WO1995034324A1

WO1995034324A1 - Cross-reacting monoclonal antibodies specific for e-selectin and p-selectin

Info

Publication number: WO1995034324A1
Application number: PCT/US1995/007302
Authority: WO
Inventors: Ellen L. Berg
Original assignee: Protein Design Labs, Inc.
Priority date: 1994-06-14
Filing date: 1995-06-07
Publication date: 1995-12-21
Also published as: ATE248605T1; US5622701A; US6210670B1; JPH10502168A; EP0765172A1; EP0765172A4; CA2191870A1; AU2821195A; EP0765172B1; HK1016018A1; DE69531679D1; DE69531679T2

Abstract

The invention provides monoclonal antibodies that specifically bind to P-selectin and to E-selectin. Humanized monoclonal antibodies also are disclosed. Many of the antibodies block the functional interactions of P-selectin and E-selectin with their respective counterreceptors.

Description

CROSS-REACTING MONOCLONAL ANTIBODIES SPECIFIC FOR E-SELECTIN AND P-SELECTIN

BACKGROUND OF THE 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 vital to the interaction of leukocytes and platelets with themselves and with the extracellular matrix and vascular endothelium.

The selectin family of receptors are so named because of their lectin-like domain and the selective nature of their adhesive functions. There are three known selectins, L-selectin (also known as LECAM-1, Mel-14 or LAM-1 or CD62L) , E-selectin (also called ELAM-1 or CD62E) and P-selectin (also known as CD62, CD62P, GMP140 or PADGEM) . The selectins are highly homologous, containing a 120 amino acid (aa) Ν-terminal lectin domain, an EGF-like domain, a variable number of multiple short consensus repeat (SCR) domains homologous to those found in complement regulatory proteins, followed by a transmembrane domain and short cytoplasmic tail. See Siegelman et al.. Science 243:1165-1172 (1989); Lasky et al.. Cell 56:1045-1055 (1989); Tedder et al., J. Exp. Med. 170:123- 133 (1989); Johnson et al.. Cell 56:1033-1044 (1989); Bevilacgua et al., Proc. Natl . Acad. Sci . USA 84:9238-9242 (1987), Bevilacgua et al. , Science 243:1160-1165 (1989), Bevilacgua et al., J. Clin . Invest. 91:379-387 (1993), Camerini et al.. Nature 280:496-498 (1989). The selectins have overlapping but distinct specificities for counterreceptors. See Bevilacgua et al., J. Clin . Invest . 91:379-387 (1993); Feize, Current Opinion in Struct . Biol . 3:701-710 (1993); Berg et al., Biochem. Biophys . Res . Comm. 184:1048-1055 (1992); Foxall et al. , J. Cell Biol . 117:895-902

(1992); Larsen et al., J. Biol . Chem. 267:11104-11110 (1992); Polley et al., Proc. Natl . Acad. Sci . USA 88:6224-6228 (1991) (each of which is incorporated by reference in its entirety for all purposes) . P-selectin is constitutively expressed by both platelets and endothelial cells where it is stored in α-granules or Weibel-Palade bodies for rapid (seconds to minutes) translocation to the cell surface upon activation by, for example, thrombin or histamine (McEver et al., J. Biol . Chem. 250:9799-9804 (1984); Hsu-Lin et al., J. Biol . Chem.

264:8121-9126 (1984)). E-selectin is expressed by activated endothelial cells (e. g. , after TNF-α or IL-1 stimulation for 6-8 hr) . Its expression is controlled at the transcriptional level (Bevilacqua et al., 1987, supra; Bevilacgua et al., 1989, supra). P-selectin and E-selectin both bind to neutrophils and monocytes (Larsen et al.. Cell 59:305-312 (1989); Johnston et al.. Cell 56:1033-1044 (1989); Bevilacgua^' et al., 1987, supra; Bevilacgua et al., 1989, supra) , as well as subsets of lymphocytes (Picker et al. , Nature 349:796-799 (1991); Shimizu et al.. Nature 349:799-802 (1991); Moore et al., BBRC 186:173-181 (1992)). L-selectin is constitutively expressed by leukocytes, and mediates lymphocyte adhesion to peripheral lymph node high endothelial venules (HEV) (Gallatin et al.. Nature 304:30-34 (1983); Berg et al., Immunol . Rev. 108:5-18 (1989); Berg et al., J. Cell . Biol . 114:343-349 (1991)), and neutrophil adhesion to cytokine-activated endothelial cells (Hallman et al., Biochem. Biophys. Res . Comm. 174:236-243 (1991); Smith et al., J. Clin. Invest. 87:609-618 (1991); Spertini et al., J. Immunol . 147:2565-2573 (1991)). L-selectin is a counter-receptor on neutrophils for both E-selectin and P-selectin (Kishi oto et al.. Blood 78:805-811 (1990), Picker et al.. Cell 66:921 (1991)), although all three selectins probably have other counter¬ receptors as well.

E-selectin, P-selectin and L-selectin mediate leukocyte-endothelial cell and platelet-leukocyte adhesive interactions during inflammation (Bevilacgua et al., 1993, supra) . All three selectins have been demonstrated to participate in an initial "rolling" interaction of leukocytes with activated endothelium (von Andrian et al., Proc. Natl . Acad. Sci . USA 88:7538-7542 (1991); Ley et al. , Blood 77:2553- 2555 (1991); Abassi et al., J. Clin . Invest. 92:2719-2730

(1993); Dore et al.. Blood 82:1308-1316 (1993); Jones et al., Biophys . J. 65:1560-1569 (1993); Mayadas et al.. Cell 74:541- 554 (1993)). This initial interaction precedes CD18-integrin- mediated adhesion and subseguent migration of neutrophils through the endothelium and into inflamed tissue sites

(Lawrence et al.. Cell 65:859-873 (1991); von Andrian et al.. Am. J. Physiol . 263:H1034-H1044 (1992)). Depending on the nature of inflammatory stimuli and time after initiation of inflammatory response, either E-selectin or P-selectin may be functionally dominant in promoting neutrophil-mediated tissue damage.

In principle, antibodies or other antagonists of the selectins could abort the adhesion process, thereby preventing neutrophils from binding to the endothelium and from extravasating into tissues. A substantial number of antibodies specific for one of the selectins have been reported. Some of these antibodies have been reported to block binding of selectins to counterreceptors in vitro. Some of the antibodies have also been reported to block selectin- mediated interactions in animal models in vivo. For example, antibodies to E-selectin have been reported to protect against neutrophil-mediated damage in an IgG complex model of lung injury in the rat (Mulligan et al., J. Clin. Invest. 88:1396 (1991)). Antibodies to P-selectin have been reported to protect against acute lung injury induced by intravenous injection of cobra venom factor (Mulligan et al., J. Clin. Invest. 90:1600-1607 (1992)), as well as in a rat model of systemic endotoxemia (Coughlan et al., J. Exp. Med. 179:329- 334 (1994)). Antibodies to P-selectin have also been reported to be protective in a cat model of myocardial ischemia and reperfusion injury (Weyrich et al., FASEB J. 7:A785 (1993)). Although some antibodies against E-selectin and P-selectin have shown blocking activity, many, if not most, antibodies specific for E-selectin or P-selectin are nonblocking (see, e. g. , Bevilacgua et al., 1989, supra; Erbe et al., J. Cell Biol . 119:215-227 (1992)). That is, these antibodies bind to epitopes in the extracellular domains of E-selectin or P-selectin that do not directly participate in counterreceptor binding or the subsequent cellular adhesion process. The prevalence of nonblocking antibodies suggests that only small regions of the extracellular domain participate directly in binding or influence binding. Thus, de novo screening of antibodies generated against E-selectin or P-selectin would be expected to generate mainly nonblocking antibodies.

Despite the large number of antibodies isolated to- date against the three selectins, there have been few reports of crossreacting antibodies that bind to more than one selectin. Crossreacting antibodies might be capable of aborting the inflammatory process at more than one level, thereby providing more broadly useful therapeutic agents for neutrophil-mediated inflammatory conditions than antibodies specific for a single selectin. One antibody has been reported to crossreact with human E-selectin and dog L-selectin but not with the two selectins from the same species (Abassi et al., J^*. Immunol . 147:2107-2115 (1991)). A second antibody has been reported to crossreact with human E-selectin and L-selectins (Jutila et al., J. Exp. Med.

175:1565-1573 (1992); WO/9324614). However, no antibody has been isolated that binds to both P-selectin and E-selectin, much less blocks the functions of both of these molecules. Accordingly, there is a need for antibodies that bind to both E-selectin and P-selectin, preferably so as to block the capacity of both of these molecules to participate in adhesion reactions with counterreceptors. The present invention fulfills this and other needs. SUMMARY OF THE INVENTION The invention provides monoclonal antibodies that have a binding site that specifically binds to P-selectin and to E-selectin. For many such antibodies, specific binding of the antibody to the P-selectin inhibits binding of the

P-selectin to a counterreceptor of P-selectin, and specific binding of the antibody to E-selectin inhibits binding of the E-selectin to a counterreceptor of E-selectin. Counterreceptors of E-selectin and P-selectin are expressed on the surface of cells such as HL-60 cells and neutrophils.

Exemplary antibodies are designated 57C.29, 2C9.11 and 1D8.10. Many of the antibodies of the invention compete with an exemplified antibody for specific binding to P-selectin and to E-selectin. Some antibodies of the invention also specifically bind to L-selectin, whereas others do not. In one embodiment the antibody recognizes an epitope of E-selectin comprising amino acids Q₂₁ ^R ₂₂' ^γ ₂₃' ^τιi₉' ^{an< A}i₂₀* In another embodiment, the antibodies bind to the same epitope of E-selectin and/or P-selectin as antibody 5C7.29. In addition to intact antibodies, the invention also provides binding fragments such as Fab, Fab', F(ab')₂, Fv or single- chain antibodies.

Some of the antibodies of the invention are non- human, e.g., mouse, whereas others are humanized or human antibodies. A humanized antibody comprises a humanized heavy chain variable region and a humanized light chain variable region. The humanized light chain variable region can comprise complementarity determining regions (e.g., CDR1, CDR2, CDR3) having amino acid seguences from the light chain of a mouse, antibody selected from the group consisting of 5C7.29, 2C9.11 and 1D8.10, and having a variable region framework sequence substantially identical to a human light chain variable region framework sequence. The humanized heavy chain variable region can comprise complementarity determining regions (e.g., CDRl, CDR2 and CDR3) having amino acid sequences from the corresponding mouse antibody heavy chain, and having a variable region framework sequence substantially identical to a human heavy chain variable region framework seguence. The antibodies optionally contain constant regions substantially identical to human constant regions.

In particular embodiments of the humanized antibodies of this invention, the humanized light chain variable region has a sequence substantially identical to the mature sequence depicted in Figure 8A [SEQ ID NO:5] and the humanized heavy chain variable region has a seguence substantially identical to the mature sequences depicted in Figure 8B [SEQ ID NO:8]. More particularly, this invention provides humanized antibodies wherein (a) the humanized light chain variable region has the sequence: X₁IX₂X₃TQSPSS LSASVGDRVT ITCSASSSX_1:LP YX₁₂HWYQQKPG KAPKLLIYDT SNX₁₃X₁₄X₁₅GVPX₄R X₇SGSGSGTX₅X₆ TX₈TISSLQPE DX₉ATYYCX₁₆X₁₇W SSDPFTFGX₁₀G TKVEIK [SEQ ID NO:9], wherein X-_L = D or Q; X₂ = Q or V; X₃ = M or L; X₄ = S or A; X₅ = S or D; X₆ = Y or F; X₇ = F or I; X₈ = L or F; X₉ = F, I or A; X₁₀ = Q, G or S; X _l = V, I or L; X₁₂ = M or L; X₁₃ = any amino acid; X₁₄ = any amino acid; X₁₅ = S or T; X₁₆ = Q, N or H; and X₁₇ = Q, N or H; and (b) the humanized heavy chain variable region has the sequence: X₃VQLVESGGG LVQPGGSLRL SCAASGFTFS SFGX₇HWVRQA PGKGLEWVX^ ISSGSSTIYY X₈X₉X₁₀X₁₁X₁₂X₁₃RFTI SRDNX₄KNX₅LY LQMX₂SLRAED TAVYYCARPL PPFAYWGQGT LVTVSX_g [SEQ ID NO:10]; wherein, X_x = A or S; X₂ = N or T; X₃ = £, Q or D; X₄ = S, A or P; X₅ = T or S; X₆ = A or S; X₇ = M, I, V or L; X₈ = any amino acid; X₉ = any amino acid; X₁₀ = any amino acid; X_l = V, A, I, L, M or F; X₁₂ = R, K or Q; and X₁₃ = G, A, D, T or S. In certain embodiments of the aforementioned antibodies, the CDR regions of the light and heavy chain variable regions have the same amino acid sequence as the CDR sequences of Figure 8A and 8B. That is, in the human light chain variable region,

X_1X = V; X₁₂ = M; X₁₃ = L; X₁₄ = A; X₁₅ = S; X₁₆ = Q; and X₁₇ = Q; and in the heavy chain variable region, X₇ = M; X₈ = A; X₉ = D; X₁₀ = T; X_1X = V; X₁₂ = R; and X₁₃ = G. In another embodiment, the variable light and heavy chain regions have the amino acid seguence depicted in Figures 8A and 8B.

In another aspect, the invention provides purified nucleic acid segments encoding a light or heavy chain variable region of one of the monoclonal antibodies discussed above. The invention also provides stable cell lines capable of producing the antibodies described above. The stable cell lines comprise nucleic acid segments respectively encoding the heavy chain and light chain of an antibody described above. The segments are operably linked to first and second promoters to allow expression of the heavy and light chains.

The invention further provides pharmaceutical compositions comprising the antibodies described above and methods of treatment using the same. The methods of treatment are particularly effective for inflammatory diseases including conditions such as ischemia-reperfusion injury, adult respiratory distress syndrome, sepsis, psoriasis and autoimmune disease. In another aspect, the invention provides methods of generating an antibody capable of blocking E-selectin and/or P-selectin mediated functions. The method comprises concurrently or consecutively immunizing a mammal with P-selectin and E-selectin. B-cells from the mammal are immortalized to generate immortalized cells producing antibodies. An immortalized cell is selected producing an antibody that specifically binds to E-selectin and to P-selectin.

The invention further provides methods of detecting E-selectin and P-selectin bearing cells in a biological sample suspected of containing the cells. The method comprises contacting the sample with an antibody as described above to form an immune complex with the E-selectin and/or P-selectin bearing cells. The presence of the immune complex is then detected to indicate the presence of the cells.

BRIEF DESCRIPTION OF THE DRAWINGS Figures 1A and IB: Crossreacting antibody 5C7.29 binds to naturally occurring human E-selectin. (a) Binding of known anti-E-selectin antibody H18/7 to activated (black histograms) and resting (grey histograms) HUVEC cells. (b) Binding of crossreacting antibody 5C7.29 to activated and resting HUVEC cells. FACS fluorescence intensity is indicated by the X axis.

Figures 2A and 2B: Crossreacting antibody 5C7.29 binds to naturally occurring P-selectin. (a) Binding of known anti-P-selectin antibody WAPS 12.2 to platelets detected by staining with secondary antibody (black histogram) , compared with staining with secondary antibody alone (control, grey histogram). (b) Binding of 5C7.29 to platelets, shown similarly. Figure 3: Crossreactivity of 5C7.29 resides in a single monoclonal antibody. 5C7.29 antibody was incubated with excess of (a, c) parent Ll-2 cells or (b, d)

Ll-2^{p_selec ιn} transfeetants, and resulting supernatants tested for reactivity with fresh samples of l-2^p"selectιn (a, b) or _L1_.₂ ^E-^selectin cells (c, d) by FACS analysis. This figure shows that Ll-2^p~selectin depletes reactivity for E-selectin.

Figure 4: Monoclonal antibody 5C7.29 blocks binding of HL-60 (neutrophil-like) cells to TNF-α-activated HUVEC cells (expressing E-selectin) . Average of four experiments. Figure 5. Monoclonal antibody 5C7.29 blocks binding of HL-60 cells to E-selectin transfectant cells. Average of four experiments.

Figure 6. Monoclonal antibodies 5C7.29, 2C9.11 and

1D8.10 block binding of platelets to HL-60 cells as shown by platelet rosetting. The chart shows the percentage of HL-60 cells with > 2 platelets bound (rosetted) . Average of three experiments.

Figures 7A-7B. Seguences of the cDNA (light chain

— SEQ ID NO:l; heavy chain — SEQ ID NO:3) and translated amino acid sequences (light chain — SEQ ID NO:2; heavy chain

— SEQ ID NO:4) of the light chain (A) and heavy chain (B) variable regions of the mouse 5C7.29 antibody. The first amino acid of each mature chain is indicated by a double underline. The three CDRs in each chain are underlined. Figures 8A-8B. Sequences of the synthetic DNA

(light chain — SEQ ID NO:5; heavy chain — SEQ ID NO:7) and translated amino acid seguences (light chain — SEQ ID NO:6; heavy chain — SEQ ID NO:8) of the light chain (A) and heavy chain (B) variable regions of the humanized 5C7.29 antibody. The first amino acid of each mature chain is indicated by a double underline. The three CDRs in each chain are underlined. Figure 9. Schematic diagram of construction of humanized 5C7.29 antibody variable region genes.

Figure 10. Humanized 5C7.29 antibody reactivity with E-selectin, P-selectin and L-selectin transfectants. Ll-2 transfectant cell lines expressing the indicated selectin were analyzed for reactivity with humanized 5C7.29 by flow cytometry.

Figures 11A and 11B. Competitive binding of mouse and humanized 5C7.29 antibodies to cells expressing E-selectin (A) or P-selectin (B) . Increasing concentrations of cold competitor antibody were incubated with the cells in the presence of radiolabeled tracer mouse 5C7.29 antibody, and the ratio of bound/free radioactivity was determined.

Figure 12. Inhibition of HL-60 cell adhesion to _CH0 ^E-s^eiec^tin _{cells by aouse} and humanized 5_C7.29 antibodies. Fluorescently labelled HL-60 cells were incubated with

_CH0 ^E-se^lec^tin _cells i_n the presence of the antibodies at the indicated concentrations. After washing, adherent cells were counted microscopically. The results from a representative experiment performed with each sample in quadruplicate (+/- standard deviation) are shown.

Figure 13. Inhibition of platelet rosetting to H^*L-60 cells by mouse and humanized 5C7.29 antibodies. Normal human platelets were incubated with HL-60 cells in the presence of the antibodies at the indicated concentrations. After fixation, the percent of HL-60 cells with greater than 2 platelets bound (rosetted) was determined. The results shown are from a representative experiment performed with each sample in triplicate (+/-standard deviation) .

DEFINITIONS

The term "substantial identity" or "substantial homology" 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 seguence identity, preferably 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 5 not identical differ by conservative amino acid substitutions. The term "substantially pure" or "isolated" 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

10. 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 by weight of all macromolecular species present in the

15 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.

20 "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. Binding fragments are produced by recombinant DNA technigues, or by enzymatic or chemical

25 cleavage of intact immunoglobulins. Binding fragments include Fab, Fab', F(ab')₂, Fv and single-chain antibodies. An antibody other than a "bispecific" or "bifunctional" antibody is understood to have each of its binding sites identical. An antibody substantially inhibits adhesion of a

30 receptor to a counterreceptor when an excess of antibody reduces the quantity of receptor bound to counterreceptor by at least about 20%, 40%, 60% or 80%, and more usually greater than about 85% (as measured in an in vitro competitive binding assay) .

35 The term epitope includes any protein determinant capable of specific binding to an immunoglobulin or T-cell receptor. Epitopic determinants usually consist of chemically active surface groupings of molecules such as amino acids or sugar side chains and usually have specific three dimensional structural characteristics, as well as specific charge characteristics.

An antibody is said to specifically bind an antigen when the dissociation constant is ≤ 1 μM, preferably ≤ 100 nM and most preferably s 10 nM.

The term patient includes human and veterinary subjects.

The term P-selectin counterreceptor denotes a protein other than an antibody that specifically binds to

P-selectin at least in part by noncovalent bonds. Specific binding maintains cells respectively bearing receptor and counterreceptor in physical proximity and may also transduce a change in physical or functional phenotype in either of the cells or both. Other selectin counterreceptors are analogously defined.

DESCRIPTION OF THE PREFERRED EMBODIMENT I. Antibodies of the Invention The invention provides antibodies that crossreact, i . e. , specifically bind, with E-selectin and P-selectin. Preferred antibodies block the functions of both of these molecules.

A. General Characteristics of Antibodies

The basic antibody structural unit is known to comprise a tetramer. Each tetramer is composed of two identical 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, IgM, IgA, IgD and IgE, respectively. Within light and heavy chains, the variable and constant regions are joined by a "J" 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 (incorporated by reference in its entirety for all purposes) .

The variable regions of each light/heavy chain pair form the antibody binding site. Thus, an intact antibody has two binding sites. Except in bifunctional or bispecific antibodies, the two binding sites are the same. The chains all exhibit the 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. 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 definitions of Kabat, Seguences of Proteins of Zπnπunolosrical Interest (National Institutes of Health,

Bethesda, MD, 1987 and 1991), or Chothia & Lesk, J. Mol . Biol . 196:901-917 (1987); Chothia et al.. Nature 342:878-883 (1989).

A bispecific or bifunctional antibody is an artificial hybrid antibody having two different heavy/light chain pairs and two different binding sites. Bispecific antibodies can be produced by a variety of methods including fusion of hybridomas or linking of Fab¹ fragments. See, e.g. , Songsivilai & Lachmann, Clin . Exp. Immunol . 79:315-321 (1990); Kostelny et al., J. Immunol . 148, 1547-1553 (1992). Production of bispecific antibodies can be a relatively labor intensive process compared with production of conventional antibodies and yields and degree of purity are generally lower for bispecific antibodies. Bispecific antibodies do not exist in the form of fragments having a single binding site (e. g. , Fab, Fab¹ and Fv) . B. Binding Specificity and Affinity

The immunoglobulins (or antibodies) of the invention exhibit specific binding to both P-selectin and E-selectin. That is, a single binding site on an antibody has affinity for 5 both P-selectin and E-selectin. Thus, the antibodies bind to epitopes that are common to both molecules. The antibodies bind to the natural and/or recombinant human forms for P-selectin and E-selectin (see Johnston et al., 1989, supra; Bevilacgua et al., 1989, supra) . Some antibodies may also

10. bind P-selectin and/or E-selectin from nonhuman species. Some of the antibodies also specifically bind to L-selectin (preferably human L-selectin (see Tedder, EPA 386,906 (1990)) whereas other antibodies of the invention do not. Surprisingly, the common epitopes bound by the crossreacting

15 antibodies of the invention are also epitopes important for both E-selectin and P-selectin to interact with their counterreceptors on activated leukocytes, such as neutrophils. Thus, most crossreacting antibodies of the invention block the functional interactions of E-selectin or P-selectin and

20 usually those of both of these molecules. Some crossreacting antibodies also block the functional interactions of L-selectin whereas others do not.

Blockage of P-selectin-mediated functions can be demonstrated in vitro. In vitro assays measure the capacity 5 of an antibody to inhibit binding of P-selectin to a counterreceptor. Suitable sources of P-selectin for such assays are purified P-selectin (or an extracellular domain thereof) , cells transfected with P-selectin, activated endothelial cells or platelets. Suitable sources of

30 counterreceptor are leukocytes, neutrophils, monocytes, or HL-60 cells (ATCC CCL 240) and appropriate cell lines transfected with L-selectin. Neutrophils can be isolated from whole blood (preferably human blood) by Ficoll-Hypaque gradient centrifugation. Neutrophils are usually pretreated 5 with rabbit serum to block Fc receptors before adding to a binding assay. When both components in the binding assay are cellular, binding can be assayed microscopically or by flow cytometry. See Kishimoto et al., supra. When one or both components is a purified protein, one component is usually immobilized to a solid phase and the other labelled. Binding is then assayed from label bound to the solid phase. Usually, the antibody is preincubated with the source of P-selectin before adding the source of counterreceptor to the incubation mixture. Blocking activity is shown when an excess of antibody, i . e. , 5-fold, 10-fold or up to 100-fold, substantially inhibits binding of P-selectin to its counterreceptor. The precise degree of inhibition will depend on the assay used. In an assay that measures inhibition of platelet binding to HL-60 cells, an excess of P-selectin blocking antibodies typically exhibits at least 50, 60, 70, 80 or 90% and usually about 80-90% inhibition.

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 2 mM EDTA (a calcium chelator) and the absence of Ca⁺⁺ in an in vitro assay) . By contrast, most blocking antibodies against P-selectin isolated to date require Ca⁺⁺ for activity. See Geng et al., J. Biol . Chem.

266:22313-22318 (1991). Antibodies requiring a Ca⁺⁺ cofactor for blocking activity may be less effective in in vivo conditions where levels of Ca⁺⁺ are expected to fluctuate.

The capacity of the antibodies of the invention to block E-selectin-mediated functions can be demonstrated by analogous in vitro assays to those employed to show blocking of P-selectin mediated functions. Suitable sources of E-selectin are mammalian cell lines transfected with E-selectin, activated endothelial cells, as well as purified E-selectin (or extracellular domains thereof) . If the assay is performed using purified E-selectin, the E-selectin can be immobilized to a solid support. Suitable sources of counterreceptors to E-selectin are leukocytes, neutrophils, monocytes, and HL-60 cells and appropriate cell lines transfected with L-selectin. The degree of binding inhibition will again depend on the components in the assay. In an assay that measures binding between activated endothelial cells and HL-60 cells, the antibodies of the invention, when present in excess, typically exhibit at least about 20, 40, 60, 80% inhibition or more typically about 25-75% or 50% inhibition.

The capacity of antibodies to block L-selectin mediated functions can be demonstrated in a variety of in vitro assays. See, e. g. , copending applications 08/160,516, filed November 30, 1993 and 08/160,074, filed November 30, 1993 (incorporated by reference in their entirety for all purposes) . A simple visual assay for detecting such interaction has been described by Kishimoto et al., supra. Briefly, monolayers of human umbilical vein cells are stimulated with IL-1. Neutrophils, with or without pretreatment with the antibody under test, are added to the monolayer under defined conditions, and the number of adhering neutrophils is determined microscopically. In one method, the neutrophils are obtained from human leukocyte adhesion deficient patients. See Anderson et al., Ann. .Rev. Med. 38:175 (1987). The neutrophils from such patients lack integrin receptors, whose binding to neutrophils might obscure the effects of blocking L-selectin binding. Preferred antibodies selectively bind a functional epitope on P-selectin and E-selectin molecules associated with a response to tissue injury and inflammation. Binding of the antibodies to a functional epitope on P-selectin and E-selectin effectively inhibits adhesion of leukocytes to the activated vascular endothelium and/or to activated platelets in vivo. Preferred antibodies impair the adhesion of leukocytes to the activated vascular endothelium to prevent or inhibit an inflammatory and/or thrombotic condition.

In vivo blocking efficacy can be demonstrated in the same animal models that have been used to show efficacy for antibodies specific for a single adhesion molecule. For example. Mulligan et al., 1991, 1992, supra, describe rat models to test the efficacy of antibodies in protecting against lung injury; Coughlan et al., 1994, describe a rat model for testing the efficacy of antibodies in treatment of systemic endotoxemia; and Weyrich et al., supra, describe a cat model for testing the protective effect of antibodies in myocardial ischemia and reperfusion injury. Other animal models for various inflammatory diseases and disorders are described by Arfors et al., Blood 69:338 (1987) (skin lesions); Tuomanen et al., J. Exp. Med. 170:959 (1989) (brain edema and death produced by bacterial meningitis) ; Lindbom et al., Clin. Immunol . Immunopath. 57:105 (1990) (tissue edema associated with delayed-type hypersensitivity reactions) ; Wegner et al.. Science 247:456 (1990) (airway hyperresponsiveness in allergic asthma); Goldman et al., FASEB J. 5:A509 (1991) (remote lung injury following aspiration) ; Gundel et al., J. Clin. Invest. 88:1407 (1991) (late-phase bronchoconstriction following antigen challenge) ; Hutchings et al.. Nature 346:639 (1990) (diabetes); Flavin et al.. Transplant, Proc. 23:533 (1991) (cardiac allograft survival); Wegner et al.. Am. Rev. Respir. Dis. 143:A544 (1991) (lung damage and dysfunction secondary to oxygen toxicity) ; Cosimi et al., J. Immunol . 144:4604 (1990) (renal allograft rejection); Jasin et al.. Arthritis Rheum. 33:S34 (1990) (antigen-induced arthritis); Thomas et al., FASEB J. 5:A509 (1991) (vascular injury and death in endotoxic shock) ; Bucky et al., Proc. Am. Burn Assoc. 23:133 (1991) (burns); Hernandez et al.. Am. J. Physiol . 253:H699 (1987) (permeability edema following ischemia reperfusion (IR) of intestine) ; Winguist et al.. Circulation 82:111 (1990); Ma et al., Cir. Res . 82:111 (1990) (myocardial damage following myocardial infarction) ; Mileski et al.. Surgery 108:206 (1990) (vascular and tissue damage following hemorrhagic shock and resuscitation) ; Clark et al.. Stroke 22:877 (1991) (central nervous system damage following I/R of the spinal cord); Mileski et al., Proc. Am. Burn Assoc. 22:164 (1990) (edema and tissue damage following frostbite and rewarming) ; Simpson et al.. Circulation 81:226 (1990) (infarct size following I/R of myocardium) . Preferred antibodies show efficacy in at least one and usually several of these inflammatory and thrombotic diseases and conditions. Many of the blocking antibodies of the invention show the same or similar binding specificity as one of the exemplary antibodies designated 5C7.29, 2C9.11 and 1D8.10. That is, the antibodies compete with at least one of the exemplified antibodies for specific binding to E-selectin and/or P-selectin. The E-selectin and P-selectin used in the test is preferably human, and may be natural or recombinant. Competition between antibodies is determined by an assay in which the immunoglobulin under test inhibits specific binding of a reference antibody (e. g. , 5C7.29) to an antigenic determinant on a P-selectin and/or E-selectin molecule. Numerous types of 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 Enzymology 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(1):7-15 (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-selectin or E-selectin bound to a solid surface or cells bearing either of these, an unlabelled test immunoglobulin and a labelled reference immunoglobulin. Competitive inhibition is measured by determining the amount of label bound to the solid surface or cells in the presence of the test immunoglobulin. Usually the test immunoglobulin is present in excess. 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 and/or E-selectin by at least 50 or 75%. The antibodies of the invention usually exhibit a specific binding affinity for P-selectin and E-selectin of greater than or equal to about 10⁶, 10⁷, 10⁸, 10⁹, or 10¹⁰ M^"1. However, antibodies do not necessarily show the same specific binding affinity for each of these ligands. Usually the upper limit of binding affinity of the antibodies is within a factor of about three, five or ten of that of one of the exemplified antibodies. Often the lower limit of binding affinity is also within a factor of about three, five or ten of that of the exemplified antibodies. The term "about" encompasses the degree of experimental error that may typically occur in the measurement of binding affinities.

A hybridoma producing the 5C7.29 antibody has been deposited with the American Type Culture Collection, 12301

Parklawn Dr. , Rockville, Maryland under the Budapest Treaty on May 25, 1994 and given the Accession No. ATCC CRL 11640. The production of this antibody is described in Example 1.

C. Production of Antibodies

(1) Nonhuman Antibodies

Mouse, or other nonhuman antibodies crossreactive with P-selectin and E-selectin can be obtained using a variety of immunization strategies. In some strategies, nonhuman animals (usually nonhuman mammals) , such as mice, are immunized with E-selectin and P-selectin antigens, either concurrently or consecutively. In other strategies, nonhuman animals are immunized with only one of these antigens. Preferred immunogens are cells stably transfected with P-selectin or E-selectin and expressing these molecules on their cell surface. Other preferred immunogens include P-selectin and E-selectin proteins or epitopic fragments of P-selectin and E-selectin containing the segments of these molecules that bind to the exemplified crossreacting antibodies.

Mouse or non-human antibodies crossreactive with all three selectins, i.e., P-selectin, E-selectin, and L-selectin, can be generated by similar strategies. Briefly, mice are immunized either simultaneously or sequentially with cells stably transfected with either P-selectin, E-selectin, or L-selectin, or purified selectin proteins or epitopic fragments thereof. Antibody-producing cells obtained from the immunized animals are immortalized and selected for the production of an antibody which specifically binds to multiple selectins. See generally, Harlow & Lane, Antibodies, A Laboratory Manual (C.S.H.P. NY, 1988) (incorporated by reference for all purposes) . The binding assays for the different selectins can be performed separately or concurrently. Concurrent analysis is conveniently performed by two-color FACS screening after incubation of hybridoma supernatants to cells transfected with selectins. For example, two populations of cells respectively expressing E-selectin and P-selectin are differentially labelled with a first label and tested for capacity to bind hybrido as supernatants. Binding is detected using an appropriate secondary antibody bearing a second label. This scheme is readily extendible to allow simultaneous detection of binding to all three selectins by differentially labelling three populations of cells respectively expressing E-selectin, P-selectin and L-selectin with different intensities of the first label. Alternatively, separate screening for E-selectin, P-selectin and, if desired, L-selectin binding, can be achieved by single color FACS analysis of supernatant binding to transfectant cells or by binding assay to immobilized E-selectin, P-selectin, or L-selectin. Crossreacting antibodies are then further screened for their capacity to block functional properties of E-selectin,

P-selectin and L-selectin using the in vitro and in vivo assays described above. Most antibodies that crossreact with P-selectin or E-selectin also block the functional capacity of both of these molecules to interact with a counterreceptor.

(2) Humanized Antibodies

The invention provides humanized antibodies having similar binding specificity and affinity to selected mouse or other nonhuman antibodies. Humanized antibodies are formed by linking CDR regions (preferably CDR1, CDR2 and CDR3) of non¬ human antibodies to human framework and constant regions by recombinant DNA techniques. See Queen et al., Proc. Natl . Acad. Sci . USA 86:10029-10033 (1989) and WO 90/07861 (incorporated by reference in their entirety for all purposes) . The humanized immunoglobulins have variable region framework residues substantially from a human immunoglobulin (termed an acceptor immunoglobulin) and complementarity determining regions substantially from a mouse immunoglobulin described above, e. g. , the 5C7.29 antibody (referred to as the donor immunoglobulin). The constant region(s) , if present, are also substantially from a human immunoglobulin.

In principal, a framework sequence from any human antibody may serve as the template for CDR grafting. However, it has been demonstrated that straight CDR replacement onto such a framework often leads to significant loss of binding affinity to the antigen (Glaser et al., J. Immunol. 149: 2606 (1992); Tempest et al., Biotechnology 9: 266 (1992); Shalaby et al., J. Exp. Med. 17: 217 (1992)). The more homologous a human antibody is to the original murine antibody, the less likely will combining the murine CDRs with the human framework be to introduce distortions into the CDRs that could reduce affinity. Therefore, homology (that is, percent sequence identity) of at least 65% between the humanized antibody variable region framework and the donor antibody variable region framework is preferred.

The heavy and light chain variable region framework residues can be derived from the same or different human antibody sequences. However, a heavy chain and light chain framework sequences chosen from the same human antibody reduce the possibility of incompatibility in assembly of the two chains. The human antibody seguences can be the seguences of naturally occurring human antibodies or can be consensus sequences of several human antibodies. See Carter et al. , WO 92/22653. Certain amino acids from the human variable region framework residues can be selected for substitution based on their possible influence on CDR conformation and/or binding to antigen. Investigation of such possible influences is by modeling, 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. For example, when an amino acid differs between a murine 5C7.29 variable region framework residue and a selected human variable region framework residue, the human framework amino acid should usually be substituted by the equivalent framework amino acid from the mouse antibody when it is reasonably expected that the amino acid:

(1) contacts antigen directly,

(2) is adjacent to a CDR region in the sequence, or

(3) otherwise interacts with a CDR region (e. g. , is within about 4-6 A of a CDR region) .

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 the donor antibody or from the equivalent positions of more typical human immunoglobulins. The variable region frameworks of humanized immunoglobulins usually show at least 85% sequence identity to a human variable region framework sequence or consensus of such sequences.

(3) Human Antibodies

In another aspect of the invention, human antibodies cross-reactive with E-selectin and P-selectin are provided. These antibodies are produced by a variety of techniques described below. Some human antibodies are selected by competitive binding experiments, or otherwise, to have the same epitope specificity as an exemplified mouse antibody, such as 5C7.29. Such antibodies are particularly likely to share similar therapeutic properties.

a. Trioma Methodology

The basic approach and an exemplary cell fusion partner, SPAZ-4, for use in this approach have been described by Oestberg et al., Hybridoma 2:361-367 (1983); Oestberg, U.S. Patent No. 4,634,664; and Engleman et al., US Patent 4,634,666 (each of which is incorporated by reference in its entirety for all purposes) . The antibody-producing cell lines obtained by this method are called triomas, because they are descended from three cells—two human and one mouse. Initially, a mouse myeloma line is fused with a human B-lymphocyte to obtain a non-antibody-producing xenogeneic hybrid cell, such as the SPAZ-4 cell line described by Oestberg, supra. The xenogeneic cell is then fused with an immunized human B-lymphocyte to obtain an antibody-producing trioma cell line. Triomas have been found to produce antibody more stably than ordinary hybridomas made from human cells.

The B-lymphocytes are obtained from the blood, spleen, lymph nodes or bone marrow of a human donor. In vivo immunization of a living human with E-selectin and/or P-selectin is usually undesirable because of the risk of initiating a harmful response. Thus, B-lymphocytes are usually immunized in vitro with an E-selectin and/or P-selectin or an antigenic fragment of either of these, or a cell bearing either of these. Specific epitopic fragments consisting essentially of the amino acid segments that bind to one of the exemplified murine antibodies are preferred for in vitro immunization. B-lymphocytes are typically exposed to antigen for a period of 7-14 days in a media such as RPMI-1640 (see Engleman, supra) supplemented with 10% human serum.

The immunized B-lymphocytes are fused to a xenogeneic hybrid cell such as SPAZ-4 by well known methods. For example, the cells are treated with 40-50% polyethylene glycol of MW 1000-4000, at about 37 degrees, for about 5-10 min. Cells are separated from the fusion mixture and propagated in media selective for the desired hybrids (e. g. , HAT or AH) . Clones secreting antibodies having the reguired binding specificity are identified by assaying the trioma culture medium for the ability to bind to E-selectin and P-selectin using the same methods as discussed above for nonhuman antibodies. Triomas producing human antibodies having the desired specificity are subcloned by, e. g. , the limiting dilution technique and grown in vitro in culture medium.

Although triomas are genetically stable they may not produce antibodies at very high levels. Expression levels can be increased by cloning antibody genes from the trioma into one or more expression vectors, and transforming the vector into a cell line such as the cell lines discussed, infra, for expression of recombinant or humanized immunoglobulins.

b. Transgenic Non-Human Mammals

Human antibodies crossreactive with P-selectin and E-selectin can also be produced from non-human transgenic mammals having transgenes encoding at least a segment of the human immunoglobulin locus. Usually, the endogenous immunoglobulin locus of such transgenic mammals is functionally inactivated. Preferably, the segment of the human immunoglobulin locus includes unrearranged sequences of heavy and light chain components. Both inactivation of endogenous immunoglobulin genes and introduction of exogenous immunoglobulin genes can be achieved by targeted homologous recombination, or by introduction of YAC chromosomes. The transgenic mammals resulting from this process are capable of functionally rearranging the immunoglobulin component seguences, and expressing a repertoire of antibodies of various isotypes encoded by human immunoglobulin genes, without expressing endogenous immunoglobulin genes. The production and properties of mammals having these properties are described in detail by, e. g. , Lonberg et al., W093/12227 (1993); Kucherlapati, WO 91/10741 (1991) (each of which is incorporated by reference in its entirety for all purposes) . Transgenic mice are particularly suitable. Crossreacting P-selectin/E-selectin human antibodies are obtained by immunizing a transgenic nonhuman mammal, such as described by Lonberg or Kucherlapati, supra, according to the same strategy as discussed for a nontransgenic nonhuman animal (section

I.C.(l)). Monoclonal antibodies are prepared by, e. g. , fusing B-cells from such mammals to suitable myeloma cell lines using conventional Kohler-Milstein technology.

c. Phage Display Methods

A further approach for obtaining human crossreacting antibodies to E-selectin and P-selectin is to screen a DNA library from human B cells as described by Dower et al., WO 91/17271 and McCafferty et al., WO 92/01047 (each of which is incorporated by reference in its entirety for all purposes) . In these methods, libraries of phage are produced in which members display different antibodies on their outer 5 surfaces. Antibodies are usually displayed as Fv or Fab fragments. Phage displaying antibodies are selected by affinity enrichment for binding to either P-selectin or E-selectin. Phage identified by the initial screen are then further screened for crossreaction with the other ligand.

10. 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. , 5C7.29) is used as a starting

15 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

20 variable regions are obtained from a library of rearranged human heavy chain variable regions. A phage showing strong specific binding for P-selectin and E-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

25 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

30 of rearranged human variable light chain regions. Again, phage showing strong specific binding for P-selectin and E-selectin are selected. These phage display the variable regions of completely human antibodies that crossreact with E-selectin and P-selectin. These antibodies usually have the

35 same or similar epitope specificity as the murine starting material (e. g. , 5C7.29). D. Bispecific Antibodies

The invention also provides bispecific or bifunctional antibodies that have one binding site that specifically binds to P-selectin and E-selectin and a second binding site that specifically binds to a second moiety. In bispecific antibodies, one heavy and light chain pair is usually from a crossreacting antibody and the other pair from an antibody raised against another epitope. This results in the property of multi-functional valency, i.e., ability to bind at least two different epitopes simultaneously, one of which is the epitope to which the anti P-selectin/E-selectin crossreacting antibody binds. The other epitope could be e.g., an epitope on L-selectin.

E. Other Therapeutic Agents Having produced an antibody having desirable properties, such as 5C7.29 and the other exemplified antibodies, other nonantibody agents having similar binding specificity/and or affinity can be produced by a variety of methods. For example, Fodor et al., US 5,143,854, discuss a technique termed VLSIPS¹", in which a diverse collection of short peptides are formed at selected positions on a solid substrate. Such peptides could then be screened for binding to an epitopic fragment recognized by 5C7.29, optionally in competition with the 5C7.29. Libraries of short peptides can also be produced using phage-display technology, see, e. g. ,

Devlin W091/18980. The libraries can be screened for binding to an epitopic fragment recognized by e. g. , 5C7.29, optionally in competition with 5C7.29.

II. Nucleic Acids

The genes encoding the heavy and light chains of immunoglobulins produced by hybridoma or trioma cell lines secreting crossreacting antibodies are cloned according to methods described in Sambrook et al., Molecular Cloning: A Laboratory Manual , 2nd ed. (Cold Spring Harbor, NY, 1989) ;

Berger & Kimmel, Methods in Enzymology, Volume 152, Guide to Molecular Cloning Techniques (Academic Press, Inc., San Diego, CA, 1987); Co et al., J. Immunol . 148:1149 (1992). For example, genes encoding heavy and light chains are cloned from a hybridoma's genomic DNA or cDNA produced by reverse transcription of RNA. Cloning is accomplished by conventional techniques including the use of PCR primers that hybridize to the seguences flanking or overlapping the genes, or segments of genes, to be cloned.

Typically, recombinant constructs comprise DNA segments encoding a complete human immunoglobulin heavy chain and/or a complete human immunoglobulin light chain of an immunoglobulin expressed by a hybridoma or trioma cell line. Alternatively, DNA segments encoding only a portion of the primary antibody genes are produced, which portions possess binding and/or effector activities. Other recombinant constructs contain segments of immunoglobulin genes fused to segments of other immunoglobulin genes, particularly segments of other human constant region sequences (heavy and/or light chain) . Human constant region sequences can be selected from various reference sources, including those listed in Kabat et al., supra. DNA segments encoding crossreacting P-selectin/

E-selectin antibodies can be modified by recombinant DNA technigues such as site-directed mutagenesis (see Gillman & Smith, Gene 8:81-97 (1979); Roberts et al.. Nature, 328:731- 734 (1987) . Such modified segments will usually retain antigen binding capacity and/or effector function. Moreover, the modified segments are usually not so far changed from the original sequences to prevent hybridization to these sequences under stringent conditions. The modified segments will usually encode an immunoglobulin showing substantial sequence identity to a reference immunoglobulin from which it was derived. Because, like many genes, immunoglobulin genes contain separate functional regions, each having one or more distinct biological activities, the genes may be fused to functional regions from other genes to produce fusion proteins (e. g. , immunotoxins) having novel properties or novel combinations of properties.

The recombinant polynucleotide constructs will typically include an expression control sequence operably linked to the coding sequences, including naturally-associated or heterologous promoter regions. Preferably, the expression control sequences will be eukaryotic promoter, systems in vectors capable of transforming or transfecting eukaryotic host cells. Once the vector has been incorporated into the appropriate host, the host is maintained under conditions suitable for high level expression of the nucleotide sequences, and the collection and purification of the crossreacting antibodies. These 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. , ampicillin-resistance or hygromycin-resistance, to permit detection of those cells transformed with the desired DNA sequences.

E. coli is one prokaryotic host particularly useful for cloning the DNA sequences of the present invention. Microbes, such as yeast are also useful for expression. Saccharomyces is a preferred yeast host, with suitable vectors having expression control sequences, an origin of replication, termination sequences and the like as desired. Typical promoters include 3-phosphoglycerate kinase and other glycolytic enzymes. Inducible yeast promoters include, among others, promoters from alcohol dehydrogenase, isocytochrome C, and enzymes responsible for maltose and galactose utilization.

Mammalian cells are a preferred host for expressing nucleotide segments encoding immunoglobulins or fragments thereof. See Winnacker, From Genes to Clones, (VCH Publishers, NY, 1987) . A number of suitable host cell lines capable of secreting intact heterologous proteins have been developed in the art, and include CHO cell lines, various COS cell lines, HeLa cells, L cells and myeloma cell lines. Preferably, the cells are nonhuman. Expression vectors for these cells can include expression control sequences, such as an origin of replication, a promoter, an enhancer (Queen et al., Immunol . Rev. 89:49 (1986)), and necessary processing information sites, such as ribosome binding sites, RNA splice sites, polyadenylation sites, and transcriptional terminator sequences. Preferred expression control seguences are promoters derived from endogenous genes, cytomegalovirus, SV40, adenovirus, bovine papillomavirus, and the like. See Co et al., J. Immunol . 148:1149 (1992). The vectors containing the DNA segments of interest can be transferred into the host cell by well-known methods, depending on the type of cellular host. For example, calcium chloride transfection is commonly utilized for prokaryotic cells, whereas calcium phosphate treatment, electroporation, lipofection, biolistics or viral-based transfection may be used for other cellular hosts. Other methods used to transform mammalian cells include the use of polybrene, protoplast fusion, liposomes, electroporation, and microinjection ( see generally, Sambrook et al. , supra) . Once expressed, crossreacting immunoglobulins of the invention can be purified according to standard procedures of the art, including HPLC purification, column chromatography, gel electrophoresis and the like (see generally, Scopes, Protein Purification (Springer-Verlag, NY, 1982)).

III. Epitope Mapping

The P-selectin epitope(s) bound by the 5C7.29 or other crossreacting antibody can be determined by providing a family of fragments containing different amino acid segments from P-selectin. Each fragment typically comprises at least

4, 6, 8, 10, 20, 50 or 100 contiguous amino acids. The family of- polypeptide fragments cover much or all of the amino acid sequence of the extracellular domain of a P-selectin polypeptide. Members of the family are tested individually for binding to e. g. , the 5C7.29 antibody. The smallest fragment that can specifically bind to the antibody being tested contains the amino acid sequence of the epitope recognized by the antibody. The E-selectin epitope bound by the antibody is mapped by an analogous strategy using a family of peptides from E-selectin. The respective epitopes on

P-selectin and E-selectin are expected to map to segments of these molecules showing a high degree of sequence identity. The epitopic fragments are useful as immunogens for generating further crossreacting antibodies. The epitopic fragments are also useful as therapeutic agents that agonize or antagonize the function of P-selectin or E-selectin.

Another method to map epitopes involves testing the ability of an antibody to bind to E-selectin or P-selectin to which random mutations have been introduced. This method is described in more detail in Example 9.

IV. Pharmaceutical Compositions The pharmaceutical compositions for use in the therapeutic methods discussed infra, typically comprise an active agent, such as crossreacting E-selectin/P-selectin antibody, dissolved in an acceptable carrier, preferably an aqueous carrier. Some compositions contain a cocktail of multiple active agents, for example, a crossreacting antibody and a thrombolytic agent. A variety of aqueous carriers can be used, e. g. , -water, buffered water, phosphate buffered saline (PBS), 0.4% saline, 0.3% glycine, human albumin solution and the like. These solutions are sterile and generally free of particulate matter. The compositions may contain pharmaceutically acceptable auxiliary substances as required to approximate physiological conditions such as pH adjusting and buffering agents, toxicity adjusting agents and the like, for example, sodium acetate, sodium chloride, potassium chloride, calcium chloride and sodium lactate. The concentration of antibody in these formulations can vary widely, i.e., from less than about 0.005%, usually at least about 1% to as much as 15 or 20% by weight and will be selected primarily based on fluid volumes, viscosities, and so forth, in accordance with the particular mode of administration selected.

Thus, a typical pharmaceutical composition for injection could be made up to contain 1 ml sterile buffered water, and 1-10 mg of immunoglobulin. A typical composition for intravenous infusion could be made up to contain 250 ml of sterile Ringer's solution, and 150 mg of antibody. Methods for preparing parenterally administrable compositions are described in Remington 's Pharmaceutical Science (15th ed. , Mack Publishing Company, Easton, PA, 1980) , which is incorporated by reference in its entirety for all purposes.

Therapeutic agents of the invention can be frozen or lyophilized for storage and reconstituted in a suitable carrier prior to use. Lyophilization and reconstitution can lead to varying degrees of antibody activity loss (e. g. , with conventional immune globulins, IgM antibodies tend to have greater activity loss than IgG antibodies) . Dosages may have to be adjusted to compensate.

V. Therapeutic Methods

The antibodies of the present invention are useful for treatment of inflammatory diseases and conditions, especially those which are mediated by neutrophils. The dual specificity of the antibodies leads to the inhibition of inflammatory events mediated by either P-selectin or E-selectin.

For example, the antibodies are suitable for therapeutic and prophylactic treatment of ischemia-reperfusion injury caused by myocardial infarction, cerebral ischemic event (e. g. , stroke), renal, hepatic or splenial infarction, brain surgery, lung injury, shock, cardiac surgery (e. g. , coronary artery bypass) , elective angioplasty, and the like. Other preferred applications are the treatment of sepsis, adult respiratory distress syndrome, and multiple organ failure. The antibodies are also useful for treating injury due to trauma, burns, frostbite or damage to the spinal cord. The antibodies will also find use in treating autoimmune diseases including rheumatoid arthritis, systemic lupus erythematosus, multiple sclerosis, type I diabetes and uveitis, in treating inflammatory diseases of the skin such as psoriasis, and in treating meningitis and encephalitis. The antibodies are also useful for treating allergic rhinitis, asthma and anaphylaxis. Other typical applications are the prevention and treatment of organ transplant rejection and graft-versus-host disease.

The pharmaceutical compositions containing the antibodies are particularly useful for parenteral administration, i.e., subcutaneously, intramuscularly or intravenously. The antibodies of the invention may also be administered, typically for local application, by gavage or lavage, intraperitoneal injection, ophthalmic ointment, topical ointment, intracranial injection (typically into a brain ventricle) , intrapericardiac injection, or intrabursal injection.

The compositions containing the present antibodies or a cocktail thereof can be administered for prophylactic and/or therapeutic treatments. In therapeutic applications, compositions are administered to a patient already suffering from an inflammatory disease, in an amount sufficient to cure or at least partially arrest the disease 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 1 to about 200 mg of antibody per dose, with dosages of from 5 to 80 mg per patient being more commonly used. Dosing schedules will vary with the disease state and status of the patient, and will typically range from a single bolus dosage or continuous infusion to. multiple administrations per day (e. g. , every 4-6 hours), or as indicated by the treating physician and the patient's condition. In life-threatening or potentially life- threatening situations, it is possible and may be felt desirable by 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 suffering from a particular disease to enhance 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 generally range from 1 to 80 mg per dose. Preferred prophylactic uses are for the prevention of adult respiratory distress syndrome in patients already suffering from sepsis or trauma; prevention of organ transplant rejection; and prevention of reperfusion injury in patients suffering from ischemia. In seriously ill patients, dosages of about 50 to 150 mg of humanized or human immunoglobulin per administration are freguently used, and larger dosages may be indicated. 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 guantity of the antibody(ies) of this invention sufficient to treat the patient effectively.

The antibodies can also be used in combination with other antibodies, particularly antibodies reactive with different adhesion molecules. For example, suitable antibodies include those specific for CDlla, CDllb, CD18,

L-selectin, and ICAM-1. Other suitable antibodies are those specific for lymphokines, such as IL-1, IL-2 and IFN-γ, and their receptors. The antibodies of the invention can also be administered in conjunction with chemotherapeutic agents. Suitable agents include non-steroidal anti-inflammatory drugs and corticosteroids, but numerous additional agents (e. g. , cyclosporin) can also be used.

In some therapeutic methods of ischemia-reperfusion therapy, crossreacting antibodies are used in combination with thrombolytic agents. In previous methods, patients with myocardial infarction or unstable angina are often treated by opening the occluded coronary artery. Reopening of the obstructed coronary artery can be achieved by administration of thrombolytic agents which lyse the clot causing the obstruction, and which, thereby, restore coronary blood flow. Reperfusion of the vessel can also be achieved by percutaneous transluminal coronary angioplasty (PTCA) by means of balloon dilation of the obstructed and narrowed segment of the coronary artery. However, restoration of coronary blood flow leads to ischemia-reperfusion injury in prior methods.

In the present methods, ischemia-reperfusion injury is reduced or prevented by combination of a thrombolytic agent or of PTCA with crossreacting E-selectin/P-selectin antibodies. Antibodies are usually administered prophylactically before, or at the same time as, administration of thrombolytic agents or initiation of PTCA. Further doses of antibody are then often administered during and after thrombolytic or angioplastic treatment. The interval between prophylactic administration of the antibodies and initiation of thrombolytic or angioplastic treatment is usually 5-60 mins, preferably 5-30 min, and most preferably 5-10 min. The antibodies are administered parentally, preferably by intravenous injection, in doses of 0.01-10 mg/kg body weight, preferably of 0.14 - 5 mg/kg and most preferably of 0.3 - 3 mg/kg. The antibodies can be given as an intravenous bolus injection, e. g. , over 1 - 5 min., as repeated injections of smaller doses, or as an intravenous infusion. The bolus injection is especially useful for the prophylactic dose or in an emergency. Further doses of antibodies can be repeated (e. g. , every 4 - 24 hr) during and after thrombolytic or angioplastic treatment of acute myocardial infarction at the same proportions as described above to achieve optimal plasma levels of the antibody.

Thrombolytic agents are drugs having the capacity, directly or indirectly, to stimulate dissolution of thrombi in vivo. Thrombolytic agents include tissue plasminogen activator (see EP-B 0 093 619) , activase, alteplase, duteplase, silteplase, streptokinase, anistreplase, urokinase, heparin, warfarin and coumarin. Additional thrombolytic agents include saruplase and vampire bat plasminogen activator. See Harris, Protein Engineering 6:449-458 (1987) ; PCT/EP 90/00194; US Patent 4,970,159. Thrombolytic agents are administered to a patient in an amount sufficient to partially disperse, or prevent the formation of, thrombi and their complications. An amount adequate to accomplish this is defined as a "therapeutically effective dose" or "efficacious dose." Amounts effective for this use will depend upon the severity of the condition, the general state of the patient, the route of administration and combination with other drugs. Often, therapeutically effective doses of thrombolytic agents and administration regimens for such agents with crossreacting antibodies to E-selectin and P-selectin are those approved by the FDA for independent uses of thrombolytic agents, e. g. , 100 mg of alteplase or 1.5 million IU of streptokinase.

VI. Methods of Diagnosis

The monoclonal antibodies of the present invention are useful for diagnosing the inflammatory conditions discussed above and monitoring the treatment thereof. The antibodies detect P-selectin and E-selectin in a tissue sample such as serum or endothelial cells, e. g. , by ELISA or RIA.

The presence of either selectin is diagnostic of inflammation. Selectin levels may be employed as a differentiation marker to identify and type cells of certain lineages and developmental origins. In such procedures, the antibody can be labelled directly (e. g. , by radioactive or fluorescent label) and immune complexes detected via the label. Usually, however, the antibody is unlabelled and the desired antigen-monoclonal antibody complex is detected with an enzyme-conjugated antibody against the monoclonal antibody. Diagnosis can also be achieved by in vivo administration of a labelled crossreacting P-selectin/E-selectin antibody and detection by in vivo imaging. The concentration of antibody administered should be su ficient 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.

VII. Other Uses

The antibodies are also useful for affinity purification of selectins and cells expressing the same on their external surfaces. The antibodies can also be used to generate anti- idiotypic antibodies that mimic a selectin domain responsible for antibody binding. Anti-idiotypic antibodies are useful as competitive inhibitors of selectin binding. For example, an anti-idiotypic antibody to a crossreacting P-selectin, E-selectin monoclonal antibody can be selected to compete with P-selectin and/or E-selectin for binding to their counterreceptors. The antibodies are also useful in screening for a therapeutic agent having the same binding specificity as a crossreacting antibody (see Section I. E) .

The following examples are provided to illustrate but not to limit the invention:

Example 1: Preparation of Cells Transfected With Selectins Ll-2 murine pre-B cell selectin transfectants are obtained by inserting the respective human selectin genes downstream of the LCMV promoter in pMRBlOl or similar plasmid (pMRBlOl is a derivative of EEb which contains the E. coli gpt gene. Mulligan et al., Proc. Natl . Acad. Sci . USA 78:2072- 2076 (1981); Stephans et al., Nucleic Acids Research 17:7110 (1989)). Plasmid DNA is introduced into Ll-2 cells by standard methods, such as electroporation, and the cells are selected for resistance to ycophenolic acid. Cells expressing high levels of the appropriate selectin are further selected by "panning" or fluorescence activated cell sorting techniques. See Lymphocytes, A Practical Approach (G.C.B. Klaus, IRL Press, Oxford, England, 1987) .

Example 2: Production of Crossreacting Monoclonal Antibodies Crossreacting antibodies were produced using two different immunization procedures. In all of these procedures, the inoculum was 10⁷ Ll-2 selectin transfectant cells (Berg et al., 1991, 1992, supra) in PBS per injection into mice. In one procedure, Balb/c mice at 4-6 weeks of age (Simonson Labs, Gilroy, CA) were injected IP with

_L1_₂ ^E-se^lec^tin transfectants at day 0 and day 14, and _L1_₂ ^P-se^lec^tin transfectants at day 46, followed by fusion of spleen cells on day 50. In a second procedure, C57/Ld mice at 4-6 weeks of age (Jackson Labs, Bar Harbor, ME) were immunized in the footpad with hypotonically lysed l-2^E"selectin cells on day 0, then with intact l-2^E"selectin cells on days 3 and 6, and with Li_2^p-^selectιn cells on day 9. The draining lymph node lymphocytes were fused on day 12. In each procedure, mouse B-cells were fused with P3X mouse myeloma cells using polyethylene glycol.

Hybridoma supernatants were screened for specific binding to both E- and P-selectin by two-color FACS analysis. _L1_₂ ^p-^selectin and Ll-2^contro1 transfectants were biotinylated by incubation with amino hexanoyl-biotin-N-hydroxy succinimide (Zymed Labs, South San Francisco, CA) at 10 μg/ml in PBS pH 8.0 for 25 min, at RT. After washing, 2 x 10⁷ cells/ml were incubated with FITC-Z-Avidin (Zymed Labs, So. San Francisco, CA) diluted 1:150 for Ll-2^p"^selectin cells and 1:1000 for

_L1_₂con^tro^l _cen_s in FACS Buffer (2% BSA/PBS/10 mM NaN₃) for 30 min at 4°C. After washing, cells were mixed with unlabelled _L1_₂ ^E-se^iec^tin _cen_s at a 1:1:1 ratio in FACS Buffer. 50 μl hybridoma supernatants were added to 200,000 mixed cells in 50 μl in 96-well plates and incubated for 1 hr on ice. After washing, secondary agent was added, 50 μl of 1:500 Goat F(ab')2 anti-mouse IgG-PE conjugated (TAGO, Burlingame, CA) for 30 min prior to washing and fixation. FACS analysis was performed on a Becton Dickinson FACScan¹" (San Jose, CA) , according to standard procedures.

Supernatants containing antibodies reacting with both P-selectin and E-selectin were identified by a shift in red fluorescence of the l-2^{E-selec ιn} transfectant (unlabelled with FITC) and the brightest FITC labelled cells (_L1_₂ ^P-se^lec^tin transfectants) . The control Ll-2 cells

(moderately labelled with FITC) did not show a shift in red fluorescence, indicating that binding.was specific for P-selectin and E-selectin. The yield of crossreacting antibodies as a ratio of supernatants screened was 1/844 and 2/57 for the two immunization schedules.

Supernatants showing binding to P-selectin and E-selectin transfectants were subcloned by limiting dilution and grown in serum free medium containing residual amounts of FBS. Three E-/P-selectin cross-reacting antibodies, designated 5C7.29, 1D8.10 and 2C9.11, were purified from these supernatants on Protein A-Sepharose (Pierce) according to the recommended protocol. Two antibodies reacting only with E-selectin, 1E4 and 2D4, and an antibody reacting only with P-selectin, 5F4, were identified by the same method. The isotypes of 5C7.29, 1D8.10, 2C9.11, 1E4, and 5F4 were determined to be IgGl, and that of 2D4 was determined to be IgG2a using an Innogenetics Inno-Lia mouse monoclonal antibody isotyping kit (Biosource International, Camarillo, CA) .

The three E-/P-selectin crossreacting antibodies were also tested for their ability to bind to the natural ligands, rather than the recombinant forms used in the initial screening assays, by single color FACS analysis. The source of natural E-selectin used in these tests was TNF-o^*-activated human umbilical vein endothelial cells (HUVEC) . In activated form, HUVEC cells express E-selectin, but do not express appreciable amounts of P-selectin. Fig. lb shows that the E-/P-cross-reactive antibody 5C7.29 reacts with TNF-α activated HUVEC (shown by black histograms) but not unactivated HUVEC (grey histograms) . Similar results were obtained for the two other cross-reacting antibodies 2C9.11 and 1D8.10. The activated cells also reacted with the anti-E-selectin blocking antibody H18/7 (Fig. la) (Becton Dickinson (San Jose, CA)) , but not with P-selectin-specific antibodies WAPS 12.2 and 5F4. (WAPS 12.2, a P-selectin blocking antibody, was provided by R. Aaron Warnock and Eugene C. Butcher (Stanford, CA).)

The source of natural P-selectin used in these tests was thrombin-activated platelets. Fig. 2b shows that 5C7.29 binds to these cells as does the known P-selectin antibody WAPS 12.2 (Fig. 2a). Similar results were obtained with 2C9.11 and ID8.10. Platelets did not significantly react with anti-E-selectin antibodies H18/7 or 1E4. The E-/P-selectin crossreacting antibodies were further analyzed for binding to Ll-2^L"selectin transfectants, and with normal human lymphocytes. Specific binding was not observed, demonstrating that the antibodies are specific for E- and P-selectins and do not bind to L-selectin. To confirm that the crossreacting antibodies were truly monoclonal, preclearing experiments were performed. 10 ng antibody (a limiting amount) was incubated with a large number (10⁷) of _Ll-2^E-^selectin cells or _Lι_₂ ^P-seⁱec^tin _{cells for} 1 hr. The supernatant was then transferred to a second aliquot of _Ll-2^E-^selectin cells or _L1_₂ ^p-^{selec in} cells (the same cell type as before) and incubated for 1 hr.. Supernatant was transferred to a third aliquot of cells of the same type as before for a further 1 hr incubation. Supernatant was then removed and examined for reactivity with l-2^E-selectιn, L_l_₂ ^P-se^lec^titι _or LI_ untransfected cells by one-color FACS analysis.

Fig. 3 shows that preincubation of a solution of the 5C7.29 antibody with Ll-2^p"selectin transfectants eliminated subsequent reactivity for both P-selectin and E-selectin. Similar results were found following preincubation with Ll-2^E"selectin transfectants. These results would be obtained only if the antibody bound to both selectins, and not if the antibody were a mixture of two different antibodies, one reactive with E-selectin and one reactive with P-selectin. Therefore, the dual specificities of 5C7.29 reside in the same antibody. Similar results were obtained for the 2C9.11 and 1D8.10 antibodies.

Example 3: Inhibition of E-Selectin-Mediated Functions

The antibody 5C7.29 was tested for the ability to block E-selectin mediated functions. In one assay, the antibody was tested for inhibition of HL-60 binding to tumor necrosis factor-o. (TNF-o.) activated human umbilical vein endothelial cells (HUVEC) . This binding assay simulates the binding of neutrophils to endothelial cells in an inflammatory response. The HL-60 cells are a promyelocytic cell line derived from a patient with acute promyelocytic leukemia. Collins et al.. Nature 270, 347-349 (1977). The HUVEC cells are endothelial cells that when activated with TNF-o. for 4-6 hours express E-selectin, and not P-selectin.

HUVEC were obtained from Clonetics (San Diego, CA) and cultured as suggested. Confluent cultures, up to passage 6, grown in 8 well plastic Lab Tek slides (Nunc,

Naperville, IL) were activated for 4 hours with 1 ng/ml TNF-o. (R&D Systems, Minneapolis, MN) . HUVEC cultures were washed and incubated in 0.15 ml Assay Buffer (10% normal bovine serum/ 10% normal rabbit serum/10 mM HEPES, pH 7.2/RPMI) containing antibodies at 17 μg/ml (i.e., in excess) for 20 min.

HL-60 cells were fluorescently labelled with 6-carboxyfluorescein diacetate acetoxy- ethyl ester (CFDA-AM, Molecular Probes, Eugene OR) (von Andrian et al., 1991, supra) by a 30 min incubation in 10 mg/ l RPMI/10 mM HEPES, pH 7.2, washed and resuspended in Assay Buffer and incubated at RT for 20 min. The resuspended cells (6 x 10⁵ cells in 0.15 ml) were then added to the HUVEC cultures.

Slides were rotated at 50 rpm on a rotator (Innova 200, New Brunswick Inc.) for 15 min at RT. The cover slips were removed and non-adherent HL-60 cells washed off by dipping slides in DMEM. Adherent cells were fixed by immersion in 1% paraformaldehyde-PBS. Slides were examined microscopically and the number of bound cells per field determined. Two treatments per slide (in quadruplicate) were analyzed.

Fig. 4 shows that the number of HL-60 cells binding to the activated HUVEC was decreased 47% by preincubation with 5C7.29. This compared favorably with blocking by the anti-E-selectin-specific antibody H18/7 (38%) . Binding was not significantly reduced by a control antibody.

Because HUVEC can also express P-selectin (although only at low levels under the present activation conditions) , 5C7.29 was also tested for HL-60 binding to CHO cells transfected with E-selectin. CHO cells permanently transfected with a truncated form of E-selectin containing the first four N-terminal domains of E-selectin fused to the transmembrane and cytoplasmic domain of another protein were produced according to standard methods. Expression was confirmed by reactivity with a control anti-E-selectin antibody (H18/7) . Inhibition of binding between fluorescently labelled HL-60 and the transfected CHO cells was performed using the same assay as for the TNF-o.-activated HUVEC. 5C7.29 was found to block adhesion by 82% (Fig. 5) . similar results were observed with 1D8.10, 2C9.11 and the E-selectin blocking antibody 1E4. The non-blocking P-selectin specific control antibody 5F4 had no significant effect in this assay.

The cross-reacting antibodies also blocked normal human peripheral blood neutrophil binding to TNF-o^*-activated HUVEC. At a final concentration of 10 μg/ml, 5C7.29 blocked 71 +/-13%, 2C9.11 blocked 62 +/-8% and 1D8 blocked 52 +/-10% of neutrophil binding to activated HUVEC, while the anti-E-selectin antibodies 1E4 and H18/7 (Bevilacgua et al., 1987, supra) blocked 68+/~ % and 68 +/"15%/ and a control mouse IgGl antibody did not block (-21% +/-!!%) , n=4. For these experiments, neutrophils were isolated from normal human blood by density gradient centrifugation and dextran sedimentation by standard procedures (Current Protocols in Immunology, Coligan et al., eds., John Wiley and Sons, New York, 1992) . Assays were performed as for HL-60 cells except neutrophils were added to HUVEC at 7.5 X 10⁴ in 0.15 ml.

Example 4: Inhibition of P-selectin-Mediated Functions

The antibodies 5C7.29, 2C7.11 and 1D8.10 were tested for their ability to block P-selectin-mediated functions.

Blocking was tested in a platelet-HL-60 rosette assay (Corral et al., 1990, supra) . The platelets provide a source of cells expressing P-selectin and the HL-60 cells simulate neutrophils. Normal human blood was collected with sodium citrate as anticoagulant and the platelet-rich plasma (PRP) prepared by centrifugation at 250g for 10 min. Platelets were isolated from PRP by centrifugation at lOOOg for 20 min and resuspended at 3 x 10⁸/ml in PBS, pH 7.2. Monoclonal antibodies (1 μg in 20 μl, i.e., an excess) were added to 20 μl platelets. In some experiments normal human thrombin

(0.3 U/μl) was added to activate the platelets as described by Corral et al., 1990, supra. After 45 min, 20 μl HL-60 cells (10⁶/^ml iⁿ PBS) were added and further incubated for 45 min. Bound platelets were fixed to HL-60 cells by addition of glutaraldehyde to 1.25%. At least 100 HL-60 cells for each sample were observed microscopically and the number of cells with bound platelets (>2 platelets per HL-60 cell) determined. Fig. 6 shows that all three crossreacting antibodies block rosetting to about the same extent as the P-selectin specific blocking antibody WAPS 12.2. Similar blocking experiments can be performed using human peripheral blood neutrophils in place of HL-60 cells. Neutrophils are prepared by the same method and used at the same concentration as described in Example 3.

Example 5: Cloning and sequencing of mouse 5C7.29 heavy chain and light chain variable region cDNA cDNAs for the heavy chain and light chain variable region genes of the mouse 5C7.29 antibody were cloned using anchored polymerase chain reactions as described (see Co et al., J". Jmmunol. 148: 1149 (1992)), using 3' primers that hybridized to the constant regions and contained Hindlll sites, and 5' primers that hybridized to the dG tails and contained EcoRI sites. The PCR amplified fragments were digested with EcoRI and Hindlll and cloned into the pUClδ or pUC19 vectors for sequencing. At least two gamma-1 specific and two kappa specific clones were seguenced. The gamma-1 clones and the kappa clones are respectively identical in sequence. The variable region cDNA sequences and the deduced amino acid sequences for the gamma-1 and kappa chains are shown in Fig. 7A-7B [SEQ ID NOS:1-4].

Example 6: Design of humanized 5C7.29 antibody variable domain

Based on a seguence homology search against the NBRF protein seguence database, the variable regions of light chain subclass I and heavy chain subclass III show good homology to the mouse 5C7.29 antibody. In particular, the antibody III-3R provides the best framework homology with 5C7.29 and was chosen to provide the framework sequences for humanization of 5C7.29. However, other members of the light chain subclass I and heavy chain subclass III would also be especially suitable for use in providing the frameworks of the respective humanized 5C7.29 chains.

The computer program ENCAD (M. Levitt et al., J. Mol. Biol. 168: 595 (1983)) was used to construct a molecular model of the 5C7.29 variable domain. The program ABMOD (B.T. Zilber et al. Biochem. 29:10032-41) is also useful. The model was used to determine the amino acids in the 5C7.29 framework that were close enough to the CDRs to potentially interact with them. To design the humanized light and heavy chain 5C7.29 variable regions, the CDRs from the mouse 5C7.29 antibody were grafted into the framework sequences of the III-3R antibody. At framework positions where the model suggested contact with the CDRs, the amino acids from the mouse 5C7.29 antibody were chosen to replace the residues in the III-3R sequence. For humanized 5C7.29, this was done at residues 69 and 70 in the light chain and at no residues in the heavy chain. Moreover, at some positions where the amino acid was unusual for human antibodies at that position, an amino acid representing a consensus among the relevant human subclass was substituted for the III-3R framework residue. For humanized 5C7.29, this was done at residues 61, 72, 82 and 99 in the light chain and residues 1, 75 and 78 in the heavy chain. The final sequence of the humanized 5C7.29 heavy and light chain variable region is shown in Figs. 8A-8B [SEQ ID NOS:5-8]. However, many of the potential CDR-contact residues are amenable to substitutions of other amino acids and may still allow the antibody to retain substantial affinity to the antigens. The following table lists a number of positions in the framework where alternative amino acids may be suitable (note LC = light chain, HC = heavy chain) :

Likewise, many of the framework residues not contacting the CDRs in the humanized 5C7.29 heavy and light chains are also amenable to substitutions with amino acids from either the human III-3R antibody, or from the corresponding position of other human antibodies, or from the mouse 5C7.29 or other mouse antibodies, while still preserving substantial affinity and non-immunogenicity of the humanized antibody. The following table lists a number of positions in the framework where alternative amino acids may be suitable:

Finally, even certain residues in the CDRs may be substituted with other residues while the antibody may retain substantially the same affinity and specificity. Structure- function studies of antibody binding reveal that not all of the CDR amino acids participate equally in specifying affinity towards a given antigen (or set of related antigens) . These studies enable prediction with some reliability of particular CDR positions least likely to change substantially the binding characteristics of an antibody. For example, Chothia and co- workers define structurally acceptable amino acids in CDR positions (Chothia et al., J. Mol . Biol . 196: 902 (1987);

Chothia et al.. Nature 342: 877 (1989); and Tramontano et al.. Proteins: Struct. Funct. Genet. 6: 382 (1989)), and many of these are not accessible to solvent (i.e, available to participate in binding), in the model of 5C7.29. Other workers have shown that residues 61-66 of CDR H2 may not participate in antigen binding (Carter et al., Proc. Natl . Acad. Sci . USA 89: 4285 (1992); Hsiao et al.. Protein Eng. 7:815 (1994)). Surveys of antibody-antigen complex structures support this notion (Glaser et al., J. Immunol . 149: 2606 (1992); Padlan, Mol . Immunol . 31: 169 (1994)). Some of these CDR residues that may be changed in humanized 5C7.29 and their potential substitutions are listed in the following table:

Selection of various combinations of alternative amino acids may be used to produce versions of humanized 5C7.29 that have varying combinations of affinity, specificity, non-immunogenicity, ease of manufacture and other desirable properties. The above examples are offered by way of illustration, not of limitation.

Example 7: Construction of humanized 5C7.29

For the construction of variable region genes for the humanized 5C7.29 antibody, nucleotide sequences were selected that encode the protein sequences of the humanized heavy and light chains, including the signal peptide, generally utilizing codons found in the mouse seguence. Several degenerate codons were changed to create restriction sites or to remove undesirable ones. The nucleotide seguences of the genes also included splice donor signals and an Xbal site at each end. The nucleotide seguences and encoded light and heavy chain variable regions of the humanized 5C7.29 antibody are shown in Figs. 8A-8B [SEQ ID NOS:5-8]. Each gene was constructed from eight overlapping synthetic oligonucleotides. Assembly and amplification of the genes were carried out in four steps as shown in Fig. 9: (1) the four pairs of complementary oligonucleotides were annealed and extended with Klenow polymerase in separate reactions; (2) the resulting four double-stranded DNA fragments were mixed in pairs, denatured, re-annealed and extended in two separate reactions using Klenow fragment; (3) the resulting two double-stranded half-gene fragments were mixed, denatured, re-annealed and extended to create the full length double stranded variable region genes; (4) the gene fragments were finally amplified, using Taq polymerase and two primers that hybridize to the 5' and the 3' end of the variable region genes and contain Xbal sites for cloning into the respective expression vectors, pVk and pVg4. Reactions were carried out under conditions well-known in the art.

The pVk vector for light chain expression and the pVgl vector for heavy chain expression have been previously described (see Co et al., J. Immunol . 148: 1149 (1992)). To produce a humanized 5C7.29 antibody of the IgG4 isotype, the heavy chain expression vector pVg4 has been constructed. To do so, the Xbal-BamHI fragment of pVgl containing the γl constant region was replaced with an approximately 2000 bp fragment of the human γ4 constant region gene (Ellison and Hood, Proc. Natl . Acad. Sci USA 79:1984 (1982)) that extended from the Hindlll side preceding the CHI exon of the γ4 gene to 270 bp after the Nsil site following the CH4 exon of the gene, using methods well-known to those skilled in the art, including polymerase chain reaction. The heavy chain and light chain plasmids were transfected into a mouse myeloma cell line Sp2/0-Agl4 (ATCC CRL 1581) . Transfection was by electroporation using a Gene Pulser apparatus (Bio-Rad) at 360 V and 25 uFD capacitance according to the manufacturer's instructions. Before transfection, the light chain- and heavy chain-containing plasmids were linearized using PvuII, extracted with phenol- chloroform, and ethanol-precipitated. All transfections were done using 30-50 μg plasmid DNA and about 10⁷ cells in PBS. The cells from each transfection were plated into 2 to 4 96-well tissue culture plates. After 48 hours, selective medium was applied.

Cells were selected for gpt expression•in DMEM + 10% FBS + HT media supplement (Sigma) + 1 μg/ml mycophenolic acid. Antibody-producing clones were screened by assaying human antibody production in the culture supernatant by ELISA. Antibody from the best-producing clones was purified by passing tissue culture supernatant over a column of protein A-Sepharose CL-4B (Pharmacia) . The bound antibodies were eluted with 0.2 M glycine-HCl, pH 3.0, and neutralized with

1 M Tris-HCl, pH 8.0. The buffer was exchanged into phosphate buffered saline (PBS) by passing over a PD10 column (Pharmacia), or by dialysis. To obtain cells producing higher levels of antibody, the transfected clones may be cultured in increasing concentrations of methotrexate.

Example 8: Properties of humanized 5C7.29

To show that humanized 5C7.29 specifically binds to E-selectin and P-selectin, _Lι_₂E-selectin and _Lι_₂P-selectin transfectants were incubated with humanized 5C7.29 or control antibodies for 1 hour. After washing, cells were incubated in a 1:400 dilution of phycoerythrin-conjugated anti-human Ig (Biosource, Camarillo, CA) , washed, then analyzed for fluorescence by flow cytometry (FACS) as previously described (Berg et al., Blood 85: 31 (1995)). Humanized 5C7.29 reacts with both _Lι-₂E-selectin _{and L1}_₂P-selectin transfectants, but not l-2^{L'selec in} transfectants (Fig. 10) demonstrating that the humanization process did not result in loss of either E-selectin or P-selectin binding or gain in the ability to bind L-selectin.

The affinity of the humanized 5C7.29 antibody for E-selectin and P-selectin was separately determined by competition with the radio-iodinated mouse 5C7.29 antibody (Fig. 11) . Purified mouse 5C7.29 antibody was labeled with Na¹²⁵I (Amersham, Arlington Heights, IL) using the lactoperoxidase procedure to 4 mCi/mg of protein. _CH0 ^E-se^lec^tin _Cells and _Ll-2^p"^selectin cells, which were obtained by transfecting the E-selectin and P-selectin genes into the respective host cells CHO and Ll-2 (Gallatin et al.. Nature 304:30 (1983)) by methods well known in the art (see, e.g., Larsen et al., J. Biol . Chem. 267: 11104 (1992)),, were used as sources for E-selectin and P-selectin. Increasing amounts of competitor antibody (mouse 5C7.29 or humanized 5C7.29) were added to 2 ng of radio-iodinated tracer mouse 5C7.29 antibody and incubated with 4 X 10⁵ _CH0^E-^selectin cells or _Ll-2^p-^{*selec in} cells in 0.2 ml of binding buffer (PBS with 2% fetal calf serum, 0.1% sodium azide) for 2 hours at 4° C with constant shaking. Cells were then washed and centrifuged, and their radioactivities measured. The ratio of bound and free tracer antibody were calculated (Figs. 11A and 11B) .

The binding affinities were calculated according to the methods of Berzofsky and Berkower (J. A. Berzofsky and I. J. Berkower, in Fundamental Jmmunologry (ed. W.E. Paul), Raven Press (New York), p. 595 (1984)). The humanized 5C7.29 had an affinity of approximately 3 x 10⁸ M^"1 for E-selectin, with no significant difference from that of mouse 5C7.29, and an affinity of approximately 1.3 X 10⁸ M^"1 for P-selectin, within about 3 to 4-fold of the mouse 5C7.29 antibody. This experiment also shows directly that humanized 5C7.29 has the ability to compete with the mouse 5C7.29 antibody for binding to both E-selectin and P-selectin. In another experiment, the affinities of mouse and humanized 5C7.29 for P-selectin were determined by the method of Scatchard (Berzofsky and Berkower, supra) to be approximately 1.9 x 10⁸ M^"1 and 6 x 10⁸ M^"1, respectively, in good agreement with the results of the competitive binding experiment. To show that the humanized 5C7.29 antibody inhibits binding of E-selectin to a counter-receptor for E-selectin, its ability to inhibit the binding of HL-60 cells to E-selectin transfectant cells was determined. Assays of the adhesion of HL-60 cells with cHO^E"selectin cells were performed as previously described (Berg et al.. Blood 85: 31 (1995), and supra) in the presence of monoclonal antibodies at the indicated concentrations. Fig. 12 shows that humanized 5C7.29 blocks binding of HL-60 cells to cHO^E_selectin transfectants as well as mouse 5C7.29. For the representative experiment shown, two treatments per slide (each treatment in quadruplicate) were analyzed and the mean and standard deviations calculated. An isotype-matched control antibody did not affect binding. To show that the humanized 5C7.29 antibody inhibits binding of P-selectin to a counter-receptor for P-selectin, its ability to inhibit the binding of HL-60 cells and activated platelets was determined. Assays of the rosetting of activated platelets to the HL-60 cells were performed as described (Berg et al.. Blood 85:31 (1995) and supra) in the presence of monoclonal antibodies at the indicated concentrations. Fig. 13 shows that humanized 5C7.29 blocks binding of platelets to HL-60 cells as well as mouse 5C7.29. An isotype-matched control antibody had no affect on binding in this assay. The representative experiment shown was performed in triplicate and the mean and standard deviations calculated.

Example 9: Epitope mapping of 5C7.29 To determine the amino acids of E-selectin involved in the binding of 5C7.29 (the epitope), the following procedure was used. DNA encoding the lectin and EGF-like domains of human E-selectin were fused to a gene encoding the human immunoglobulin lambda constant region (C_λ) , which served as a tag. The chimeric DNA was inserted in a plasmid vector, which provided a lac promoter and pelB signal sequence for expression and secretion of the chimeric (fusion) protein in E. coli . The E-selectin domains were randomly mutagenized by error-prone polymerase chain reaction (PCR) utilizing AmpliTaq enzyme (Perkin Elmer) and Mn⁺⁺, and the amino acid substitutions were determined by DNA sequencing. E. coli strain TGl-irecA was transformed with the wild-type and mutant plasmids, and chimeric proteins were overexpressed by growing transformed E. coli in 2YT broth. After 8 hours of induction with lmM IPTG, culture supernatants containing the chimeric proteins were collected. All operations were performed according to methods well-known in the art of molecular biology.

Next, 96-well plates were coated with the 5C7.29 antibody (or control antibodies) . After blocking, the plates were incubated with the E. coli supernatants and then with HRP-conjugated anti-human C_λ antibodies (Biosource, Camarillo, CA) . After washing, bound enzyme was detected with TMB substrate. Supernatants containing mutant E-selectin-C_λ chimeric protein to which 5C7.29 could still bind gave a positive signal, while supernatants containing mutant E-selectin to which 5C7.29 could not bind gave a negative signal. The results are shown in the following table, where the symbol AXB means a mutant in which the Xth amino of E-selectin form the mature N-terminus is changed from the normal A to mutant B.

Mutant Reactivity

Q21R —

R22G —

Y23H —

Y48H +

E92G +

N105S +

K111E +

T119A —

A120T - Because mutating amino acids Q₂₁, R₂₂, Y₂₃, T₁₁₉ and A₁₂₀ in E-selectin prevented binding of antibody 5C7.29, these amino acids must be in the epitope of 5C7.29. The full amino acid seguence of E-selectin is given in Bevilacqua, supra and in United States Patent 5,272,263 (ELAM-1) . (Another anti-E-selectin antibody was able to bind to these mutants, showing that they did not disrupt the overall structure of E-selectin) . Other E/P cross-reacting antibodies that show a different pattern of reactivity with these E-selectin mutants must have a different epitope in E-selectin. The epitope of 5C7.29 in P-selectin may be determined by a similar procedure using P-selectin mutants, and may be similarly compared to the epitope of other E/P cross-reacting antibodies. The epitopes of 5C7.29 in E-selectin and P-selectin are preferred epitopes, because antibodies such as 5C7.29 that bind to them may have high affinity and blocking activity.

While the foregoing invention has been described in some detail for purposes of clarity and understanding, it will be clear to one skilled in the art from a reading of this disclosure that various changes in form and detail can be made without departing from the true scope of the invention. All publications and patent documents cited in this application are incorporated by reference in their entirety for all purposes to the same extent as if each individual publication or patent document were so individually denoted.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: Berg, Ellen L.

(ii) TITLE OF INVENTION: Cross-Reacting Monoclonal Antibodies Specific for E-Selectin and P-Selectin

(iii) NUMBER OF SEQUENCES: 10

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Townsend and Townsend Khourie and Crew

(B) STREET: One Market Plaza, Steuart Tower, Suite 2000

(C) CITY: San Francisco

(D) STATE: California

(E) COUNTRY: USA

(F) ZIP: 94105

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

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

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

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: WO

(B) FILING DATE:

(C) CLASSIFICATION:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: 08/259,963

(B) FILING DATE: 14-JUNE-94

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Smith, William M.

(B) REGISTRATION NUMBER: 30,223

(C) REFERENCE/DOCKET NUMBER: 11823-005810PC

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: 415-326-2400

(B) TELEFAX: 415-326-2422

(2) INFORMATION FOR SEQ ID N0:1:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 384 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 1..384

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

ATG GAT TTT CAA GTG CAG ATT TTC AGC TTC CTG CTA ATC AGT GCC TCA 48 Met Asp Phe Gin Val Gin lie Phe Ser Phe Leu Leu lie Ser Ala Ser 1 5 10 15

GTC ATA ATA TCC AGA GGA CAA ATT GTT CTC ACC CAG TCT CCA GCA ATC 96 Val He He Ser Arg Gly Gin He Val Leu Thr Gin Ser Pro Ala He 20 25 30

ATG TCT GCA TCT CCA GGG GAG AAG GTC ACC ATG ACC TGC AGT GCC AGC 144 Met Ser Ala Ser Pro Gly Glu Lys Val Thr Met Thr Cys Ser Ala Ser 35 40 45

TCA AGT GTG CCT TAC ATG CAC TGG TAT CAG CAG AAG TCA GGC ACC TCC 192 Ser Ser Val Pro Tyr Met His Trp Tyr Gin Gin Lys Ser Gly Thr Ser 50 55 60

CCC AAA TTA TGG ATT TAT GAC ACA TCC AAT CTG GCT TCT GGA GTC CCT 240

Pro Lys Leu Trp He Tyr Asp Thr Ser Asn Leu Ala Ser Gly Val Pro

65 70 75 80

GCT CGC TTC AGT GGC AGT GGG TCT GGG ACC TCT TAC TCT CTC ACA ATC 288 Ala Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu Thr He 85 90 95

AGC AGC ATG GAG GCT GAA GAT GCT GCC ACT TAT TAC TGC CAG CAG TGG 336 Ser Ser Met Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gin Gin Trp 100 105 110

AGT AGT GAC CCA TTC ACG TTC GGC TCG GGG ACA AAG TTG GAA ATA AAG 384 Ser Ser Asp Pro Phe Thr Phe Gly Ser Gly Thr Lys Leu Glu He Lys 115 120 125

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 128 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

^• (ii) MOLECULE TYPE: protein

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

Met Asp Phe Gin Val Gin He Phe Ser Phe Leu Leu He Ser Ala Ser 1 5 10 15

Val He He Ser Arg Gly Gin He Val Leu Thr Gin Ser Pro Ala He 20 25 30

Met Ser Ala Ser Pro Gly Glu Lys Val Thr Met Thr Cys Ser Ala Ser 35 40 45

Ser Ser Val Pro Tyr Met His Trp Tyr Gin Gin Lys Ser Gly Thr Ser 50 55 60

Pro Lys Leu Trp He Tyr Asp Thr Ser Asn Leu Ala Ser Gly Val Pro 65 70 75 80

Ala Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Ser Leu Thr He 85 90 95

Ser Ser Met Glu Ala Glu Asp Ala Ala Thr Tyr Tyr Cys Gin Gin Trp 100 105 no

Ser Ser Asp Pro Phe Thr Phe Gly Ser Gly Thr Lys Leu Glu He Lys 115 120 125

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 405 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 1..405

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

ATG GAC TCC AGG CTC AAT TTA GTT TTC CTT GTC CTT ATT TTA AAA GGT 48 Met Asp Ser Arg Leu Asn Leu Val Phe Leu Val Leu He Leu Lys Gly 1 5 10 15

GTC CAG TGT GAT GTA CGA CTG GTG GAG TCT GGG GGA GGT TTA GTG CAG 96 Val Gin Cys Asp Val Arg Leu Val Glu Ser Gly Gly Gly Leu Val Gin 20 25 30

CCT GGA GGG TCC CGG AAA CTC TCC TGT GCA GCC TCT GGA TTC ACT TTC 144 Pro Gly Gly Ser Arg Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe 35 40 45

AGT AGC TTT GGA ATG CAC TGG GTT CGT CAG GCT CCT GAT AAG GGG CTG 192 Ser Ser Phe Gly Met His Trp Val Arg Gin Ala Pro Asp Lys Gly Leu 50 55 60

GAG TGG GTC GCA TTC ATT AGC AGT GGC AGT AGT ACC ATC TAC TAT GCT 240

Glu Trp Val Ala Phe He Ser Ser Gly Ser Ser Thr He Tyr Tyr Ala

65 70 75 80

GAC ACA GTG AGG GGC CGA TTC ACC ATC TCC AGA GAC AGT CCC AAG AAC 288 Asp Thr Val Arg Gly Arg Phe Thr He Ser Arg Asp Ser Pro Lys Asn 85 90 95

ACC CTG TTC CTG CAA ATG ACC AGT CTA AGG TCT GAG GAC ACG GCC ATG 336 Thr Leu Phe Leu Gin Met Thr Ser Leu Arg Ser Glu Asp Thr Ala Met 100 105 110

TAT TAC TGT GCA AGA CCT TTA CCC CCG TTT GCT TAC TGG GGC CAA GGG 384 Tyr Tyr Cys Ala Arg Pro Leu Pro Pro Phe Ala Tyr Trp Gly Gin Gly 115 120 125

ACT TTG GTC ACT GTC TCT GCA 405

Thr Leu Val Thr Val Ser Ala

130 135 (2) INFORMATION FOR SEQ ID N0:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 135 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

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

Met Asp Ser Arg Leu Asn Leu Val Phe Leu Val Leu He Leu Lys Gly 1 5 10 15

Val Gin Cys Asp Val Arg Leu Val Glu Ser Gly Gly Gly Leu Val Gin 20 25 30

Pro Gly Gly Ser Arg Lys Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe 35 40 45

Ser Ser Phe Gly Met His Trp Val Arg Gin Ala Pro Asp Lys Gly Leu 50 55 60

Glu Trp Val Ala Phe He Ser Ser Gly Ser Ser Thr He Tyr Tyr Ala 65 70 75 80

Asp Thr Val Arg Gly Arg Phe Thr He Ser Arg Asp Ser Pro Lys Asn 85 90 95

Thr Leu Phe Leu Gin Met Thr Ser Leu Arg Ser Glu Asp Thr Ala Met 100 105 110

Tyr Tyr Cys Ala Arg Pro Leu Pro Pro Phe Ala Tyr Trp Gly Gin Gly 115 120 125

Thr Leu Val Thr Val Ser Ala 130 135

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 384 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: CDNA

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 1..384

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5 :

ATG GAT TTT CAA GTG CAG ATT TTC AGC TTC CTG CTA ATC AGT GCC TCA 48 Met Asp Phe Gin Val Gin He Phe Ser Phe Leu Leu He Ser Ala Ser 1 5 10 15

GTC ATA ATA TCC AGA GGA GAT ATT CAA ATG ACC CAG TCT CCA TCT AGC 96 Val He He Ser Arg Gly Asp He Gin Met Thr Gin Ser Pro Ser Ser 20 25 30

TTA TCT GCA TCT GTA GGG GAT AGG GTC ACC ATA ACC TGC AGT GCC AGC 144 Leu Ser Ala Ser Val Gly Asp Arg Val Thr He Thr Cys Ser Ala Ser 35 40 45

TCA AGT GTG CCT TAC ATG CAC TGG TAT CAG CAG AAG CCA GGC AAA GCC 192 Ser Ser Val Pro Tyr Met His Trp Tyr Gin Gin Lys Pro Gly Lys Ala 50 55 60

CCC AAA TTA TTG ATT TAT GAC ACA TCC AAT CTG GCT TCT GGA GTC CCT 240

Pro Lys Leu Leu He Tyr Asp Thr Ser Asn Leu Ala Ser Gly Val Pro

65 70 75 80

TCT CGC TTC AGT GGC AGT GGG TCT GGG ACC TCT TAC ACT CTC ACA ATC 288 Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Thr Leu Thr He 85 90 95

AGC AGC CTG CAG CCT GAA GAT TTT GCC ACT TAT TAC TGC CAG CAG TGG 336 Ser Ser Leu Gin Pro Glu Asp Phe Ala Thr Tyr Tyr Cys Gin Gin Trp 100 105 110

AGT AGT GAC CCA TTC ACG TTC GGC CAG GGG ACA AAG GTG GAA ATA AAG 384 Ser Ser Asp Pro Phe Thr Phe Gly Gin Gly Thr Lys Val Glu He Lys 115 120 125

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 128 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

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

Met Asp Phe Gin Val Gin He Phe Ser Phe Leu Leu He Ser Ala Ser 1 5 10 15

Val He He Ser Arg Gly Asp He Gin Met Thr Gin Ser Pro Ser Ser 20 25 30

Leu Ser Ala Ser Val Gly Asp Arg Val Thr He Thr Cys Ser Ala Ser 35 40 45

Ser Ser Val Pro Tyr Met His Trp Tyr Gin Gin Lys Pro Gly Lys Ala 50 55 60

Pro Lys Leu Leu He Tyr Asp Thr Ser Asn Leu Ala Ser Gly Val Pro 65 70 75 80

Ser Arg Phe Ser Gly Ser Gly Ser Gly Thr Ser Tyr Thr Leu Thr He 85 90 95

Ser Ser Leu Gin Pro Glu Asp Phe Ala Thr Tyr Tyr Cys Gin Gin Trp 100 105 110

Ser Ser Asp Pro Phe Thr Phe Gly Gin Gly Thr Lys Val Glu He Lys 115 120 125

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 405 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 1..405

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

GTC CAG TGT GAA GTA CAA CTG GTG GAG TCT GGG GGA GGT TTA GTG CAG 96 Val Gin Cys Glu Val Gin Leu Val Glu Ser Gly Gly Gly Leu Val Gin 20 25 30

CCT GGA GGG TCC CTT CGT CTC TCC TGT GCA GCC TCT GGA TTC ACT TTC 144 Pro Gly Gly Ser Leu Arg Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe 35 40 45

AGT AGC TTT GGA ATG CAC TGG GTT CGT CAG GCT CCT GGT AAG GGG CTG 192 Ser Ser Phe Gly Met His Trp Val Arg Gin Ala Pro Gly Lys Gly Leu 50 55 60

GAG TGG GTC GCA TTC ATT AGC AGT GGC AGT AGT ACC ATC TAC TAT GCT 240

Glu Trp Val Ala Phe He Ser Ser Gly Ser Ser Thr He Tyr Tyr Ala

65 70 75 80

GAC ACA GTG AGG GGC CGA TTC ACC ATC TCC AGA GAC AAC ACC AAG AAC 288 Asp Thr Val Arg Gly Arg Phe Thr He Ser Arg Asp Asn Ser Lys Asn 85 90 95

ACC CTG TAT CTG CAA ATG AAC AGT CTA AGG GCT GAG GAC ACG GCC GTG 336 Thr Leu Tyr Leu Gin Met Asn Ser Leu Arg Ala Glu Asp Thr Ala Val 100 105 110

ACT TTG GTC ACT GTC TCT GCA 405

Thr Leu Val Thr Val Ser Ala

130 135 (2) INFORMATION FOR SEQ ID NO:8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 135 amino acids

(B) TYPE: amino acid (D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:

Met Asp Ser Arg Leu Asn Leu Val Phe Leu Val Leu He Leu Lys Gly 1 5 10 15

Val Gin Cys Glu Val Gin Leu Val Glu Ser Gly Gly Gly Leu Val Gin 20 25 30

Pro Gly Gly Ser Leu Arg Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe 35 40 45

Ser Ser Phe Gly Met His Trp Val Arg Gin Ala Pro Asp Lys Gly Leu 50 55 60

Glu Trp Val Ala Phe He Ser Ser Gly Ser Ser Thr He Tyr Tyr Ala 65 70 75 80

Asp Thr Val Arg Gly Arg Phe Thr He Ser Arg Asp Asn Ser Lys Asn 85 90 95

Thr Leu Tyr Leu Gin Met Asn Ser Leu Arg Ala Glu Asp Thr Ala Val 100 105 110

Tyr Tyr Cys Ala Arg Pro Leu Pro Pro Phe Ala Tyr Trp Gly Gin Gly 115 120 125

Thr Leu Val Thr Val Ser Ala 130 135

(2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 106 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(l)

(D) OTHER INFORMATION: /note= "Xaa is Asp or Gin."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(2)

(D) OTHER INFORMATION: /note= "Xaa is Gin or Val."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(4,32)

(D) OTHER INFORMATION: /note= "Xaa is Met or Leu."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(29)

(D) OTHER INFORMATION: /note= "Xaa is Val, Leu or He."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(53,54)

(D) OTHER INFORMATION: /note= "Xaa is any amino acid."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(55)

(D) OTHER INFORMATION: /note= "Xaa is Ser or Thr."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(59)

(D) OTHER INFORMATION: /note= "Xaa is Ser or Ala."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(61)

(D) OTHER INFORMATION: /note= "Xaa is Phe or He." (ix) FEATURE :

(A) NAME/KEY: Region

(B) LOCATION: one-of(69)

(D) OTHER INFORMATION: /note= "Xaa is Ser or Asp."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(70)

(D) OTHER INFORMATION: /note= "Xaa is Tyr or Phe."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(72)

(D) OTHER INFORMATION: /note= "Xaa is Leu or Phe."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-θf(82)

(D) OTHER INFORMATION: /note= "Xaa is Phe or He."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(88,89)

(D) OTHER INFORMATION: /note= "Xaa is Gin, Asn or His."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(99)

(D) OTHER INFORMATION: /note= "Xaa is Gin, Gly or Ser."

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

Xaa Xaa He Xaa Thr Gin Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 1 5 10 15

Asp Arg Val Thr He Thr Cys Ser Ala Ser Ser Ser Xaa Pro Tyr Xaa 20 25 30

His Trp Tyr Gin Gin Lys Pro Gly Lys Ala Pro Lys Leu Leu He Tyr 35 40 45

Asp Thr Ser Asn Xaa Xaa Xaa Gly Val Pro Xaa Arg Xaa Ser Gly Ser 50 55 60

Gly Ser Gly Thr Xaa Xaa Thr Xaa Thr He Ser Ser Leu Gin Pro Glu 65 70 75 80

Asp Xaa Ala Thr Tyr Tyr Cys Xaa Xaa Trp Ser Ser Asp Pro Phe Thr 85 90 95

Phe Gly Xaa Gly Thr Lys Val Glu He Lys 100 105 (2) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 116 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(l)

(D) OTHER INFORMATION: /note= "Xaa is Glu, Gin or Asp."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of (34)

(D) OTHER INFORMATION: /note= "Xaa is Met, He, Val or Leu."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(49)

(D) OTHER INFORMATION: /note= "Xaa is Ala or Ser."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(61,62,63)

(D) OTHER INFORMATION: /note= "Xaa is any amino acid."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(64)

(D) OTHER INFORMATION: /note= "Xaa is Val, Ala, He, Leu or Met."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of (65)

(D) OTHER INFORMATION: /note= "Xaa is Arg, Lys or Gin."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(66)

(D) OTHER INFORMATION: /note= "Xaa is Gly, Ala, Asp, Thr or Ser."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of (75)

(D) OTHER INFORMATION: /note= "Xaa is Ser, Ala or Pro." (ix) FEATURE :

(A) NAME/KEY: Region

(B) LOCATION: one-of(78)

(D) OTHER INFORMATION: /note= "Xaa is Thr or Ser."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-θf(84)

(D) OTHER INFORMATION: /note= "Xaa is Asn or Thr."

(ix) FEATURE:

(A) NAME/KEY: Region

(B) LOCATION: one-of(116)

(D) OTHER INFORMATION: /note= "Xaa is Ala or Ser."

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:

Xaa Val Gin Leu Val Glu Ser Gly Gly Gly Leu Val Gin Pro Gly Gly 1 5 10 15

Ser Leu Arg Leu Ser Cys Ala Ala Ser Gly Phe Thr Phe Ser Ser Phe 20 25 30

Gly Xaa His Trp Val Arg Gin Ala Pro Gly Lys Gly Leu Glu Trp Val 35 40 45

Xaa Phe He Ser Ser Gly Ser Ser Thr He Tyr Tyr Xaa Xaa Xaa Xaa 50 55 60

Xaa Xaa Arg Phe He He Ser Arg Asp Asn Xaa Lys Asn Xaa Leu Tyr 65 70 75 80

Leu Gin Met Xaa Ser Leu Arg Ala Glu Asp Thr Ala Val Tyr Tyr Cys 85 90 95

Ala Arg Pro Leu Pro Pro Phe Ala Tyr Trp Gly Gin Gly Thr Leu Val 100 105 110

Thr Val Ser Xaa 115

Claims

WHAT IS CLAIMED IS:

1 1. A monoclonal antibody having a binding site that

2 specifically binds to P-selectin and to E-selectin, said

3 antibody having an affinity for each of P-selectin and

4 E-selectin of at least 10⁸ M^'1.

1 2. The antibody of claim 1, wherein

2 the specific binding of the antibody to the P-selectin

3. inhibits binding of the P-selectin to a counterreceptor of

4 P-selectin; and

5 the specific binding of the antibody to the E-selectin

6 inhibits binding of the E-selectin to a counterreceptor of

7 E-selectin.

1 3. The antibody of claim 2, wherein the

2 counterreceptors are expressed on an HL-60 cell or a

3 neutrophil.

1 4. The antibody of claim 2 that competes with

2 antibody 5C7.29, ATCC accession number CRL 11640, for specific

3 binding to P-selectin and to E-selectin.

5. The antibody of claim 2 that is a mouse antibody.

1 6. The antibody of claim 2 that is monoclonal antibody 5C7.29, ATCC accession number CRL 11640.

1 7. The antibody of claim 2 that is a Fab, Fab', F(ab')₂, Fv fragment, or a single-chain antibody.

8. The antibody of claim 2 that is a human antibody.

9. The antibody of claim 1 that does not specifically bind to L-selectin.

10. The antibody of claim 1 that specifically binds to L-selectin.

11. The antibody of claim 1 that recognizes an epitope of E-selectin comprising amino acids Q₂₁, R₂₂, Y₂₃, T₁₁₉, and A₁₂₀.

12. A humanized antibody that specifically binds to P-selectin and inhibits the binding of the P-selectin to a counterreceptor of P-selectin; and that specifically binds to E-selectin and inhibits the binding of the E-selectin to a counterreceptor of E-selectin, said antibody comprising a humanized light chain variable region and a humanized heavy chain variable region wherein (1) the humanized light chain variable region comprises complementarity determining regions having amino acid sequences from a non-human antibody light chain and comprises a variable region framework seguence substantially identical to a human light chain variable region framework seguence; and (2) the humanized heavy chain variable region comprises complementarity determining regions having amino acid sequences from a non-human antibody heavy chain, and comprises a variable region framework sequence substantially identical to a human heavy chain variable region framework seguence.

13. The humanized antibody of claim 12 wherein the humanized light chain variable region has a sequence substantially identical to the sequence: DIQMTQSPSS LSASVGDRVT ITCSASSSVP YMHWYQQKPG KAPKLLIYDT SNLASGVPSR FSGSGSGTSY TLTISSLQPE DFATYYCQQW SSDPFTFGQG TKVEIK [SEQ ID NO:6] and the humanized heavy chain variable region has a seguence substantially identical to the seguence: EVQLVESGGG LVQPGGSLRL SCAASGFTFS SFGMHWVRQA PGKGLEWVAF ISSGSSTIYY ADTVRGRFTI SRDNSKNTLY LQMNSLRAED TAVYYCARPL PPFAYWGQGT LVTVSA [SEQ ID NO:8].

14. The humanized antibody of claim 13 wherein (a) the humanized light chain variable region has the sequence: X₁IX₂X₃TQSPSS LSASVGDRVT ITCSASSSX_ι;LP YX₁₂HWYQQKPG KAPKLLIYDT SNX₁₃X₁₄X₁₅GVPX₄R X₇SGSGSGTX₅Xg TXgTISSLQPE DX₉ATYYCX₁₆X₁₇W SSDPFTFGX₁₀G TKVEIK [SEQ ID NO : 9 ] , Wherein X = D or Q; X₂ = Q or V; X₃ = M or L; X₄ = S or A; X₅ = S or D; X₆ = Y or F; X₇ = F or I; X₈ = L or F; X₉ = F, I or A; X₁₀ = Q, G or S; X ₁ = V, I or L; X₁₂ = M or L; X₁₃ = any amino acid; X₁₄ = any amino acid; X₁₅ = S or T; X₁₆ = Q, N or H; and X₁₇ = Q, N or H; and (b) the humanized heavy chain variable region has the seguence: X₃VQLVESGGG LVQPGGSLRL SCAASGFTFS SFGX₇HWVRQA PGKGLEWVX-_LF ISSGSSTIYY X₈X₉X₁₀X_ι:LX₁₂X₁₃RFTI SRDNX₄KNX₅LY LQMX₂SLRAED TAVYYCARPL PPFAYWGQGT LVTVSXg [SEQ ID NO:10]; wherein, X_x = A or S; X₂ = N or T; X₃ = E, Q or D; X₄ = s, A or P; X₅ = T or S; X₆ = A or S; X₇ = M, I, V or L; X₈ = any amino acid; X₉ = any amino acid; X₁₀ = any amino acid; X _l = V, A, I, L, M or F; X₁₂ = R, K or Q; and X₁₃ = G, A, D, T or S.

15. The humanized antibody of claim 13 wherein in the humanized light chain variable region, X_1X = V; X₁₂ = M; X₁₃ = L; X₁₄ = A; X₁₅ = S; X₁₆ = Q; and X₁₇ = Q; and wherein in the humanized heavy chain variable region, X₇ = M; X₈ = A; X₉ = D; X₁₀ = T; X₁ = V; X₁₂ = R; and X₁₃ = G.

16. The humanized antibody of claim 13 wherein the humanized light chain variable region has the sequence: DIQMTQSPSS LSASVGDRVT ITCSASSSVP YMHWYQQKPG KAPKLLIYDT SNLASGVPSR FSGSGSGTSY TLTISSLQPE DFATYYCQQW SSDPFTFGQG TKVEIK [SEQ ID NO:6] and the humanized heavy chain variable region has the sequence: EVQLVESGGG LVQPGGSLRL SCAASGFTFS SFGMHWVRQA PGKGLEWVAF ISSGSSTIYY ADTVRGRFTI SRDNSKNTLY LQMNSLRAED TAVYYCARPL PPFAYWGQGT LVTVSA [SEQ ID NO:8].

17. The humanized antibody of claim 12 further comprising light chain and heavy chain constant regions substantially identical to human light chain and heavy chain constant regions.

18. A purified nucleic acid segment encoding a light or heavy chain variable region of the antibody of claim 2.

19. A purified nucleic acid segment encoding a light or heavy chain variable region of the antibody of claim 12.

20. The purified nucleic acid segment of claim 19 further comprising a light chain or heavy chain constant region substantially identical to a human light chain or heavy chain constant region.

21. A stable cell line comprising: a nucleic acid segment encoding the heavy chain of the antibody of claim 2, the segment operably linked to a first promoter to allow expression of the heavy chain; a second nucleic acid segment encoding the light chain of the antibody of claim 2, the second segment operably linked to a second promoter to allow expression of the light chain; wherein the stable cell line can produce the antibody of claim 2.

22. A stable cell line comprising: a nucleic acid segment encoding the heavy chain of the antibody of claim 12, the segment operably linked to a first promoter to allow expression of the heavy chain; a second nucleic acid segment encoding the light chain of the antibody of claim 12, the second segment operably linked to a second promoter to allow expression of the light chain; wherein the stable cell line can produce the antibody of claim 12.

23. A pharmaceutical composition comprising the monoclonal antibody of claim 2.

24. A pharmaceutical composition comprising the monoclonal antibody of claim 12.

25. A method of treating an inflammatory disease or condition, comprising administering to a human patient a therapeutically effective dose of the pharmaceutical composition of claim 12.

26. A method according to claim 25, wherein the inflammatory disease or condition is selected from the group consisting of ischemia-reperfusion injury, adult respiratory distress syndrome, trauma, stroke, sepsis, psoriasis, and autoimmune disease.

27. The method of claim 26 wherein the inflammatory disease or condition is ischemia-reperfusion injury after myocardial infarction or stroke.

28. The method of claim 27 further comprising the step of administering a therapeutically effective dose of a thrombolytic agent.

29. A method of generating an antibody capable of blocking E-selectin and P-selectin mediated functions, the method comprising: immunizing a mammal with P-selectin; immunizing the mammal with E-selectin; immortalizing B-cells from the mammal to obtain immortalized cells producing antibodies; and selecting an immortalized cell producing an antibody that specifically binds to E-selectin and to P-selectin.

30. A method of detecting E-selectin and P-selectin bearing cells in a biological sample suspected of containing the cells, the method comprising: contacting the sample with the antibody of claim 2 to form an immune complex with the E-selectin and P-selectin bearing cells; and detecting the presence of the immune complex to indicate the presence of the cells.

31. A method of detecting E-selectin and P-selectin bearing cells in a biological sample suspected of containing the cells, the method comprising: contacting the sample with the antibody of claim 12 to form an immune complex with the E-selectin and P-selectin bearing cells; and detecting the presence of the immune complex to indicate the presence of the cells.

32. A monoclonal antibody that specifically binds to E-selectin and P-selectin, said antibody binding to the same epitope of E-selectin as antibody 5C7.29, ATCC accession number CRL 11640.

33. The monoclonal antibody of claim 32, said antibody further binding to the same epitope of P-selectin as antibody 5C7.29, ATCC accession number.