dilution (in DPBS+1% BSA) of biotinylated antibody was added. The plate was then incubated on a platform shaker for 1 hr. After washing, 50 l of a 1 to 1000 dilution of a streptavidin horseradish peroxidase conjugate was added, and incubated for 1 hr. After washing with DPBS, the plate was developed with the substrate TMB as above.
Fig. 6 shows the ability of the mu MAbs PB1.3, PNB1.6, and 84/26 to interfere with the binding of each other to P-selectin. Fig. 6A shows that only mu MAb PNB1.6 is able to block the binding of biotinylated mu MAb PNB1.6 to P-selectin, demonstrating that mu MAbs PB1.3 and 84/26 must recognize different epitopes from mu MAbs PNB1.6. Likewise, In Fig. 6B, mu MAb PB1.3, but not mu MABs PNB1.6 or 84/26, is able to block the binding of biotinylated PB1.3 to selectin, demonstrating that the other two monoclonals must recognize different epitopes. Fig. 6C, both mu MAbs 84/26 and PB1.3, but not mu MAb PNB1.6, are able to block the binding of biotinylated mu MAb 84/26 to P-selectin.
V. Blocking P-selectin antibodies do not inhibit thrombin-induced aaaregation of human platelets:
This example demonstrates that the P-selectin antibodies, mu MAbs PB1.3 and PNB1.6 do not inhibit thrombin-induced aggregation of platelets. Fresh human platelets were isolated as previously described and suspended in Tyrode-Hepes Buffer, pH 7.2 at a concentration of 2x108/ml.

SUBSTITUTE SHEET (RULE 26) Platelet aggregometry was performed in a Lumi aggregometer (Chrono-Log Corp.). For the assay, 0.45 ml of the platelet suspension was placed in a standard siliconized cuvette. To this was added 10 l of the antibody or vehicle control. After 5 min. at 37 C, 0.1 unit of thrombin (Human, Sigma) was added and aggregation was recorded.
When platelet aggregation was measured under control conditions, in the absence of antibody, an aggregation response of 71 units was obtained. An equivalent extent of aggregation was obtained in the presence of 10 l of undiluted culture supernatants from cells producing mu MAb PB1.3 antibody (Fig. 7). Different concentrations of mu MAb PB1.3 (1-100 g/ml) were added to the platelet suspension. At no concentration of mu MAb PB1.3 was platelet aggregation different from the response determined in the absence of antibody (Fig. 7).

Examnle 9 In vivo Protective Effects of Blockina P-Selectin Antibody.
mu MAb PB1.3 in Feline Myocardial Ischemia and Reperfusion A model of myocardial ischemia and reperfusion injury in cats was performed according to the methods of Tsao et al. (Circulation 82:1402-1412 (1990)). Briefly, male cats (2.5-3.5 Kg) were anesthetized with sodium pentobarbital (30 mg/Kg, i.v.). An intratracheal tube was inserted through a mid-line incision and all cats were given intermittent positive pressure ventilation by a Harvard small animal respirator. A polyethylene catheter was inserted into the external jugular vein and the right femoral vein was cannulated and connected to a Statham P23Ac pressure transducer for the measurement of arterial blood pressure. A
midline thoracotomy was performed, the pericardium was opened and the heart was exposed. A 2-0 silk suture was carefully placed around the left anterior descending artery (LAD) 10-12 mm from its origin. After a 30 min period of stabilization myocardial ischemia was initiated by complete ligation of the LAD for 1.5 h of ischemia followed by 4.5 h of reperfusion.
Blocking (mu MAb PB1.3) and non-blockina (mu MAb PNB1.6) anti-SUBSTIME SHEET (RULE 26) P-selectin antibodies were administered intravenously at a dose of 1 mg/kg 10 min prior to the initiation of reperfusion.
The ischemic myocardium was determined as that portion of tissue which did not stain with nitroblue tetrazolium and was expressed as a percentage of the area at risk. Area at risk was determined by reoCclusion of the LAD
at the end of the reperfusion period followed by the injection of Evans blue dye into the left atrium. The area of risk was, therefore, determined by negative staining.
Endothelial dependent relaxation of coronary artery rings was determined by measuring the acetylcholine-induced relaxation of rings previously contracted with the thromboxane A2 mimetic U46619. Data are expressed as percentage relaxation.
Neutrophil accumulation in the myocardium was measured as the myeloperoxidase activity of polytron homogenized myocardial tissue.
The results of these experiments are shown in Fig. 8. In cats treated with the non-blocking anti-P-selectin antibody mu MAb PNB1.6 the extent of myocardial ischemia was 33 5% of the area at risk. In contrast, the extent of ischemia in PB1.3 treated animals'was significantly (P<0.01) less at 15 3% of the area at risk. Endothelial dependent relaxation to acetylcholine was also significantly preserved in ischemic-reperfused coronary arteries taken from cats treated with mu MAb PB1.3 compared to mu MAb PNB1.6 (67 6 vs 11 3%, P<0.01). That these phenomena are attributable to a reduction in myocardial leukocyte infiltration is suggested by a significant (P<0.1) reduction in the number of PMNs in myocardial tissue from mu MAb PB1.3-treated animals (64 7 PMN/mm2) compared with mu MAb PNB1.6 treated animals (136 12 PNIId/mm2) .

SUBSTITUTE SHEET (RULE 26) xa le 10 Production of Humanized Antibodies to P-Selectin A. Clonina and Sequencincg of Mu MAb PB1.3 Variable Regions The isolation of mu MAb PB1.3 is described in Example 2. Tcatal RNA was isolated from hybridoma cells producing mu MAb-'.PB1.3. The cDNAs encoding the variable regions of the heavy and light chains of mu MAb PB1.3 were amplified by polymerase chain reaction (PCR). The entire mouse variable region was amplified using a mixture of degenerate primers that hybridized within the leader sequence and a single primer that hybridized to the mouse constant region near the V-C junction. See Jones & Bendig, Bio/Technology 9:88-89 (1991) and Bio/Technology 9:579 (1991).
The PCR fragments were cloned and sequenced. The nucleotide sequences and the corresponding amino acid sequences for the variable regions of the heavy and light chains of mu MAb PB1.3 are given in Figs. 9 and 10, respectively.

B. Modelling the Structure of the Mu MAb PB1.3 Variable Regions A molecular model of the VL and VH regions of mu MAb PB1.3 was built. The model was built on a Silicon Graphics IRIS 4D workstation running under the UNIX operating system and using the molecular modelling package QUANTA (Polygen Corp., USA). The amino acid sequence of the mu MAb PB1.3 variable regions were compared with the sequences of structurally-solved immunoglobulin variable regions. See Tables 1 and 2.

C. Design of Reshaped MAb PB1.3 Variable Reaions Human variable regions were selected that have framework regions most similar to those found in the mu MAb PB1.3 variable regions. The human mAb DEN light chain variable region and the human mAb 21/28'CL heavy chain variable region were selected as frameworks to join the corresponding complementarity determining regions (CDRs) of mu MAb PB1.3. The human framework regions were modified using the methodology described in Monoclonal Antibodies, 2:

SUBSTITUTE SHEET (RULE 26) AppL_~cations in Clinical Oncology (Editor A. Epenetos:
Chapnan & Hall, 1992;
see especially, Bendig et al., The Humanization of Mouse Monoclonal Antibodies by CDR-Grafting: Examples with Anti-Viral and Anti-Tumor Cell Antibodies contained therein, which is incorporated herein by reference). The alignment of amino acids of the variable regions leading to the design of the reshaped humanized PB1.3 variable regions are given in Tables 1 and 2.
Three versions of the heavy chain variable region were modelled and were designated CY1748RHA, CY1748RHB, and CY1748RHc. The nucleotide sequences and corresponding amino acid sequences of these variable regions and signal peptides are given in Figs. 11, 12, and 13, respectively. In the case of CY1748RH8 the first two mouse residues conserved in CY1748RHa (glutamic acid and alanine) were changed back to the original human amino acids (glutamine and valine) found in the donor human FRi. In the case of CY1748RHc, computer modelling suggested that the murine lysine at position 73 might form a salt bridge with residue 55 (aspartic acid) in CDR2. If this were the case, then the lysine would be important for the stabilization of the H2 loop. The human threonine residue at this position would be unable to form such a salt bridge.
Therefore, version CY1748RHc substitutes the murine lysine at position 73 of CY1748RIia. Version CY1748RH$ contains the highest number of human amino acid residues. A summary of the amino acid differences among these three versions of the heavy chain is given in Table 3.
Four versions of the light chain variable region were modelled and were designated CY1748RLA, CY1748RL,B, CY1748RLc, and CY1748RLD. The nucleotide sequences and corresponding amino acid sequences of these variable regions and signal peptides are given in Figs. 14, 15, 16 and 17, respectively. Computer modelling suggested that the murine residue glutamic acid at position 60 might be important in P-selectin binding. This residue was incorporated in version CY1748LRA but replaced with the original human framework residue serine in version CY1748LRa. The murine aspartic acid SUBSTiME SHEET (RULE 26) residue might have a charge interaction with residue 24 in loop 1(L1), therefore CY1748LRc is identical to CY1748LRA
except for the replacement of the mouse residue aspartic acid at position 70 for the human framework amino acid glutamic acid. CY1748LRD incorporates both these amino acid replacements and of the four versions contains the greatest number of human amino acid residues. A summary of the amino acid differences among these four versions of the light chain is given in Table 4.
D. Subcloning of cDNAs Encoding Heavv and Liaht Chain Variable Regions Into Exnression Vectors (1) Construction of Exoression Vectors (a) nHCMV-1748RL-KR-neo The expression vector contains the human cytomegalovirus (HCMV) enhancer and promoter to drive transcription of the recombinant immunoglobulin light chain, the coding sequences of the bacterial neo gene to act as a dominant selectable marker during stable transformation, and the SV40 origin of replication to give high levels of transient expression in cos cells. To construct the vector, a HindIII-SacIi fragment from a pUCB vector containing the PstI-m fragment of HCMV (Boshart et al., Cell 41:521-530 (1985)) was converted to an EcoRI-HindIiI fragment using appropriate adaptor oligonucleotides. In a three-way ligation, the 1.2 kb EcoRI-HindiII fragment containing the HCMV enhancer-promoter was ligated to a 5.05 kb BamH-EcoRI fragment from pSV2neo (Southern and Berg, J. Mol. App. Goner. 1:327-341 (1982)) and a 0.5 kb HindIII-BamHI fragment containing the VLlys kappa light variable region (Foote and Winter, J. Mol. Biol.
224:487-499 (1992)). The Hindlii site in pSV2neo had previously been removed by filling-in with Klenow polymerase producing a vector designated pHCMV-VLlys-neo.
A"2.6 kb EcoRI fragment of the genomic human kappa region was cloned (Rabbitts et al., Curr. Top. Microbiol.
Immunol. 113:166-171 (1984)) and subcloned into a pUC vector, pUCKR17 (Mengle-Gaw, EMBO J. 6:1959-1965 (1987)). BamHI sites were added to both ends of the fraqment usinq EcoRI-BamHI

SUBSTIME SHEET (RULE 26) adapters. This fragment was then inserted into the BamHI site of pHCMV-VLlys-neo, in the appropriate orientation, to create pHCMV-VLlys-KR-neo. To facilitate insertion of reshaped human variable regions, the BamHI site on the 3'-side of the human kappa constant region was removed by filling-in with Klenow polymerase. In the resultlhg vector, designated pHCMV-1748RL-KR-neo, the HindIII-BamHI fragment containing the VLlys is easily replaced with the VL of a reshaped antibody.

(b) pHCMV-1748-y1C-neo The construction of this vector began with the same three way ligation as described above for generation of pHCMV-VLlys-neo above save for the insertion of a 0.7 kb HindIII-BamHI fragment containing the VHlys heavy variable region (Verhoeyen et al., Science 239:1534-1536 (1988)); Foote and Winter, supra) in place the 0.5 kb HindIII-BamHI fragment containing the VLlys kappa light chain. The cDNA coding for human 71 constant region was inserted into the BamHI site to produce a vector designated pHCMV-VHlys-neo. cDNA coding for human ryl constant region was cloned from a human cell line secreting a human ryl antibody by PCR amplification. BamHI
sites were created at each end of the cDNA, and a splice acceptor site and a 65bp intron sequence were introduced at the 5-end of the cDNA sequence. The BamHI fragment (1176bp) containing the human 71 cDNA plus the splice acceptor site and intron sequence was then cloned into the expression vector.
As with the light chain expression vector, the BamHI site at the 3'-side of the human 71 constant region was removed by filling-in with Klenow polymerase. With this vector, designated pHCMV-1748-71C-neo, the HindIII-BamHI fragment containing the unwanted VHlys is easily replaced with the VH
of a reshaped antibody.

(2) Insertion of Immunoglobulin Codina Seauences The cDNAs encoding the heavy chain variable region of mu MAb PB1.3 and the three humanized versions were inserted into the expression vector HCMV-71C-neo by standard techniques SUBSTITUTE SHEET (RULE 26) ~. ~

generating expression plasmids HCMV-1747CH-y1C-neo, HCMV-1748RHA-ylC-neo, HCMV-1748RH8-y1C-neo, and HCMV-1748RHC-71C-neo. The nucleotide sequence of the human IgGi constant region cDNA and corresponding protein sequence is given in 5 Ellison et al., Nuc. Acids Res., 10:4071-4079 (1982). The cDNAs encoding the light chain variable regions of CY1747(PB1.3) and the four humanized versions were inserted into the expression vector HCMV-EIt-neo generating plasmids HCMV-1747CL-Kt-neo, HCMV-1748RLA-Kt-neo, etc. The nucleotide 10 sequence of the human kappa constant region gene and corresponding protein sequence is shown in Hieter et al., Cell 22:197-207 (1980) Schematic maps of the heavy chain and light chain expression vectors are given in Figs. 18 and 19.
15 By coexpressing appropriate combinations of heavy and light chains, several antibodies can be expressed. For example, coexpression of HCNIV-1748RHA--YiC-neo and HCMV-1748RLA KR-neo gives rise to the reshaped MAb PB1.3 (HA/LA). Coexpression of HCMV-1747CH-71C-neo and HCMV-1747CL-Kt-neo gives rise to the 20 chi MAb PB1.3 (i.e., chimeric antibody comprising murine variable domains and human constant domains).

E. Flectrotioration of COS Cells The DNA constructions were tested in COS cells for 25 the transient expression of chi MAb PB1.3 and various reshaped versions of MAb PB1.3 antibodies. DNA was- introduced into the COS cells by electroporation using the Gene Pulser apparatus (Biorad). COS cells were trypsinized and washed once in phosphate buffered saline (PBS). DNA (10 g of each heavy 30 chain plasmid and appropriate light chain plasmid) and a 0.8 ml aliquot of 1 x 10'7 cells/ml in PBS were placed in a sterile Gene Pulser*Cuvette (Biorad, 0.4 cm gap). A pulse was delivered at 1900 volts, 25 microfarads capacitance. After a 10 minute recovery period at ambient temperature, the 35 electroporated cells were added to 20 ml of DMEM media (GIBCO/BRL) containing 10% fetal bovine serum. After a 48 h incubation, the medium was collected, centrifuged to remove * Trademark SUBS'iTTIJTE SHEET (RULE 26) cellular debris, and stored under sterile conditions at 4 C
for short periods of time.

ExamQle 11: Analysis of Humanized Antibodies Media from the transfected COS cells were assayed by ELISA to quantify levels of human antibody produced and analyze capacity to bind to recombinant soluble P-selectin (rsP-selectin). Immulon microtiter,plates (Dynatech, #011-010-3355) were coated with 100 l/well of a solution of goat anti-human IgG (whole molecule, Sigma #1-1886) to a final concentration of 12.5 g/ml or a solution of rsP-selectin (1.2 mg/ml) diluted in Dulbecco's phosphate buffered saline (DPBS).
Plates were coated overnight at 4 C or for 2 h at 37 C. The plates were then washed three times with DPBS. The plates were then blocked with 400 l/well of DPBS + 1% bovine serum albumin (BSA, Sigma, #A-7888) for one h or longer at room temperature. The plates were then washed three times with DPBS.
Serial dilutions of the monoclonal antibodies in DPBS + 1% BSA were added to the plates (100 i/well), and the plates were incubated for 1 h at room temperature. The plates were then washed three times with DPBS. 100 l/well of a second antibody (goat anti-human IgG peroxidase conjugate, Sigma #A-8867) diluted 1 to 1000 in DPBS + 1% BSA was added to the plates, and the plates were incubated for one h at room temperature. The plates were then washed three times with DPBS. 100 l/well of tetramethylbenzidine peroxidase substrate/H202 (TMB, Kirkegaard and Perry Laboratories #50-76-00) was added to the plates, and the color was allowed to develop for an appropriate period of time (usually 3 to 15 min), and the reaction was quenched by the addition of 100 l/well of 1-M phosphoric acid. The plates were read at 450 nm in a Titertek Multiskan MCC/340.
The concentration of the antibodies produced in all COS cell supernatants was determined by comparing the IgG
ELISA results to a human IgGl antibody (kappa light chain, Sigma #1-3889) of known concentration.

SUBSTtTUTE SHEET (RULE 26) P-selectin binding was plotted as optical density against the concentration of IgG. Binding curves of chi PB1.3 and various reshaped versions of reshaped MAb PB1.3 are given in Figs. 20, 21, and 22. The objective was to select a humanized antibody showing substantially indistinguishable binding characteri5tics from chi MAb PB1.3. If more than one humanized antibodies showed such characteristics, the one containing the most human amino acid residues (or the fewest murine amino acids) would be preferred, because this version would be least likely to induce a HAMA response in clinical use. All four of the reshaped light chains coexpressed with heavy chain CY1748RHA generated antibodies that bound to rsP-selectin as well as the chi MAb PB1.3. Of these, light chain variable region CY1748LRD was selected for further experiments, because it contains the most human amino acid residues. The heavy chain variable regions CY1748RHA and CY1748RH8 when coexpressed with light chain variable regions CY1748RLA and CY1748RL, also generated antibodies that bound to rsP-selectin to the same extent as chi MAb PB1.3 (Fig. 22).
Of these heavy chains, the CY1748RH, heavy chain is preferable because it contains the most human amino acid residues. Thus, a preferred humanized antibody is formed from light chain D
and heavy chain B and is designated reshaped MAb PB1.3 (Hg/I,D) .
Examole 12 Stable Exnression of Reshaoed MAb PB1.3 (H-/I.) in CHO Cells A. Introduction of Humanized Immunoalobulin Chains in CHO
cells (1) Construction of Vectors (a) pHCMV-1748RH-21C-dhfr This vector is identical to pHCMV-1748RH-71C-neo described in Example 11 except that the neo gene is replaced by the dihydrofolate reductase (dhrf) gene linked to a defective SV40 promoter-enhancer sequence. In order to remove the enhancer sequence from the SV40 early promoter, the plasmid DNA, pSV2-dhfr (Subramani et al., Mol. Cell. Biol.
1:854-864 (1981)) was digested with SphI and PvuII, filled-in ~gSTiTUTE SHEET (RULE 26) with Klenow polymerase, and self-ligated to yield pSV2-dhfr-AE. A-3.7 kb EcoRI fragment containing the HCMV promoter, the VHlys heavy chain variable region, and a genomic DNA clone of the human yi constant region was excised from the plasmid pHCMV-VHlys-yl-neo with EcoRI. This fragment was ligated to EcoRI-digested pSV2-dhfr-AE to create pRCMV-VHlys-yl-dhfr.
The genomic clone of the y1 constant region was replaced with a BamHI fragment containing the cDNA clone of the human yl constant region-described in Example 11. The BamHI site on the 3'-side of the human yi constant region was then removed by filling in with Klenow polymerase as in Example 11. The resulting vector was designated pHCMV-174RH-yC-dhfr. The HindIII-BamHI fragment of this vector containing the unwanted VHlys is easily replaced with the VH of a reshaped antibody.
(b) DHCMV-1748RH-3~4-dhfr This vector is identical to pHCMV-1748RH-y1C-dhfr described above, but for the replacement of the cDNA clone of the human yl constant region with a genomic clone of the human y4 constant region. To construct the vector, a" 7.0 HindIII
fragment of DNA containing the genomic clone of the human y4 constant region (BrUggeman et al., J. Exp. Med. 166:1351-1361 (1987)) was subcloned into the HindiII site of pUC19 to create a plasmid called 428D. Because the human 74 fragment already had a BamHI site near its 3-'end, the subclone was inserted into pUC19 in the orientation that placed this site distal to the BamHI site in the polylinker of pUC19. The 7.0 kb human y4 fragment was excised from 428D using BamHI and ligated into pHCMV-Vxlys-neo to create the plasmid pHCMV-VHlys-y4-neo.
The final plasmid was constructed by a three-way ligation of a 5.4 kb BamHI-HindIII fragment containing the HCMV enhancer-promoter and the pSV2-dhfr-AE plasmid sequence, a 0.5 kb HindIII-BamHi fragment containing the VH of reshaped human 1748 antibody and the 7.0 kb BamHI-BamHI fragment containing the genomic clone of the human y4 constant region.
SUBSTITUTE SHEET (RULE 26) (2) Insertion of Immunoclobulin Coding Seauences The cDNA encoding 1748RHA heavy chain variable region was cloned into expression vector HCMV-1C-dhfr generating HCMV-1748RHA-1C-dhfr. The cDNA encoding 1748RH8 was cloned into expression vectors HCMV-1C-dhfr and HCMV-4C-dhfr generating the plasmids HCMV-1748RH8-1C-dhfr and HCMV-1748RHB-4C-dhfr, respectively. The nucleotide sequence of the human IgG4 constant region gene and corresponding amino acid sequence is given by Ellison et al., DNA, 1:11-18 (1981). The vector HCMV-1C-dhfr expresses an IqGl isotype antibody and vector HCMV-74C-dhfr expresses an IgG4 isotype antibody when they are each coexpressed with the kappa light chain vector.
The schematic maps of the dhfr-expression vectors are given in Figs. 23 and 24, respectively.
CHO dhfr-deficient cells were grown in aMEM (+
nucleosides, GIBCO/BRL) and 10% fetal bovine serum. Plasmid DNA was introduced into the COS cells by electroporation using the Gene Pulser apparatus. CHO cells were trypsinized and washed once in phosphate buffered saline (PBS). DNA (10 g of each heavy chain plasmid and appropriate light chain plasmid) and a 0.8 ml aliquot of 1 x 107 cells/ml in PBS were placed in a sterile Gene Pulser Cuvette (0.4 cm gap). A pulse was delivered at 1900 volts, 25 microfarads capacitance. After a 10 minute recovery period at room temperature, the electroporated cells were added to 20 ml of aMEM (+
nucleosides)/10% FBS. After a 24-48 h incubation cells were trypsinized and plated into 100 mm dishes in aMEM (-nucleosides)/10% dialyzed FBS (to select for the expression of the dhfr-containing plasmid) supplemented with 500 g/ml G418 (GIBCO/BRL, to select for the expression of the neo-containing plasmid). Media was changed every 3-4 days until colonies emerged. Single clones were isolated via cloning cylinders, expanded and analyzed for IgG production via ELISA.
Three sets of DNAs were introduced into the CHO
dhfr-deficient cells: 1) HCMV-1748RHA-71C-dhfr and HCMV-1748RLA-KR-neo expressing reshaped MAb PB1.3 (HA/LA), IgGl isotype; 2) HCMV-1748RHB-71C-dhfr and HCMV-1748RLD-Kt-neo expressing reshaped MAB PB1.3 (HB/LD) IgG1 isotype; and HCMV-SUBSTIME SHEET (RULE 26) 1748RHB--y4C-dhfr and HCMV-1748RLp-KR-neo expressing a reshaped MAb PB1.3 (HB/Lp), IgG4 isotype.
CHO cells were expanded and seeded into 10-chamber (6000 cm2 total surface area) Nunc cell factories. The medium was harvested at 72 h and passed over a protein-A Sepharose Fast Flow (Pharmacia) column. The colubn was washed and bovine IgG was eluted at pH 4.5. Thp humanized antibodies were eluted at pH 3.5, dialyzed in PBS or 20 mM acetate, 0.15 M NaCl, pH 5.5. Antibody concentration was determined spectrophotometrically and confirmed by an ELISA for human IgG. (A purified human IgG4 antibody (kappa, Sigma 1I-4639) was used as the.control for mu MAb PB1.3 (HB/LD) IgG4 isotype.) Purity was confirmed by SDS-PAGE.

B. Competitive Bindina Assay The relative binding affinities of the various versions of reshaped MAb PB1.3 to P-selectin were determined by competitive ELISA assay. The assay measures the concentration of unlabelled antibody that inhibits by 50% the binding of labelled mu MAb PB1.3 to P-selectin.
Two ml of purified MAb PB1.3 (1 mg/ml) was mixed with 100.8 l of NHS-LC-biotin (1 mg/ml, Pierce) and incubated overnight at 4 C. The 2.1 ml sample was loaded on a NAP 25 column (size exclusion column, Pharmacia). The column was eluted with 0.3 ml fractions with PBS. Each fraction was read at 280 nm. The highest O.D.280 nm fractions were pooled. The biotinylated-MAb PB1.3 was titered against rsP-selectin to select an appropriate concentration to use for the competitive binding assay.
A 96-well plate (Microtest III, Falcon #3912) was coated with 50 l/well of rsP-selectin (diluted to 2 g/ml in DPBS) overnight at 4 C or 90 min at 37 C. The plate was inverted and residual fluid was tapped away.
The coated plate and an additional blank plate were blocked with 200 l/well of DPBS + 1% bovine serum albumin (BSA, Sigma #A-7888), for 1 h at room temperature. The blank plate was inverted and the residual blocking fluid was tapped away. The rsP-selectin coated plate was allow to continue SUBSTITUTE SHEET (RULE 26) blocking during the antibody dilution period. Serially diluted sample antibodies (in DPBS + 1% BSA) were added to the blank plate at 50 l/well. The initial concentrations of the antibodies were 100 g/ml. Biotinylated-PB1.3 (diluted to 0.3 g/ml in DPBS + 1% BSA) was added to the sample antibodies in the blank plate, 50 l/well, and mixed by pipetting up and down. The rsP-selectin coated plate was washed three times with DPBS. The sample antibody/biotinylated -PB1.3 mixture was pipetted into the plate coated with rsP-selectin, 50 l/well, and allowed to incubate for one h at room temperature. The plate was then washed three times with DPBS.
Streptavidin horse radish peroxidase conjugate (Pierce 021124) diluted 1:2000 in DPBS +1% BSA was added to the plate, 50 l/well and allowed to incubate for 1 h at room temperature. The plate was then washed three times with DPBS.
50 l/well of tetramethylbenzidine peroxidase substrate/H202 (T1rID, Kirkegaard and Perry Laboratories #50-76-00) was added to the plates, and the color was allowed to develop for an appropriate period of time (usually 3 to 15 min), and the reaction was quenched by the addition of 50 l/well 1 M phosphoric acid. The plates were read at 450 nm in a Titertek Multiskan MCC/340.
Apparent binding affinity is expressed as the concentration of antibody that binds rsP-selectin at one-half maximum optical density at 450 nm. The ratio of concentrations at one-half maximum binding of the sample antibody to that of the control (unlabelled PB1.3) antibody provides an estimate of the relative binding affinity of a test antibody to rP-selectin.
Results from two typical competitive binding experiments are given in Figs. 25 and 26. In the first experiment, twice as much reshaped MAb PB1.3 (HALA), IgGl isotype (2.0 g/ml) as PB1.3 (1.0 g/ml) is needed to inhibit the binding of labelled PB1.3 to 50% of maximum. Therefore this humanized antibody binds rsP-selectin 50% as well as PB1.3. Similarly, in the second experiment PB1.3/(HBLfl) (IgG4 isotype) binds to rsP-selectin 50% as well as PB1.3.

SUBSTIME SHEET (RULE 26) L'162 149 Example 13 Increasing Antibody Production in CHO Cells Via DHFR Amplification CHO cells were plated into flasks and allowed to reach confluence before changing media (8 ml for a T-25 cm2 flask or 20 ml for a T-75cm2 flask). After incubating the cells for 48-72 h the media was harvested'a,nd assayed for IgG
production by ELISA. The cells were trypsinized and counted.
Antibody production rate was expressed as the amount of antibody in micrograms secreted by one million cells in a 24 h period.
The three to six highest-producing clones were amplified separately, and the remaining seven best clones were pooled and amplified. Cells were plated at a density of 1 x 105 cells/100 mm dish in selection media. For the first round of amplification cells were plated in three sets of dishes containing aMEM(- nucleosides)/10% dialyzed FBS supplemented with 500 g/ml G418 and either 10, 20 or 50 nM methotrexate (Sigma IA-6770). Cultures were fed every four days. At 10-14 days single colonies were isolated by cloning cylinders, expanded, and assayed for IgG production by ELISA. Additional rounds of amplification were performed on single clones or pools of clones at methotrexate concentrations 5-10 times the initial methotrexate concentration.
Table 5 provides a summary of the antibody production for the CHO cells expressing three humanized versions of PB1.3. One amplified clone expressing reshaped MAb PB1.3 (HBLD) (IgG4 isotype) arose from a pool of clones that expressed less than 0.2 g/106 cells/day, amplified to 2.0 g/106 cells/day in the first round, and reached 33.0 g/106 cells/day after the third round. This clone was selected to be grown for large scale production of reshaped MAb PB1.3.

SUBSTITUTE SHEET (RULE 26) ~~ U y 3~Z~vo % PCTIUS94104935 -~.- 73 Example 14 Bioreactor Run 160-g collagen-coated microcarrier beads (150-200 microns, JRH Biosciences #60142-100) were hydrated in 800-m1 of sterile water for 30 min and the water decanted. The beads were resuspended in 800-mi water and autoclaved for 30 min.
After cooling to room temperature the beads were washed twice with sterile serum free aGMEM(-nucleosides) and resuspended in 1-1 complete media aMEM(-nucleosides)/5% dialyzed FBS.
The cell line CHO-48B4, which expresses reshaped MAb PB1.3 (HBZ.n) (IgG4 isotype) at 33.0 g/106 cells/day, was expanded into 36 T-225 cm2 in acMEM(-nucleosides)/10t dialyzed FBS/500 nm methotrexate. When the cells reached confluence they were trypsinized and resuspended in 2-1 complete media (no methotrexate) at a final concentration of 5.0 - 8.0 x 10 cells/ml.
1-1 beads, 2-1 cell suspension, and 2-1 fresh complete media were aseptically added to the 10-1 autociaved vessel of a Proteus* 2000 Bioreactor (Wheaton). The headspace was continuously perfused with air and 5% CO2. Cells were allowed to attach to the microcarriers by programming the Proteus to stir the cell/microcarrier suspension at 70 rpm for 1 min and 0 rpm for 59 min. After a total of 48 cycles, 5-1 media was added to the bioreactor, which was programmed to maintain the suspension at 37 C, pH at 7.0, dissolved oxygen at 30% that of air, and stirred continuously at 50 rpm.
After 3 days the bioreactor was programmed to replace complete media at 5-1/day utilizing a gravity filter attached to the harvest port (to leave microcarriers in the bioreactor). The harvested media was stored at 4 C until processing. The bioreactor was run for 5-7 days until accumulated cell debris clouded the media harvest (due to aeration damage).
The media harvest was clarified by passing though a Millipa)~-60 0.45 micron filter (Millipore). 20-1 filtered supernatant was passed over a column (2.5 x 12 cm) of protein-A Sepharose Fast Flow (Pharmacia) at a flow rate of 0.55 1/hr.
After all of the supernatant had passed through the column, * Trademark SU$STITUTE. SHFFT roi it c~~.

the column was washed with 1.5-1 PBS (BioWhittaker, #17-516Y).
The column was pre-eluted with 8 x 40-m1 0.2-M sodium acetate, pH 4.5, and eluted with 8 x 40-m1 0.2-M glycine, 0.5-M NaCL, pH 3.5, both into tubes containing 10 ml 2-M Tris-HC1, pH 8Ø
The single fraction with the highest absorbance at 280 nm was dialyzed into 20-mM acetate, 0.15-M NaCl, pH 5.5, and stored at 4 C.
200-mg antibody was harvested per day for 5-7 days.
The purification yield was approximately 50%. Therefore, one bioreactor run yields 500-700 mg purified antibody.

ExamDle 15 The Effect of Mu MAb PB1.3 on Tissue Injurv Produced by Tschemia Renerfusion Following Renlantation of the Rabbit Ear New Zealand White rabbits were anesthetized with a combination of ketamine and xylazine followed by infiltration of lidocaine into the base of the ear. The left ear was partially amputated leaving the central artery, vein and a bridge of cartilage intact. All nerves were divided to render the ear totally anesthetic. The ear was then reattached and a microvascular clip was placed across the artery to produce total ischemia. Rabbits were kept in a room maintained at 23.5 C for 6 h then the microvascular clamp was removed and the ear allowed to reperfuse. Treatment was administered immediately prior to reperfusion with either mu MAb PB1.3(2 mg/kg) or with either saline or an isotype (IgGl) matched murine monoclonal antibody designated PNB1.6 (2 mg/kg). Ear volume was measured daily for 7 days by water displacement to quantify tissue edema. On day 7 necrosis was estimated as a percentage of total surface area.
A significant increase in ear volume was determined in control animals administered either saline or PNB1.6 following reperfusion of the ears (Fig. 27). Because no differences between the two control groups were noted, their data have been combined. An increase in edema was also seen in animals treated with MAb PB1.3, the magnitude of the increase was significantly less than that observed in the control group at all time points measured. The extent of SUBSTITUTE SHEET (RULE 26) necrosis (Fig. 28), measured on day 7, was also significantly reduced in mu MAb PB1.3-treated animals compared with control animals.
Thus, ischemia followed by reperfusion of the replanted rabbit ea.Y,produces an edematous response initially followed by extensive tissue necrosis. These two manifestations of tissue injury are substantially and significantly reduced in animals pre-treated with the anti-P-selectin antibody mu MAb PB1.3.
Example 16 The Effect of Mu MAb PB1.3 on Vascular Patency Following Ischemia and Renerfusion of the Canine Skeletal Muscle This experiment utilizes a model of ischemia and reperfusion of the isolated canine gracilis muscle. One aspect of the reperfusion injury is a phenomenon referred to as the "no-reflow" phenomenon in which blood flow through the microcirculation of a previously ischemic tissue is absent or severely reduced following reperfusion.
An isolated canine gracilis muscle preparation was utilized as described previously (Carden et al. Circ. Res.
66:1436-1444 (1990)). Briefly, adult mongrel dogs were anesthetized with pentobarbital sodium. The skin overlying the gracilis muscle was divided and the muscle was freed from connective tissue by blunt dissection. The obturator nerve was sectioned. The proximal caudal artery and vein were cannulated and all other collateral vessels ligated. The femoral artery cannula was connected to a constant flow perfusion pump and the outflow from the femoral vein drained in to a reperfusion reservoir. Arterial and venous pressures were monitored via side branches and blood flow was adjusted to maintain perfusion pressure at 100 mm Hg. Vascular isolation was assumed complete if arterial perfusion pressure decreased to less than 20 mm Hg when the perfusion pump was switched off.
Microvascular patency was assessed by perfusion of the isolated gracilis muscle with contrast media (india ink) at the end of the experimental protocol. Comouterized video SUBSTfTUTE SHEET (RULE 26) imaging was used to quantitate the number of ink-containing microvessels (< 10 m diameter) per muscle fiber in histologic sections obtained from isolated canine gracilis muscles subjected to 4.5 h of continuous perfusion, 4 h of ischemia followed by 0.5 h of reperfusion or ischemia reperfusion in the presence of the anti-P-selectin antibody mu MAb PB1.3 at a concentration of 40 g/ml in the perfusate.
The ischemia and reperfusion protocol reduced the number of patent microvessels to 36 + 3$ of the value determined in control animals. When mu MAb PB1.3 was included in the perfusate the number of patent microvessels following ischemia and reperfusion was 119 18 % of control.
Thus, the anti-P-selectin antibody, mu MAb PB1.3, completely prevented the development of the "no-reflow"
phenomenon, determined by the number of patent microvessels, in the ischemic and reperfused canine gracilis muscle.
ExamDle 17 The Effect of Mu MAb PB1.3 on Leukocyte Endothelial Cell Interactions Induced by the Degranulation of Tissue Mast Cells I. Methods a) Intravital microscopv Male Wistar rats (200-250g) were obtained from Harlan Sprague Dawley, Indianapolis, and were fasted for 12-24 hrs before the surgical procedure. Surgical anaesthesia was induced by an intramuscular injection of ketamine/rompun/
acepromazine. A catheter was placed in the left jugular vein, and a tracheostomy was performed to facilitate spontaneous breathing. The abdomen was shaved and washed, and a 3/4 inch long incision was made through the midline into the peritoneal cavity, taking care to avoid any bleeding. The rat was placed on a microscope stage specifically designed for the intravital microscopy procedure. A well-vascularized posterior loop of the ileum was exteriorized, and the mesentery draped over a viewing pedestal heated to 37 C. Throughout the experiment, the mesentery was superfused at 2 ml/min with a heated bicarbonate-buffered saline solution (37 C, pH 7.4).
Additionally, small pieces of Saran-Wrap (previously soaked in SUBSTITUTE SHEET (RULE 26) alcohol and rinsed in saline) were placed over the ileum and mesenteric areas peripheral to the viewed area, in order to keep these areas moist. The microcirculatory beds were observed through an intravital microscope (Mikron Instruments, San Diego) equipped with lOx oculars and a 20x objective (water immersion, numerical aperture 0.4). A high-resolution video system, consisting of a video camera (Hitachi color CCD) clamped onto the microscope, a video timer (American Video Equipment), a VCR (Sony SVO-9500MD) and a monitor (Sony Trinitrori), permitted video recordings of the microscope image. In each mesentery preparation, 3 postcapillary venular segments of approximately 180 m length were selected for repeated recordings according to the following criteria: (1) A
true postcapillary venule receiving blood from confluent capillaries, diameter 15-30 m; (2) Brisk blood flow, so that individual red blood cells could not be distinguished;
(3) Some baseline rolling of leukocytes (less than 8 leukocytes present at any given time in the vascular segment), but no firmly adhering leukocytes. A leukocyte was defined as being firmly adhering if it stayed immobile at the vessel wall for more than 30 seconds.

b) FxDerimental Drotnrrnl One-minute video recordings of the baseline leukocyte interaction in the selected venules were made at 20 and 25 min after completion of the surgical procedure. All treatments were administered intravenously before the second baseline recording. At 30 min after surgery, infusion of compound 48/80 (Sigma, catalog #C4257) into the superfusion buffer was begun, and continued throughout the rest of the experiment. This yielded a final concentration of 10 g/ml at the mesentery, and was done in order to induce mast cell degranulation. Venules were recorded again at 20 and 30 min after the beginning of 48/80 application. The extent of leukocyte-endothelial cell interactions was determined at the end of the experiment by analyzing the recorded video tapes.
The number of clearly visible leukocytes (i.e., cells interacting with the vessel wall) was determined from frames * Trademark SUBMTM SHEET (RULE 26) frozen at exactly 0, 15, 30, 45, and 60 seconds of the recorded sequence, and the mean from these five determinations was calculated. Hence, counted leukocytes included rolling cells as well as those firmly adherent.
c) Statistics Each experimental group conta,ilned 5-6 animals with 2-3 venules from each animal. The mean and its standard error is given. Significant differences among group means were tested through analysis of variance followed by the Tukey-Kramer's HSD test. Differences before and after treatments in individual venules were studied by paired t-tests. All tests were run using JMP software running on a Macintosh IIsi, and significance was accepted at p<0.05.
II. Results Topical application of compound 48/80 to the rat mesentery resulted in visible degranulation of more than 90%
of the mast cells, which took place within 3 min after application. Subsequently, an increase in leukocyte rolling and adhesion was seen. No significant difference in leukocyte accumulation was detected between the 20 and 30 minute timepoints post 48/80 application,'the mean of these two timepoints was therefore used in the statistical analysis.
The venule-to venule variability in the model was quite high, with baseline interaction ranging from 0 to 4.8 leukocytes present in individual venules, and mast cell-induced interaction ranging from 2.8 to 25.6 leukocytes present in the PBS-treated group (coefficient of variability, CV, 37%).
Animal means showed much lower variability (CV, 16%); hence, each individual venule was treated as a separate experimental entity for statistical purposes.
As seen in Fig. 29, intravenous treatment with mu MAb PB1.3 inhibited the mast cell-induced intravascular leukocyte accumulation by 90% when compared with the formulation buffer control group. The non-reactive antibody P6H6 did not have any effect in this model.

SUBSTITUTE SHEET (RULE 26) =~v y~._:vo, YCT/US94104935 Thus, administration of the anti-P-selectin antibody, mu MAb Pbl.3, to rats completely prevents the interaction of leukocytes with the vascular endothelium and their subsequent migration into the tissues induced by application of the mast cell degranulating agent 48/80.
Example 18 Maooing the Eoitome of Mu MAb PB1.3 P-selectin is an integral membrane glycoprotein found in secretory granules of platelets and endothelial cells. After cellular activation, P-selectin is rapidly redistributed to the plasma membrane. The mature molecule is a protein with multiple domains, including a "lectin" region, "EGF" region, nine tandem consensus "C3b-C4b Regulatory Protein" (CRP)repeats, a transmembrane region, and a cytoplasmic domain. See Johnston et al., Ce31, 56:1033-1044 (1989) for the cloning and description of P-selectin.
Fig. 30 shows a proposed folding pattern of the domains of P-selectin.

P-Selectin Deletion Analyses In order to map the epitope specifically bound by mu MAb PB1.3, mutants of recombinant P-selectin were constructed by deleting consecutive CRP domains (Fig. 31). This site-specific mutagenesis was accomplished by PCR amplification using primers specific to each CRP domain (Table 6). The template used for PCR-mutagenesis was a pUC DNA clone of P-selectin which was truncated and mutated at DNA sequence position 2354 (sequence positions are in accordance with the published sequence of P-selectin of Johnston et al., supra.
This template plasmid pGe1,T which was derived initially by in vitro mutagenesis, lacks the trans-membrane domain and the cytoplasmic carboxy-terminus of P-selectin, and encodes instead the carboxy-terminal 11 amino acid residues of SV40 Large T antigen. This SV40-derived C-terminus is recognized by the monoclonal antibody KT-3. (See MacArthur et al., Journal of Virology 52:483-491 (1984.)) After PCR amplification, the C1SRC'TTTI rTt c't.lr-r+ ,==~ ~ ~ +r~

cM6 de'_:ted DNA molecules were purified, ligated, transformed into bac=eria, cloned and sequenced as described below. DNAs with the appropriate linkage of CRP to the T-antigen tag sequence were subcloned into the pcDNAl expression vector (Invitrogen) 5 anc transfected into COS cells.
PCR mutagenesis was performed on plasmid pGMPO1,T.
Af=er phosphorylation at the 5'-end, the primers in Table 6 were added to individual PCR reactions at 1 M, together with the T-antigen sequence primer lys-1. The PCR was performed in 10 a Perkin Elmer-Cetus machine for 30 cycles at 95 C for 50 sec, 500C for 1 min. and 73 C for 4 min. using Pfu polymerase (Stratagene) according to the manufacturer's recommended conditions. After purification by agarose gel electrophoresis and elution by Qiaex (Qiagen Corp.), the mutated pUC DNAs were 15 ligated and transformed into E. coli InvF'a cells (Invitrogen). Cloned DNAs carrying the desired mutation were identified by DNA sequence analysis and the Sall-HpaI
fragments containing the mutated P-selectin gene cloned into the XhoI and XbaI sites of the expression vector, pcDNAI
20 (Invitrogen). The pcDNA clones were transformed into E. coli strain MC1061 (Invitrogen) and purified plasmid DNAs used for transfection into COS.
DEAE-Dextran Transfection Method This method is described in Kriegler, M., Gene 25 Transfer and Expression: A Laboratory Manual, W.H. Freeman and Company (1990). COS cells were seeded at 1 x 106 cells/100 mm dish in DMFM (Biowhittaker), 10t fetal bovine serum (FBS). On day two, plasmid DNA was ethanol precipitated, and resuspended at a concentration of 20 g/mL in sterile TE (10 mM Tris, pH
30 8.0, 1 mM EDTA). 150 L of DNA was mixed with 300 l of sterile TBS (Tris Buffered Saline, 140 mM NaCl, 5 mM KC1, 1.4 mM Na2HPO4, 25 mM Tris-base, pH 7.5, 1.0 mM CaC12, and 0.5 mM
MgC12) and with 300 l of sterile DEAE dextran (Sigma, #D-9885, 1 mg/ml in TBS). The growth media was aspirated, and 35 the cell monolayers were washed once with PBS, and once with TBS. 750 l of the DNA/DEAE dextran/TBS mixture was added to the monolayer. The dish was incubated at ambient temperature inside a laminar flood hood rocking the dish every 5 min for * Trademark SUBSTITUTE SHEET (RULE 26) _.~ .,._ _ ._ a. _ _w........ - -- --- ~. __. .. _. .

1 h. After the 1 h incubation, the DNA solution was aspirated and the cells were washed once with TBS and then once with PBS. The cells were incubated in a complete medium supplemented with 100 M chloroquine (Sigma, OC-6628), 37 C, 5% CO2. After 4 h, the medium was replaced with complete media, and the cells were incubated at 37 C and 5% COZ. After 48 h post-transfectibn, the cells were fed with DMEM growth medium lacking serum. 24 h later the medium was harvested, the cell debris removed by centrifugation at 1500 rpm for 5 min in a tabletop clinical centrifuge.
Capture ELISA to Analyze Secreted P-Selectin Mutant Proteins A 96-well=Costar plate was coated with KT-3 antibody (produced in ascites and purified by Protein G Sepharose, Pharmacia) 100 Al/well at 40 g/ml at 4 C overnight. The plate was washed 3x with DPBS, blocked for 1 h at ambient temperature with 250 i/well of DPBS + 1% BSA. The plate was washed 3x with DPBS, and incubated with COS cell supernatants 100 Al/well for 2 h at ambient temperature. After washing the plate 3x with DPBS the bound P-selectin was detected by adding either 100 l/well of (a) 1 ng/ml rabbit anti-P-selectin polyclonal Ig, or (b) biotinylated mu MAb PB1.3, or (c) biotinylated P6H6 diluted in DPBS/1% BSA for 1 h at ambient temperature. After washing the plate 3x with DPBS the rabbit antibody was detected using 100 Al/well of a 1:1000 dilution of HRP-conjugated goat anti-rabbit antibody (Biorad), blocked with tissue culture media from the KT-3 cell line (10%) in DPBS/1% BSA for 30 min at ambient temperature. The biotinylated antibodies were detected using 100 Al/well of a 1:1000 dilution of HRP-streptavidin (Pierce) for 30 min at ambient temperature. The bound HRP was incubated with TMB and quenched with 1 M phosphoric acid. Figure 32 shows the signal less the background from supernatants of mock transfected COS
cells. The "Mock" column is the supernatant from cells transfected with a full length form of P-selectin, which lacks the SV40 T-antigen tag, and therefore is not detected in this KT-3 capture ELISA. The X-axis specifies the number of CRPs expressed by the various mutant P-selectin cDNAs, from all 9 SUBSTITUTE SHEET (RULE 26) CRPs in column 2 to one CRP in column 10. The results show that, although P-selectin mutants lacking the carboxy-terminal CRPs can be detected by the polyclonal rabb-it Ig, these molecules do not react with PB1.3. This suggests that the 5 binding site for PB1.3 lies within the fifth CRP repeat domain.
In order to confirm the results from epitope mapping, the deletion analysis was extended to delete only the fifth CRP domain, leaving the carboxy-terminal sequences of P-selectin intact (see Fig. 30). PCR amplimers were designed to remove the region of P-selectin encoding amino acids 407 to 468 (Table 6) using the same protocols described above. Cell supernatants were analyzed directly by ELISA with antibodies Mu MAb PB1.3 and P6H6 (Fig. 33). Although mu MAb PB1.3 binds the intact P-selectin, it fails to bind the deleted molecule.
P6H6 binds both molecules equivalently.
Synthetic Peptide Analyses The mu MAb PB1.3 epitope was further characterized by synthesizing peptides spanning the 62 amino acid region of the fifth CRP of P-selectin (Fig. 34) and screening for mu MAb PB1.3 binding by ELISA. The fifth CRP extends from amino acid sequence 407 to 468 according to the nomenclature of Johnston, et al. The peptide sequences are 968.01, a.a.
408-426; 968.02, a.a. #418-436; 968.04, a.a. #408-433; 968.05, an oxidized version of 968.04; 968.06, a.a. #428-447; and 968.08, a.a. #448-467. Each peptide was dissolved to 2mg/ml in dimethyl sulfoxide (DMSO) and coated onto a 96-well microtiter plated by dilution of 1 microgram into 100 l PBS
and dried at 37 C overnight. Coated wells were washed with PBS and blocked with 250 l/well of a solution of PBS/1% BSA
for 30 min at ambient temperature. The wells were washed with PBS and incubated with 100 l/well of PB1.3 diluted to l g/ml in PBS/1%BSA for 1 h at room temperature. Bound mu MAb PB1.3 was detected with HRP-conjugated goat anti-mouse antibody reagent and TMB substrate as described before. Fig. 34 shows a comparison PB1.3 reactivity with the peptides. The carboxy-terminal peptide, number 968.08 reacted strongly with PB1.3.

SUBSTITUTE SHEET (RULE 26) ~ v y. c u0 PCT/US94J04935 Therefore the epitope recognized by PB1.3 maps to amino acids 448 to 467 in the fifth CRP of human P-selectin.
For the purposes of clarity and understanding, the invention has been described in these examples and the above disclosure in some detail. it will be apparent, however, that certain changes and modifications may be practiced within the scope of the appended claims.

SUBSTITUTE. SHEET (RULE 26) b ac~ wr 41 aaaOc7dw>~aa4Uaw >b4>0 uc44vsc7~~w~z~- >~x >aada.t7~c~a N >
--I ~
o N ~ > aa>a~Udca>aa ~a.~~~>~>~Ua~~v~t7~EwE-~n~d~x I~>a~CY4c.C7ar,Ga N
~ L Q~ a,. O'v~UdW>D4~4Ud~n>M >v~u5r3d~nC75E-~F+rn;+4~,v~ >O~dC7C~U-t N
tD VVJ7 ~ 4 o k~ W > aaCY a0 Uawa> 0 ~~>a~~.vac~~d~nC7~E-{wE-+0 >+x >~4CYv~aUa~~~a d W d dacyO~v,Uawa> Wa.Udv,>M >rA u6ddcn c7~HwHz ?+>:~ x I r>RAapdo.UvC7,.4 O b a a ---------- ----------------- ~A-----Ci~ _..._--.... _..-.
~ O.N M'd A O t~ OO Q~ ~'CV N~ CY N~!~7 lV ~i1 tt1 tr1 t~1 A M
t*1 't r1 ~4 CW Q~ =-ti r r .r .-i ...ti .~ -.~ N
l~+1 M M(~Y 4Q Nd~ *Jt "d 0001Q-!V tA.tvt Wh OOt?t "' N c1 Y W'i 'O h .
~ ..a .r ..r - .I " -i ,-4 -q R N('~~tV~NNf1~MC*icMCM~Mt~n(*1tn<<1MM f-1~~~
14rtl RA lvql x~ W~~Or~~za ~ ~xAna~4~zwo4wa4xa>H~Hv~~ AH~~v~ ~+~~~wa~~.-axv~wAN~>~rc~
N 'C7 >
zd 0 xC7z[-4 p4F-qrrio'A4wC7Hca ., a;,:9 wa~nc~~aacv,wAE+d:ar+>+u O {/j O (,V1 N ~, ~~ w~~c~3~za,~ I~zc~aHZ~aaoawac~asyH~Hd A~~H=~ ~.d~~wa~~~a~~waH~c~;~~c>
0) r-i ~ ~j~ t7 W~~..C7A.+xa. I(C7~OC7E=rr~-~'.~0~~4t~~4C9Dd<E-~r=lE-~A
AD4v~~nr.~ E.=~C,~;~O'viE-wAto Qr- > t,) N
O

~~ wa~c~v,~za I~zAC~a~~zwawt~zt~eHaH~ A~4~~~ >>wu ---~~-----------------~~_____ ________-----___ ___ ~
I"n~ ~~~r-~~4000000~~ ~ -~ ~; a r -1 tV ~ N N M d ~p ~+.pp(T Q=-~ e} y'~ ~p (~.ppQ~ O=-r CV M et Vy ~(7 0o Q~ p .-r tv N N N i+1 h t- ppQ~ r t!
VSVIYIV~ Vl V1v1v~~ryY'fYyfl'r Vy VSt(}~O WN~~o~O~O~O~O~D N(~ f' i'Cf' h[- t'-00000000 QU dGe0O600ObQO6d60 ~CrnCri TABLE 1-continued ALIGNMENT OF AMiN4 ACID SEQUENCES LEADING TO THE DESlu-Id OF
RBSHAPFD HUMAN PB1.3 HEAVY CHATN VARIAF3LE REGION

FR or u-otse mouse human human RH Vo, Kabat # CDR 1'747 II A I 21128'CI. 1747 Camm.e_t 93 97 1 fi A A A A
94 98 FR3 R R R R Rs 96 100 t R X P - R

100 104 i D S S - D
--- --LOQD I - X C -- -IOfE I - A Y - --IOQH I - Y D G --140J 106 i Y A - G Y

101 108 t D D D N D

103 110 FR4 W W w w w 105 112 ! A Q Q Q Q

109 116 I v v v v v 11l 118 I V V V V V

>a FUp4d~"nCYx I>4~+x~> f3>4 0CYDA0 Wda~d A'~CY~E-~Cyrnwv~E~,-~cndv~?t7Ac4?P1 N

a.,a~va(n ww F a0 0 >t~Aae>F0.4 HU aFw ao I ,tn a~.aa a~aaaawc~w~~~
?
o N
N
m se Ar.a~gFoO aaw 0 dw?L7AoC?E-,... Uadv~av~a> I i...cnC7'apAt~O~4- wp~
N 00 a o k ~v~dcn,aC7AaG> F~.. uad~n(v A I(( z?.,z~~+aaaaoa~a U

H A">~Fawaww F~aad~az I I I i I ydFZ>;> 3~a'aocwa aw w~a ___------__ _____------ -~~--------------- -----__--~Ik vv'I~Oh- 0041 = ~ -4 ~=-1 ~ ~=-~=~-~~ =~=a9 N NANN~~lV fYt'~ tN+)MM M~
MMfh'~ =t ml,'7 P4 ~, l'1 M~d' v] ~O 1'W O~ C,) =~r =-N~ =fn4 .~4-7 ~ ti~~~V N tV fY ~ M N nl f'7 N~ N~ fV ~ M c~1 cN*1 M 't VN" "~#~' ~ V~

TABLE 2-continued ALIGNMF.NT OF AMINO ACII) SEQUENCES LEADING TO THE DESIGN OF
RESHAPED PB 1.3 I1GHT CHAIN VARIABI.E REGION

FR or mouse mouse human human RH VL
Kabat #1 CDR 1747 xV ic-I DEN 1747 Conm:tnt 46 46 i A L L L A mtty be 8i4partin8 L2, L may sher ft cxeientation of 48 48 1 I I I I I*

50 50 CDR2 T Y A G T"
51 51 I A A A A A*
52 52 1 S S S S S"' 59 59 I P P P P p 60 60 I E 5 S S E ckose io residue 54 in L2,anthe auzf9ce avd m9 birid A8 64 64 1 G G G G Gy' 6t) 69 1 T T T T T
70 70 1 D D D E D may havt clma iateraction with rtslduc 24inL1.
71 71 I F Y F F F*

tc 80 1 S Q P S S
81 8i 1 E E E D D

83 83 ! L I F F F

TABLE 2-continued ALIGNiVIENT OF AhflN0 ACID SEQUENCES IEADING TO T1B; DESIGN OF
RESHAPED PB 13 LIGHT CIiAIN VARIABLE REGION

FR or IIIOliSC IIIOUSfl l1llll]8II }3uIBai1 RH VL
Kabat # CDR 1747 tcV K-I DEN 1747 CowmerA

86 86 1 Y Y Y Y y 90 90 t Q Q Q Q Q*
91 91 I Y G Y Y Y*
92 92 I N N N DN N*
93 93 l N T S S N*
94 94 1 Y L L F Y*
95 95 I P P P P P*

96 96 I Y R w y Yo 99 99 ! C3 G 0 G a 106A. I - - - -- ---i47 107 FR4 Q K K K K

DIFFERENCES IN AMlNO ACID SEQUENCES FOR THE
HEAVY CHAIN VARIABLE DOMAINS FOR THE
THREE RESHAPED VERSIONS OF P81.3 CONSTRUCT AMINO ACID

1748RHc GLU ALA LYS

DIFFERENCES IN AMINO ACID SEQUENCES FOR THE
UGHT CHAIN VARIABLE DOMAINS FOR THE
FOUR RESHAPED VERSIONS OF P81.3 CONSTRUCT AMINO ACID

1748RLC ~ GLU GLU

ANTIBODY PRODUCTION RATES FOR AMPLIFIED CHO CELL UNES
PRODUCING THREE DIFFERENT VERSIONS OF HUMANIZED PB1.3 (CY1748) (micrograms1106 ceils/day) Antibody Unamlified 1st round 2nd round 3rd round 1748A-IgGl 0.36 2.1 8.2 29.6 0.36 2.1 13.7 21.8 0.36 0.90 9.9 19.7 17488-igG1 0.10 1.6 23.9 0.24 1.5 21.1 < 0.2 7.3 18.6 0.10 1.6 17.8 1748B-IgG4 < 0.3 < 2.0 33.0 < 0.3 < 2.0 26.9 0.75 2.6 23.2 < 0.3 3.0 20.3 0.80 6.5 20.8 < 0.3 < 2.0 1 9 . 4 0.'$ 2.5 1 8.6 0.78 8.8 17.8 TABLE fi DNA SEQUENCE OF THE PCR AMPLIMERS USED TO GENERATE
HUMAN P-SELECTIN CRP DELETION MUTANTS

PCR 3'-amplimer to Predicted amino rem4Ye CRP Number: acid seauence of mutant CRP TAg 9 5'-TTTCACAGCTCTGCATGC-3' ...CRAVK TIQEA LK...
8- 9 5'-TGCTATGCCTTTGCAGGT-3' ...CKGIA T1QEA LK...
7-9 5'-CTTGATGGCTTCACACAT-3' ...CEAIK TIQEA LK...
6-9 5'-GGGAATGGCTTGGCATTC-3' ...CQAIP TIQEA LK...
5-9 5'-CTGCAAAGCTTGACAGAC-3' ...CQALO TIQEA LK...
4- 9 5'-CGAAATAGCCTCACAGGT-3' ...CEAIS TIQEA LK...
3- 9 5'-CTGCACAGCTTTACACAC-3' ...CKAVQ TIQEA LK...
2-9 5'-CTGGGCAGCTAAACACTG-3' ...CLAAQ TIQEA LK...
5'-amplimer, lys-1 5'-AAACCTCCCACACCCCCT-3' To Delete onty CRP#5 these two ampiirners were used:

a i'>i~ii rii c r ..~~'- T Gr A C .. ~ ~p-TTT G CT R n G r~ - ~
~ ' 3'-?motimer 5'-CTGCAAAGCT" GA C~. GAC- ~'

Claims (31)

CLAIMS:

1. A humanized immunoglobulin comprising a humanized heavy chain and a humanized light chain;

(1) the humanized light chain comprising three complementarity determining regions (CDR1, CDR2 and CDR3 having amino acid sequences from the corresponding complementarity determining region of a mouse PB1.3 immunoglobulin light chain, produced by the cell line designated ATCC Accession No. HB11041, and a variable region framework from a human kappa light chain variable region framework sequence; and (2)the humanized heavy chain comprising three complementarity determining regions (CDR1, CDR2 and CDR3 having amino acid sequences from the corresponding complementarity determining regions of a mouse PB1.3 immunoglobulin heavy chain, produced by the cell line designated ATCC Accession No. HB11041, and a variable region framework from a human heavy chain variable region framework sequence;
wherein the humanized immunoglobulin specifically binds to P-selectin, and wherein:

the human kappa light chain variable region framework sequence is substituted at positions L21 and L46 by an amino acid present in the equivalent position of the mouse PB1.3 immunoglobulin light chain variable region framework sequence; and the human heavy chain variable region framework sequence is substituted a position H71 by an amino acid present in the equivalent position of the mouse PB1.3 heavy chain variable region framework sequence.

2. The humanized immunoglobulin of claim 1 wherein the humanized light chain variable region is further substituted at position L60 or L70 by an amino acid present in the equivalent position of the mouse PB1.3 light chain variable region framework sequence.

3. The humanized immunoglobulin of claim 1 or claim 2 wherein the humanized heavy chain variable region is further substituted at position H1, H2, H73 by an amino acid present in the equivalent position of the mouse PB1.3 heavy chain variable region framework sequence.

4. The humanized immunoglobulin of any one of claims 1 to 3 wherein the humanized heavy chain variable region framework is a 21/28'CL heavy chain variable region framework sequence.

5. The humanized immunoglobulin of claim 4, wherein the humanized light chain variable region framework is a DEN
light chain variable region framework sequence.

6. The humanized immunoglobulin of any one of the preceding claims, wherein the humanized light chain comprises the 1748RLA mature light chain variable region amino acid sequence as shown in SEQ ID NO. 13.

7. The humanized immunoglobulin of any one of claims 1 to wherein the humanized light chain comprises the 1748RLB
mature light chain variable region amino acid sequence as shown in SEQ ID NO. 15.

8. The humanized immunoglobulin of any one of claims 1 to 5, wherein the humanized light chain comprises the 1748RLC mature light chain variable region amino acid sequence as shown in SEQ ID NO. 17.

9. The humanized immunoglobulin of any one of claims 1 to 5, wherein the humanized light chain comprises the 1748RLD mature light chain variable region amino acid sequence as shown in SEQ ID NO. 19.

10. The humanized immunoglobulin of any one of claims 1 to 5, wherein the humanized heavy chain comprises the 1748RHA mature heavy chain variable region amino acid sequence as shown in SEQ ID NO. 7.

11. The humanized immunoglobulin of any one of claims 1 to 5, wherein the humanized heavy chain comprises the 1748RHB heavy chain variable region amino acid sequence as shown in SEQ ID NO. 9.

12. The humanized immunoglobulin of any one of claims 1 to 5, wherein the humanized heavy chain comprises the 1748RHC mature heavy chain variable region amino acid sequence as shown in SEQ ID NO. 11.

13. The humanized immunoglobulin of claim 9, wherein the humanized heavy chain comprises the 1748RHB mature heavy chain variable region amino acid sequence as shown in SEQ
ID NO. 9.

14. The humanized immunoglobulin of any one of claims 1 - 13 that is a Fab, Fab', F(ab')2, Fabc or Fv fragment.

15. The humanized immunoglobulin of claim 13 further comprising a constant region domain.

16. The humanized immunoglobulin of claim 15, wherein the constant region domain has an effector function.

17. The humanized immunoglobulin of claim 15, wherein the constant region domain lacks an effector function.

18. The humanized immunoglobulin of claim 16, wherein the effector function is capable of complement fixation or antibody dependent cellular toxicity.

19. A nucleic acid encoding the heavy chain of the humanized immunoglobulin of any one of claims 1 - 18.

20. A nucleic acid encoding the light chain of the humanized immunoglobulin of any one of claims 1 - 18.

21. A pharmaceutical composition comprising the humanized antibody of claim 13, and a pharmaceutically acceptable carrier therefor.

22. A pharmaceutical composition comprising the humanized immunoglobulin of any one of claims 1 to 18, and a pharmaceutically acceptable carrier therefor.

23. The pharmaceutical composition of claim 22, wherein the humanized immunoglobulin is a Fab fragment.

24. An in vitro method for detecting P-selectin, the method comprising:
administering the humanized immunoglobulin of claim 13 to a tissue sample from a patient; and detecting complexes formed by specific binding between the immunoglobulin and P-selectin present in the target sample.

25. Use of a humanized immunoglobulin of claim 13 in the preparation of a diagnostic reagent for detecting complexes formed by specific binding between the immunoglobulin and P-selectin.

26. Use of a humanized immunoglobulin of any one of claims 1 to 18 in the preparation of a medicament for the treatment of thrombosis or ischemia-reperfusion injury.

27. The use of claim 26, wherein the medicament is for the treatment of epidermal, myocardial, renal, cerebral, splenic, hepatic, spinal, splanchnic, pulmonary, partial-body or whole-body ischemia.

28. A stable cell line comprising:

a nucleic acid segment encoding the heavy chain of the humanized immunoglobulin of any one of claims 1 to 18, the segment operably linked to a promoter to allow expression of the heavy chain;
a second nucleic acid segment encoding the light chain of the humanized immunoglobulin of any one of claims 1 to 18, the second segment operably linked to a second promoter to allow expression of the light chain;
wherein the stable cell line is capable of producing the humanized immunoglobulin of any one of claims 1 to 18.

29. A stable cell line comprising:

a nucleic acid segment encoding the heavy chain of the humanized immunoglobulin of claim 13, the segment operably linked to a promoter to allow expression of the heavy chain;
a second nucleic acid segment encoding the light chain of the humanized immunoglobulin of claim 13, the second segment operably linked to a second promoter to allow expression of the light chain;
wherein the stable cell line is capable of producing the humanized immunoglobulin of claim 13.

30. Use of a humanized immunoglobulin of any one of claims 1 to 18 for the treatment of thrombosis or ischemia-reperfusion injury.

31. The use of claim 30, wherein the ischemia-reperfusion injury is epidermal, myocardial, renal, cerebral, splenic, hepatic, spinal, splanchnic, pulmonary, partial-body or whole-body ischemia.