US 5,880,268 AGrant

Modulators of the Interaction Between ICAM-R and .alpha..sub.d /CD18

Issue Date:1999-03-09

•1 Claims

•34 Drawing Sheets

Abstract

DNA sequences encoding a novel human intercellular adhesion molecule polypeptide (designated "ICAM-R") and variants thereof are disclosed along with methods and materials for production of the same by recombinant procedures. Binding molecules specific for ICAM-R and variants thereof are also disclosed as useful in both the isolation of ICAM-R from natural cellular sources and the modulation of ligand/receptor binding biological activities of ICAM-R. Specifically, antibody substances which modulate the interaction between ICAM-R and ad/CD18 are provided.

Metadata

Assignee

ICOS Corporation

Inventors

W. Michael Gallatin
Rosemay Vazeux

Application Information

Application Number:US 4839323

Filing Date:1995-06-07

Priority Date:1992-01-27

Art Unit:165

Classifications

IPC:

C12P 2108C07K 1600

Field of Search:

530300;350;387.1;387.9;388.1;388.22;388.25;388.7;388.73;388.75;389.1;389.6 ;387.3

Patent Drawings (34 sheets)

Description

Field of the Invention

The present invention relates generally to cellular adhesion molecules and more particularly to the cloning and expression of DNA encoding a heretofore unknown human polypeptide designated "ICAM-R" which possesses structural relatedness to the intercellular adhesion molecules ICAM-1 and -2.

Background of the Invention

Research spanning the last decade has significantly elucidated the molecular events attending cell-cell interactions in the body, especially those events involved in the movement and activation of cells in the immune system. See generally, Springer, Nature, 346: 425-434 (1990). Cell surface proteins, and especially the so-called Cellular Adhesion Molecules ("CAMs") have correspondingly been the subject of pharmaceutical research and development having as its goal intervention in the processes of leukocyte extravasation to sites of inflammation and leukocyte movement to distinct target tissues. The isolation and characterization of cellular adhesion molecules, the cloning and expression of DNA sequences encoding such molecules, and the development of therapeutic and diagnostic agents relevant to inflammatory processes, viral infection and cancer metastasis have also been the subject of numerous U.S. and foreign applications for Letters Patent. See Edwards, Current Opinion in Therapeutic Patents, 1(11): 1617-1630 (1991) and particularly the published "patent literature references" cited therein.

Of fundamental interest to the background of the present invention are the prior identification and characterization of certain mediators of cell adhesion events, the "leukointegrins," LFA-1, MAC-1 and gp 150.95 (referred to in WHO nomenclature as CD18/CD11a, CD18/CD11b, and CD18/CD11c, respectively) which form a subfamily of heterodimeric "integrin" cell surface proteins present on B lymphocytes, T lymphocytes monocytes and granulocytes. See, e.g., Table 1 of Springer, supra, at page 429. Also of interest are other single chain adhesion molecules (CAMs) that have been implicated in leukocyte activation, adhesion, motility and the like, which are events attendant the inflammatory process. For example, it is presently believed that prior to the leukocyte extravasation which characterizes inflammatory processes, activation of integrins constitutively expressed on leukocytes occurs and is followed by a tight ligand/receptor interaction between the integrins (e.g., LFA-1) and one or both of two distinct intercellular adhesion molecules (ICAMs) designated ICAM-1 and ICAM-2 which are expressed on blood vessel endothelial cell surfaces and on other leukocytes.

Like the other CAMs characterized to date, e.g., vascular adhesion molecule (VCAM-1) as described in PCT WO 90/13300 published Nov. 15, 1990; and platelet endothelial cell adhesion molecule (PECAM-1) as described in Newman et al., Science, 247: 1219-1222 (1990) and PCT WO 91/10683 published Jul. 25, 1991!, ICAM-1 and ICAM-2 are structurally homologous to other members of the immunoglobulin gene superfamily in that the extracellular portion of each is comprised of a series of domains sharing a similar carboxy terminal motif. A "typical" immunoglobulin-like domain contains a loop structure usually anchored by a disulfide bond between two cysteines at the extremity of each loop. ICAM-1 includes five immunoglobulin-like domains; ICAM-2, which differs from ICAM-1 in terms of cell distribution, includes two such domains; PECAM-1 includes six; VCAM includes six or seven, depending on splice variations, and so on. Moreover, CAMs typically include a hydrophobic "transmembrane" region believed to participate in orientation of the molecule at the cell surface and a carboxy terminal "cytoplasmic" region. Graphic models of the operative disposition of CAMs generally show the molecule anchored in the cell membrane at the transmembrane region with the cytoplasmic "tail" extending into the cell cytoplasm and one or more immunoglobulin-like loops extending outward from the cell surface.

A variety of therapeutic uses have been projected for intercellular adhesion molecules, including uses premised on the ability of ICAM-1 to bind human rhinovirus. European Patent Application 468 257 A published Jan. 29, 1992, for example, addresses the development of multimeric configurations and forms of ICAM-1 (including full length and truncated molecular forms) proposed to have enhanced ligand/receptor binding activity, especially in binding to viruses, lymphocyte associated antigens and pathogens such as Plasmodium falciparum.

In a like manner, a variety of uses have been projected for proteins immunologically related to intercellular adhesion molecules. WO91/16928, published Nov. 14, 1991, for example, addresses humanized chimeric anti-ICAM-1 antibodies and their use in treatment of specific and non-specific inflammation, viral infection and asthma. Anti-ICAM-1 antibodies and fragments thereof are described as useful in treatment of endotoxic shock in WO92/04034, published Mar. 19, 1992. Inhibition of ICAM-1 dependent inflammatory responses with anti-ICAM-1 anti-idiotypic antibodies and antibody fragments is addressed in WO92/06119, published Apr. 16, 1992.

Despite the fundamental insights into cell adhesion phenomena which have been gained by the identification and characterization of intercellular adhesion proteins such as ICAM-1 and lymphocyte interactive integrins such as LFA-1, the picture is far from complete. It is generally believed that numerous other proteins are involved in inflammatory processes and in targeted lymphocyte movement throughout the body. Quite recently, for example, Springer and his co-workers postulated the existence of a third counter-receptor for LFA-1 de Fougerolles et al., J. Exp. Med., 174: 253-267 (1991)! and subsequently reported success in immunoprecipitating a "third" ICAM ligand, designated "ICAM-3" de Fougerolles, et al., J. Exp. Med., 175: 185-190 (1992)!. This molecule was reported to bind soluble LFA-1 and to be highly expressed by resting lymphocytes, monocytes and neutrophils. Unlike ICAM-1 and ICAM-2, however, the new ligand was not found to be expressed by endothelial cells. The immunoprecipitated product was noted to display a molecular weight of about 124,000 and to be heavily glycosylated, as revealed by a drop in apparent molecular weight to about 87,000 upon N-glyanase treatment. More recently, another research group described a cDNA sequence for a counter-receptor for LFA-1 which was also designated "ICAM-3" see Fawcett et al., Nature, 360: 481-484 (1992)!. Even more recently, two articles were published by Springer and his co-workers de Fougerolles et al., J. Exp. Med., 177: 1187-1192 (1993) and Juan et al., Eur. J. Immunol., 23: 1508-1512 (1993)! which respectively report the amino acid sequence for ICAM-3 as being identical to that of ICAM-R and note the identity of ICAM-3 to the differentiation antigen CDw50 based on patterns of immunological reactivity of antibodies specific for each protein.

There thus continues to be a need in the art for the discovery of additional proteins participating in human cell-cell interactions and especially a need for information serving to specifically identify and characterize such proteins in terms of their amino acid sequence. Moreover, to the extent that such molecules might form the basis for the development of therapeutic and diagnostic agents, it is essential that the DNA encoding them be elucidated. Such seminal information would inter alia, provide for the large scale production of the proteins, allow for the identification of cells naturally producing them, and permit the preparation of antibody substances or other novel binding proteins specifically reactive therewith and/or inhibitory of ligand/receptor binding reactions in which they are involved.

Brief Summary

In one of its aspects, the present invention provides purified and isolated polynucleotides (e.g., DNA sequences and RNA transcripts thereof, both sense and antisense strands) encoding a novel human polypeptide, "ICAM-R," as well as polypeptide variants (including fragments and analogs) thereof which display one or more ligand/receptor binding biological activities and/or immunological properties specific to ICAM-R. ICAM-R-specific ligand/receptor binding biological activities encompass interactions of both the ICAM-R extracellular and cytoplasmic domains with other molecules (e.g., in processes of cell-cell adhesion and/or signal transduction). Preferred DNA sequences of the invention include genomic and cDNA sequences as well as wholly or partially chemically synthesized DNA sequences. Biological replicas (i.e., copies of isolated DNA sequences made in vivo or in vitro) of DNA sequences of the invention are contemplated. Also provided are autonomously replicating recombinant constructions such as plasmid and viral DNA vectors incorporating ICAM-R sequences and especially vectors wherein DNA encoding ICAM-R or an ICAM-R variant is operatively linked to an endogenous or exogenous expression control DNA sequence.

According to another aspect of the invention, host cells, especially unicellular host cells such as procaryotic and eucaryotic cells, are stably transformed with DNA sequences of the invention in a manner allowing the desired polypeptides to be expressed therein. Host cells expressing such ICAM-R and ICAM-R variant products can serve a variety of useful purposes. To the extent that the expressed products are "displayed" on host cell surfaces, the cells may constitute a valuable immunogen for the development of antibody substances specifically immunoreactive with ICAM-R and ICAM-R variants. Host cells of the invention are conspicuously useful in methods for the large scale production of ICAM-R and ICAM-R variants wherein the cells are grown in a suitable culture medium and the desired polypeptide products are isolated from the cells or from the medium in which the cells are grown.

Novel ICAM-R and ICAM-R variant products of the invention may be obtained as isolates from natural cell sources, but are preferably produced by recombinant procedures involving host cells of the invention. The products may be obtained in fully or partially glycosylated, partially or wholly de-glycosylated, or non-glycosylated forms, depending on the host cell selected for recombinant production and/or post-isolation processing.

Products of the invention include monomeric and multimeric polypeptides having the sequence of amino acid residues numbered -29 through 518 as set out in SEQ ID NO: 1 herein. As explained in detail infra, this sequence includes a putative signal or leader sequence which precedes the "mature" protein sequence and spans residues -29 through -1, followed by the putative mature protein including, in order, five putative immunoglobulin-like domains (respectively spanning about residues 1 to 90, 91 to 187, 188 to 285, 286 to 387, and 388 to 456), a hydrophobic "transmembrane" region extending from about residue 457 to about residue 481 and a "cytoplasmic" region constituting the balance of the polypeptide at its carboxy terminus. Based on amino acid composition, the calculated molecular weight of the mature protein lacking glycosylation or other post-translational modification is approximately 52,417. ICAM-R variants of the invention may comprise water soluble or insoluble monomeric, multimeric or cyclic ICAM-R fragments which include all or part of one or more of the domain regions specified above and having a biological or immunological property of ICAM-R including, e.g., the ability to bind to a binding partner of ICAM-R and/or inhibit binding of ICAM-R to a natural binding partner. ICAM-R variants of the invention may also comprise polypeptide analogs wherein one or more of the specified amino acids is deleted or replaced: (1) without loss, and preferably with enhancement, of one or more biological activities or immunological characteristics specific for ICAM-R; or (2) with specific disablement of a particular ligand/receptor binding function. Analog polypeptides including additional amino acid (e.g., lysine or cysteine) residues that facilitate multimer formation are contemplated.

Also comprehended by the present invention are antibody substances (e.g., monoclonal and polyclonal antibodies, antibody fragments, single chain antibodies, chimeric antibodies, CDR-grafted antibodies and the like) and other binding proteins (e.g., polypeptides and peptides) which are specific (i.e., non-reactive with the ICAM-1 and ICAM-2 intercellular adhesion molecules to which ICAM-R is structurally related) for ICAM-R or ICAM-R variants. Antibody substances can be developed using isolated natural or recombinant ICAM-R or ICAM-R variants or cells expressing such products on their surfaces. Specifically illustrating antibodies of the present invention are the monoclonal antibodies produced by the hybridoma cell lines designated 26E3D-1, 26I18F-2, 26I10E-2, 26H11C-2 which were deposited with the American Type Culture Collection (ATCC), 12301 Parklawn Drive, Rockville, Md. 20852, on Jun. 2, 1992 as Accession Nos. HB 11053, HB 11054, HB 11055 and HB 11056, respectively, in support of U.S. Ser. No. 07/894,061; the hybridoma cell line designated 43H7C which was deposited with the ATCC on Dec. 16, 1992 as Accession No. HB 11221 and the hybridoma cell lines designated 42C5H and 42D9B which were deposited with the ATCC on Jan. 15, 1993 as Accession Nos. HB 11235 and HB 11236, respectively, in support of U.S. Ser. No. 08/009,266; the hybridoma cell lines 46D7E and 46I2H which were deposited with the ATCC on Jan. 7, 1993 as Accession Nos. HB 11232 and HB 11231, respectively, also in support of U.S. Ser. No. 08/009,266; and the hybridoma cell lines 63E11D, 63G4D, 63H4C, 63H6H, 63I1C and 63I6G which were deposited with the ATCC on Jul. 15, 1993 as Accession Nos. HB 11405, HB 11409, HB 11408, HB 11407, HB 11406 and HB 11404, respectively, in support of U.S. Ser. No. 08/102,852; and the hybridoma cell line 81K2F, which was deposited with the ATCC on Jul. 27, 1994, as Accession No. HB 11692 in support of this application. Various distinguishing properties of binding proteins of the invention are illustrated by these antibodies and are summarized in Table 11 of Example 21 herein. Such properties include the ability to modulate CD18-dependent binding (e.g., to LFA-1 and .alpha..sub.d /CD-18) and CD18-independent binding (e.g., to VLA-4) of ICAM-R to cells and cell surface molecules as well as the ability to modulate lymphocyte activation by SEA and/or alloantigen. Binding proteins of the invention are additionally susceptible to characterization in terms of binding site structure (e.g., epitopes and/or sensitivity of binding properties to modifications in ICAM-R amino acid sequence).

Binding proteins are useful, in turn, in compositions for immunization as well as for purifying polypeptides of the invention and identifying cells displaying the polypeptides on their surfaces. They are also manifestly useful in modulating (i.e., blocking, inhibiting or stimulating) ligand/receptor binding biological activities involving ICAM-R, especially those ICAM-R effector functions involved in specific and non-specific immune system responses. Anti-idiotypic antibodies specific for anti-ICAM-R antibody substances and uses of such anti-idiotypic antibody substances in modulating immune responses are also contemplated. Assays for the detection and quantification of ICAM-R on cell surfaces and in fluids such as serum may involve, for example, a single antibody substance or multiple antibody substances in a "sandwich" assay format.

The scientific value of the information contributed through the disclosures of DNA and amino acid sequences of the present invention is manifest. As one series of examples, knowledge of the sequence of a cDNA for ICAM-R makes possible the isolation by DNA/DNA hybridization of genomic DNA sequences encoding ICAM-R and specifying ICAM-R expression control regulatory sequences such as promoters, operators and the like. DNA/DNA hybridization procedures carried out with DNA sequences of the invention and under stringent conditions are likewise expected to allow the isolation of DNAs encoding allelic variants of ICAM-R, other structurally related proteins sharing one or more of the biological and/or immunological properties specific to ICAM-R, and non-human species (e.g., rodent) proteins homologous to ICAM-R. DNAs of the invention are useful in DNA/RNA hybridization assays to detect the capacity of cells to synthesize ICAM-R. Also made available by the invention are anti-sense polynucleotides relevant to regulating expression of ICAM-R by those cells which ordinarily express the same. As another series of examples, knowledge of the DNA and amino acid sequences of ICAM-R makes possible the generation by recombinant means of ICAM-R variants such as hybrid fusion proteins (sometimes referred to as "immunoadhesions") characterized by the presence of ICAM-R protein sequences and immunoglobulin heavy chain constant regions and/or hinge regions. See, Capon et al., Nature, 337: 525-531 (1989); Ashkenazi et al., F.N.A.S. (USA), 88: 10535-10539 (1991); and PCT WO 89/02922, published Apr. 6, 1989. ICAM-R variant fusion proteins may also include, for example, selected extracellular domains of ICAM-R and portions of other cell adhesion molecules.

The DNA and amino acid sequence information provided by the present invention also makes possible the systematic analysis of the structure and function of ICAM-R and definition of those molecules with which it will interact on extracellular and intracellular levels. The idiotypes of anti-ICAM-R monoclonal antibodies of the invention are representative of such molecules and may mimic natural binding proteins (e.g., peptides and polypeptides) through which ICAM-R intercellular and intracellular activities are modulated or by which ICAM-R modulates intercellular and intracellular events. Alternately, they may represent new classes of modulators of ICAM-R activities. Anti-idiotypic antibodies, in turn, may represent new classes of biologically active ICAM-R equivalents.

In vitro assays for identifying antibodies or other compounds that modulate the activity of ICAM-R may involve, for example, immobilizing ICAM-R or a natural ligand to which ICAM-R binds, detectably labelling the nonimmobilized binding partner, incubating the binding partners together and determining the effect of a test compound on the amount of label bound wherein a reduction in the label bound in the presence of the test compound compared to the amount of label bound in the absence of the test compound indicates that the test agent is an inhibitor of ICAM-R binding.

Another type of assay for identifying compounds that modulate the interaction between ICAM-R and a ligand involves immobilizing ICAM-R or a fragment thereof on a solid support coated (or impregnated with) a fluorescent agent, labelling the ligand with a compound capable of exciting the fluorescent agent, contacting the immobilized ICAM-R with the labelled ligand in the presence and absence of a putative modulator compound, detecting light emission by the fluorescent agent, and identifying modulating compounds as those compounds that affect the emission of light by the flourescent agent in comparison to the emission of light by the fluorescent agent in the absence of a modulating compound. Alternatively, the ICAM-R ligand may be immobilized and ICAM-R may be labelled in the assay.

Yet another method contemplated by the invention for identifying compounds that modulate the interaction between ICAM-R and a ligand involves transforming or transfecting appropriate host cells with a DNA construct comprising a reporter gene under the control of a promoter regulated by a transcription factor having a DNA-binding domain and an activating domain, expressing in the host cells a first hybrid DNA sequence encoding a first fusion of part or all of ICAM-R and either the DNA binding domain or the activating domain of the transcription factor, expressing in the host cells a second hybrid DNA sequence encoding part or all of the ligand and the DNA binding domain or activating domain of the transcription factor which is not incorporated in the first fusion, evaluating the effect of a putative modulating compound on the interaction between ICAM-R and the ligand by detecting binding of the ligand to ICAM-R in a particular host cell by measuring the production of reporter gene product in the host cell in the presence or absence of the putative modulator, and identifying modulating compounds as those compounds altering production of the reported gene product in comparison to production of the reporter gene product in the absence of the modulating compound. Presently preferred for use in the assay are the ADHI promoter, the lexA DNA-binding domain, the GAL4 transactivation domain, the lacZ reporter gene, and yeast host cells.

A modified version of the foregoing assay may be used in isolating a polynucleotide encoding a protein that binds to ICAM-R by transforming or transfecting appropriate host cells with a DNA construct comprising a reporter gene under the control of a promoter regulated by a transcription factor having a DNA-binding domain and an activating domain, expressing in the host cells a first hybrid DNA sequence encoding a first fusion of part or all of ICAM-R and either the DNA binding domain or the activating domain of the transcription factor, expressing in the host cells a library of second hybrid DNA sequences encoding second fusions of part or all of putative ICAM-R binding proteins and the DNA binding domain or activating domain of the transcription factor which is not incorporated in the first fusion, detecting binding of an ICAM-R binding protein to ICAM-R in a particular host cell by detecting the production of reporter gene product in the host cell, and isolating second hybrid DNA sequences encoding ICAM-R binding protein from the particular host cell.

The DNA sequence information provided by the present invention also makes possible the development, by homologous recombination or "knockout" strategies see, e.g., Kapecchi, Science, 244: 1288-1292 (1989)!, of rodents that fail to express a functional ICAM-R protein or that express a variant ICAM-R protein. Such rodents are useful as models for studying the activities of ICAM-R and ICAM-R modulators in vivo.

Modulators which affect the interaction between ICAM-R and LFA-1, .alpha..sub.d /CD18, VLA-4, tubulin, and the 14.3.3 family of proteins are specifically contemplated as useful therapeutic compounds.

Inflammatory conditions which may be treated or monitored with ICAM-R related products of the invention include conditions resulting from a response of the non-specific immune system in a mammal (e.g., adult respiratory distress syndrome, multiple organ injury syndrome secondary to septicemia, multiple organ injury syndrome secondary to trauma, reperfusion injury of tissue, acute glomerulonephritis, reactive arthritis, dermatosis with acute inflammatory components, stroke, thermal injury, hemodialysis, leukapheresis, ulcerative colitis, Crohn's disease, necrotizing enterocolitis, granulocyte transfusion associated syndrome, atherosclerosis and cytokine-induced toxicity) and conditions resulting from a response of the specific immune system in a mammal (e.g., psoriasis, organ/tissue transplant rejection and autoimmune diseases including Raynaud's syndrome, autoimmune thyroiditis, EAE, multiple sclerosis, rheumatoid arthritis, diabetes, and lupus erythematosus). ICAM-R products of the invention may also be useful in monitoring and treating asthma, tumor growth and/or metastasis, and viral infection (e.g., HIV infection).

In particular, disease processes in which T cell activation plays a central and essential triggering role may be impacted beneficially by ICAM-R related products of the invention described herein. The therapeutic use of ICAM-R analogs incorporating specific amino acid substitutions (e.g., analogs E37T or D231H) chosen to enhance or diminish their specific immunomodulatory properties are useful in this regard. Specific examples of T cell dependent diseases for which ICAM-R related products may have utility include but are not limited to asthma, psoriasis, diabetes, graft vs. host disease, tissue transplant rejection, and multiple sclerosis. The use of products of the invention to modulate diseases wherein macrophages play a central generative role is also indicated. Moreover, monoclonal antibodies specific to ICAM-R may be used therapeutically either on their own or when conjugated to other moieties (e.g., toxins, radionuclides) to therapeutically target and/or detect the presence of neovascularizing sites.

Brief Description of the Drawing

Numerous other aspects and advantages of the present invention will be apparent upon consideration of the following detailed description thereof, reference being made to the drawing wherein:

FIG. 1(A through G) depicts an isolated cDNA clone insert (SEQ ID NO: 2) derived from HL60 cells encoding ICAM-R and the deduced amino acid sequence (SEQ ID NO: 1) of an open reading frame therein;

FIG. 2(A through B) comprises bar graphs illustrating the results of Northern blot hybridization of transfected L cells using ICAM-R and ICAM-1 DNA probes;

FIG. 3(A through F) presents photomicrographs depicting the results of in situ hybridizations of transfected L cells using ICAM-R or ICAM-1 RNA probes;

FIG. 4A comprises bar graphs illustrating the results of assays for the adhesion of PMA-stimulated or unstimulated lymphoblastoid cells from patients with leukocyte adhesion deficiency to soluble ICAM-R in the presence and absence of anti-CD18 antibody, while FIG. 4B comprises bar graphs illustrating the results of assays for the adhesion of various other PMA-stimulated or unstimulated cell lines to soluble ICAM-R in the presence and absence of anti-CD18 or anti-CD11a antibody;

FIG. 5 illustrates in histogram format the results of FACS analyses of indirect immunofluorescence staining of transfected L cells using monoclonal antibodies specific for ICAM-R, ICAM-1 or ICAM-2;

FIG. 6 is a diagram of three chimeric ICAM-R proteins utilized to map epitopes of anti-ICAM-R monoclonal antibodies of the invention;

FIG. 7(A through B) presents bar graphs depicting the results of actin-normalized Northern blot hybridization of human leukocyte cell lines and umbilical cord endothelial cells using ICAM-R or ICAM-1 DNA probes;

FIG. 8(A through B) comprises photographs of Western blots of immunoprecipitations of lysates from human cells lines using ICAM-R specific monoclonal antibodies;

FIG. 9(A through G) presents photomicrographs of immunohistologic staining of various human tissues with an anti-ICAM-R monoclonal antibody;

FIG. 10 is a bar graph depicting the effects of anti-ICAM-R monoclonal antibodies on the stimulation of lymphocyte proliferation by anti-CD3 antibodies;

FIG. 11(A through B) comprises bar graphs illustrating the effects of anti-ICAM-R monoclonal antibodies on superantigen-induced proliferation of human peripheral blood lymphocytes, while FIG. 11C is a graph comprising logistic dose response curves of the effects of anti-ICAM-R monoclonal antibodies on superantigen-induced proliferation of human peripheral blood lymphocytes;

FIG. 12 is a bar graph depicting the effects of anti-ICAM-R monoclonal antibodies on alloantigen-induced T-cell proliferation; and

FIG. 13 is a bar graph illustrating the effect of anti-ICAM-R monoclonal antibodies on superantigen-induced proliferation of "memory" T cells;

FIG. 14 comprises a bar graph depicting the effect of anti-ICAM-R monoclonal antibodies on superantigen-induced proliferation of "resting" T cells;

FIG. 15 comprises a bar graph illustrating that crosslinking distinct ICAM-R epitopes differentially affects ICAM-R association with the cytoskeleton.

FIG. 16 is a schematic depiction of the three-dimensional structure of the extracellular domain 1 of ICAM-R.

Detailed Description

The present invention is illustrated by the following examples relating to the isolation of a full length cDNA clone encoding ICAM-R from a cDNA library derived from human HL60 promyelocytic cells (ATCC CCL 240) and to the expression of ICAM-R DNA in L cells. More particularly, Example 1 addresses the design and construction of oligonucleotide probes for PCR amplification of ICAM related DNAs. Example 2 addresses the use of the probes to amplify a genomic DNA fragment homologous to, but distinct from, DNAs encoding ICAM-1 and ICAM-2. Example 3 treats the screening of cDNA libraries with the genomic fragment to isolate additional ICAM-R coding sequences. Example 4 refers to the further screening of cDNA libraries to isolate a full length human cDNA encoding ICAM-R. Example 5 provides a characterization of DNA and amino acid sequence information for ICAM-R, relates the structures thereof to ICAM-1 and ICAM-2, describes the chromosomal localization of the ICAM-R gene and describes the isolation of human ICAM-R genomic sequences. Example 6 relates to the development of mammalian host cells expressing ICAM-R. Example 7 describes preliminary experiments indicative of ICAM-R participation in intercellular adhesion events involving CD18-dependent and CD18-independent pathways. Example 8 presents experiments illustrating inhibition of cell adhesion to ICAM-R by ICAM-R derived peptides. Example 9 relates to the construction and expression of a soluble variant of human ICAM-R and various assays useful for identifying ICAM-R ligands and modulators of ICAM-R activities. Example 10 describes the construction and expression of ICAM-R variants having point mutations in their extracellular domains. Example 11 describes the preparation and preliminary characterization of anti-ICAM-R antibodies and the preparation of Fab' fragments thereof. Example 12 relates to assays determining the capability of ICAM-R specific monoclonal antibodies to inhibit binding of CD18.sup.+ cells to recombinant soluble human ICAM-R. Example 13 details the humanization of ICAM-R specific monoclonal antibodies of the invention. Example 14 relates to mapping of the ICAM-R epitopes recognized by the anti-ICAM-R monoclonal antibodies of the invention. Examples 15, 16, 17 and 18 relate to assessment of the distribution and biochemical characterization of ICAM-R polypeptide and RNA encoding the same in normal cells and tissues as well as in various cell lines. Example 19 describes assays for the involvement of ICAM-R in homotypic cell-cell adhesion. Example 20 addresses experiments indicating that ICAM-R is involved in immune cell activation/proliferation. Example 21 comprises a summary of characteristics of ICAM-R specific monoclonal antibodies of the invention. Example 22 describes experiments showing differential phosphorylation of and cytoskeletal associations with the cytoplasmic domain of ICAM-R. Examples 23 and 24 set out experiments characterizing the interaction between ICAM-R and various cytoplasmic ligands utilizing dihybrid screening techniques. Example 25 describes the interaction between ICAM-R and LFA-1 while Examples 26 and 27 describe the interaction between ICAM-R and .alpha..sub.d /CD18 and ICAM-R and VLA-4, respectively. Example 28 provides evidence that elevated levels of soluble ICAM-R are observed in human serum various immune-mediated diseases. Example 29 describes various therapeutic applications of subject matter of the invention.

Example 1

Nucleic acid and amino acid alignments of individual sets of CAMs (e.g., ICAM-1 and ICAM-2) did not manifest sufficient conservation between molecules to yield information useful in the design of consensus-type probes for isolating related novel genes. The strategic focus of attempts to isolate unknown DNAs encoding cellular adhesion molecules therefore involved the development of degenerate consensus oligonucleotides representing putative spaced apart DNA sequences of various known molecules and the use of these oligonucleotides as primers for polymerase chain reaction (PCR) amplification of DNA replicas of intermediate gene sequences which resemble, but are not identical to, the known DNAs. The starting point for oligonucleotide primer design was the notation that the amino acids in regions surrounding cysteines which form immunoglobulin-like loops of certain CAMs are somewhat conserved. At the amino terminal side of the motif, the sequence:

SEQ ID NO: 3

G-X-X-(V or L or I)-X-(V or L or I)-X-C

is found, while at the carboxy terminal side of the motif, the sequence:

SEQ ID NO: 4

N-X-G-X-Y-X-C-X-(V or A)

is typical. See Hunkapiller et al., Nature, 323: 15-16 (1986); Williams et al, Ann. Rev. Immunol., 6: 381-405 (1988); and Newman et al, supra.! In and of themselves the two amino acid motifs are much too general and do not allow the construction of degenerate sets of oligonucleotides useful as probes for unknown DNAs which might share the motif. In an attempt to solve this problem, each individual CAM sequence was split into a domain of sub files defined by the cysteine motif termini described above. Subfiles were generated for each of the seven domains of human vascular adhesion molecule (VCAM-1), the six domains of human platelet endothelial cell adhesion molecule (PECAM-1), the five domains of ICAM-1, the two domains of ICAM-2, three of the four domains of both human myeloglobin-related glycoprotein and human fibroblast growth factor receptor, and the five domains of mouse neural cell adhesion molecule (NCAM). All the subfiles were pooled and segregated independently from the CAM of origin using a multialignment homology computer algorithm designated "Multalin" Corpet, Nucleic Acids Research, 16(22): 10881-10890 (1988)! providing a tree of alignment allowing the ascertainment of consensus sequences around cysteine motifs. A consensus sequence representing the amino terminal cysteine motif was determined to be:

SEQ ID NO: 5

G-K-(N or S)-(L or F)-T-(L or I)-(R or E)-C

while the carboxy terminal consensus sequence was determined to be:

SEQ ID NO: 6

(D or E)-(H or D)-(H or G)-(G or H)-(A or R)-N-F-S-C.

Employing human preferences for codon usage to partially eliminate degeneracy, three separate sets of degenerate oligonucleotides totalling 1152 probes were generated for use as top strand PCR primers for amplification from a putative amino terminus of the motif. The specific degenerate sequences of the three pools are set out below in IUPAC nomenclature.

SEQ ID NO: 7

ATTCTGCAGGCAARAAYCTSACHMTBMGSTG

SEQ ID NO: 8

ATTCTGCAGGCAARAGYTTYACHMTBGARTG

SEQ ID NO: 9

ATTCTGCAGGCAARTCYTTYACHMTBGARTG

Each of the primers included a PstI restriction endonuclease recognition site (CTGCAG) to facilitate cloning of amplified products.

A total of 768 probes were designed as bottom strand primers as set out below in IUPAC nomenclature for amplification from a putative carboxy terminus of the motif. Each of these primers included an XbaI recognition site (TCTAGA) to facilitate cloning of amplified products.

SEQ ID NO: 10

ATTTCTAGARAARTTRGCSCCRTGRTSRTC

SEQ ID NO: 11

ATTTCTAGARAARTTSCKRTGSCCRTSKTC

Oligonucleotides were synthesized with an automated Applied Biosystems, Inc. (Foster City, Calif.) Model 394 DNA synthesizer using an 0.2 micromolar scale synthesis program and employing beta-cyanoethyl chemistry. Protective groups were then removed by heating at 55.degree. C. for in excess of six hours. Oligonucleotides were then lyophilized to dryness, rehydrated in TE (10 mM Tris, pH 7.0, 1 mm EDTA) and desalted in TE by size exclusion chromatography with G25-150 Sephadex.

Example 2

The two sets of probes whose design and synthesis are described in Example 1 were employed in PCR amplification procedures applied to a human genomic DNA template. Briefly put, PCR-generated fragments of a size similar to that of the immunoglobulin-like loop regions of ICAM-1 and ICAM-2 were isolated, subcloned into Bluescript plasmid (Stratagene, La Jolla, Calif.) and screened both directly by sequencing and hybridization in arrays for homology to ICAM-2 DNA. Approximately 50% of the fragments were identical to ICAM-1 or ICAM-2 (except, of course, in the regions of the degenerate primer). One subclone, designated 13-3C7, was found to have an open reading frame homologous to ICAM-1 and ICAM-2 in the region of their respective second domains. It did not correspond to any known sequence present in the Genbank data base. The specific manipulations leading up to the isolation of subclone 13-3C7 were as follows.

The degenerate oligonucleotides were mixed to a final concentration of 10 .mu.g/ml in a PCR reaction to amplify human genomic DNA obtained either from peripheral blood leukocytes or Hela cells. The DNA amplification was performed in PCR buffer (2 mM MgCl.sub.2, 25 mM KCl, 10 mM Tris pH 8.3) with 2 mM deoxynucleotides. After a 94.degree. C. denaturation for 4 minutes, 30 PCR cycles were performed with annealing at 60.degree. C. for 2 minutes, elongation at 72.degree. C. for 4 minutes and denaturation at 94.degree. C. for 1 minute. A DNA band migrating at about 0.2 kb was extracted from a 6% polyacrylamide gel by electroelution, digested by XbaI and Pst 1 restriction enzymes, and ligated into the Bluescript vector (Stratagene). The plasmid was electroporated into XL 1-blue strains of E.coli (Stratagene) and colonies were selected on X-gal IPTG, carbenicillin agarose plates. Single strand templates were obtained from 6 white colonies after addition of M13K07 helper phage (Stratagene), carbenicillin, and kanamycin to a 2 ml culture of each colony. For sequence analysis, the single strand templates were then sequenced using the Sanger method both by DNA automatic sequencing (Applied Biosystems) and with a sequenase kit (UCB, Belgium). Four sequences (clones 1.1, 1.3, 1.4, 1.6) were obtained which were 184-185 base pairs (bp) long and were 92-95% homologous to the second domain of ICAM-2. In addition, a 182 bp long DNA sequence (clone 1.5) was obtained which contained a frameshift in the open reading frame of an ICAM-1-like domain along with a 66 bp DNA (clone 1.2) corresponding to a truncated immunoglobulin-like domain.

The sequence of clones 1.6, 1.5, 1.2 was used to design three oligonucleotide probes (RM16, RM15, RM12) that were used in subsequent tests to eliminate from further consideration additional colonies containing cDNAs that were highly homologous to the previous isolated clones. The sequences of probes RM16, RM15 and RM12 are set out below.

Probe RM16 (SEQ ID NO: 12)

GAGACTCTGCACTATGAGACCTTCG

Probe RM15 (SEQ ID NO: 13)

CAGGTGATTCTCATGCAGAGTCCAGG

Probe RM12 (SEQ ID NO: 14)

CCGACATGCTGGTAAGTGTGTCCAA

In a second round of tests, new colonies were obtained from the original PCR products that had been XbaI and Pst 1 digested and from additional PCR products that had been rendered blunt-ended by treatment with the Klenow fragment of polymerase I and subcloned by blunt-end ligation. The colonies containing the vector with an insert were selected on carbenicillin L broth agarose plates containing X-gal and IPTG. Single strand templates were then synthesized in 96-well plates by growing individual white colonies in 300 .mu.l L broth, to which was added M13K07 phage, carbenicillin and kanamycin. Ten .mu.l of each template was transferred with a pronging device to a nylon membrane, denatured and fixed with UV light. (Ten .mu.l of each template were transferred to three different nylon membranes for each 96-well plate.) Oligonucleotides RM16, RM15, RM12 were labelled by phosphorylation using .lambda.-.sup.32 P!ATP. The nylon membranes were pre-hybridized in 20% formamide, 5.times.SSC, 5.times.Denhardt's solution and 0.5% SDS for 3 hours at 42.degree. C. then hybridized overnight with the different radiolabelled oligonucleotide probes under the same conditions. The membranes were then washed in 0.2.times.SSC, 0.5% SDS three times for 15 minutes each at room temperature then washed in the same buffer at 37.degree. C. for 15 minutes, rinsed in 2.times.SSC and exposed. Each template that did not hybridize with either of the three oligonucleotide probes was further sequenced using the Sanger technique by DNA automatic sequencing and by sequenase kit. Using this technique, the 170 bp DNA sequence of a clone designated 13-3C7 was determined.

Example 3

The cDNA insert of subclone 13-3C7 isolated in Example 2 was used as a hybridization probe to screen four different lambda phage cDNA libraries prepared from human spleen, human placenta (two libraries) and human leukocyte cell line U937 (ATCC CRL 1593). Briefly summarized, one hundred and twenty positive clones were picked (from among the approximately 1.6 million clones screened), subcloned, rescreened with the 13-3C7 probe, and the rescreening positive were size selected for inserts of greater than approximately 500 bp by analytical PCR with primers corresponding to the plasmid DNA flanking the insertion for DNAs. A 1.3 kb clone derived from U937 cDNA, designated clone 19C, was sequenced and revealed DNA regions encoding two immunoglobulin-like domains separated by what appeared to be an intervening sequence (intron) resulting from improper or incomplete mRNA splicing prior to cDNA formation. The two regions displayed significant homology, but overall distinctness, in comparison to domains 2 and 3 of ICAM-1 and less homology to domains 1 and 2 of ICAM-2.

The specific procedures leading up to isolation of clone 19C were as follows. The four libraries were constructed in lambda gt10 phage (.lambda.gt10) using cDNA obtained from the U937 cell line, from the spleen of a patient with chronic myelomonocytic leukemia and from human placenta. Exact match oligonucleotides designated 1 Hr-5' and 1 Hr-3' were designed corresponding to the 5' and 3' sides of the domain-like region of subclone 13-3C7 (including bases attributable to incorporation of the original degenerate primer). The sequences of the 1 Hr-5' and 1 Hr-3' oligonucleotide primers are set out below.

Primer 1 Hr-5' (SEQ ID NO: 15)

GACCATGAGGTGCCAAG

Primer 1 Hr-3' (SEQ ID NO: 16)

ATGGTCGTCTCTGCTGG

Using these oligonucleotides in a PCR reaction with the 13-3C7 insert template and .sup.32 P-dCTP, a 148 bp long DNA probe was generated. The cDNA libraries were plated and transferred to nylon membranes. The membranes were pre-hybridized in 40% formamide, 5.times.SSC, 5.times.Denhardt's, 0.5% SDS at 42.degree. C. for at least 15 minutes, then hybridized overnight with the probe in the same buffer at 42.degree. C. The membranes were washed several times at room temperature in 2.times.SSC and exposed. Most of the phage plaques that hybridized with the probe were derived from the U937 cDNA library. These phages were further purified and tested by PCR (using 1 Hr-5' and 1 Hr-3' as primers) for the presence of the domain inside the cDNA clones. The phage were also tested by PCR to determine the length of the clones and the location of the domain within the cDNA fragment (using a combination of 13-3C7 specific primers and primers homologous to flanking .lambda.gt10 vector sequences). Two clones were selected. Clone 1F was 0.7 kb long and clone 19C was 1.3 kb long. These cDNAs were digested with EcoRI and subcloned in the Bluescript vector. In addition, the largest cDNA (clone 19C) was sonicated to obtain small pieces which were subcloned into Bluescript for sequencing. By homology with the ICAM-1 molecule, clone 19C cDNA contains 2 regions having homology to domains 2 and 3 of ICAM-1, respectively, with an intervening sequence of unrelated DNA. Hereinafter, these DNA regions are referred to as domains 2 and 3 of ICAM-R.

Example 4

The 1.3 kb (clone 19C) DNA isolated in Example 3 and having regions encoding immunoglobulin-like loops resembling domains 2 and 3 of ICAM-1 was then employed to generate a probe for the screening of additional cDNA libraries in an attempt to isolate a full length cDNA clone. Briefly, the domain 2 and 3 regions within clone 19C were each amplified by PCR using unique probes designated to match respective amino (5') and carboxy (3') terminal portions of the domains. These amplified DNAs, in turn, provided probes for screening of cDNA libraries derived from: (1) the HL60 myelomonocytic cell line; (2) lipopolysaccharide-activated human monocytes; (3) HUT-78 T-cells (ATCC T1B161); and (4) activated peripheral blood leukocytes. The latter two libraries yielded no positive upon rescreening. Positives derived from HL60 and monocyte cDNA libraries were then screened with a probe representing domain 2 of ICAM-1 DNA (GenBank, Accession No. 22634) in order to eliminate ICAM-1 clones. A single phagmid clone derived from lambda 345 and designated pVZ-147, repeatedly tested positive for hybridization with the probe(s) based on the DNA isolated in Example 4 and negative for hybridization with the ICAM-1 DNA probe. The approximately 1.7 kb insert from clone pVZ-147 was isolated and sequenced to provide the 1781 bp sequence set out in SEQ ID NO: 2. The deduced amino acid sequence of the polypeptide encoded by this DNA is set out in SEQ ID NO: 1. The polypeptide was designated "ICAM-R" on the basis of its structural relatedness to ICAM-1 and ICAM-2. The DNA and deduced amino acid sequences of ICAM-R were published after the priority dates of this application in Vazeux et al., Nature, 360: 485-488 (1992). The open reading frame of the DNA sequence of ICAM-3 published after the priority dates of this application in Fawcett et al., supra, differs at two nucleotide positions from the coding region of the DNA sequence of ICAM-R presented in FIG. 1(A through G) herein. (See nucleotide positions 194 and 1275.)

The specific manipulations involved in the isolation of lambda phage clone pVZ147 are as follows. All cDNA libraries were constructed in .lambda.gt10 except for the HL60 library which cloned into phage lambda 345. Oligonucleotides for use in library screening and rescreening had the following sequences.

Probe IHr2-5' (SEQ ID NO: 17)

TTCACCCTGCGCTGCCAA

Probe IHr2-3' (SEQ ID NO: 18)

AAAGGGGCTCCGTGGTCG

Probe IHr 3-5' (SEQ ID NO: 19)

CCGGTTCTTGGAGGTGGAA

Probe IHr 3-3' (SEQ ID NO: 20)

CATGACTGTCGCATTCAGCA

Probe Icam 1-5 (SEQ ID NO: 21)

GCAAGAACCTTACCCTAC

Probe Icam 1-3 (SEQ ID NO: 22)

GAAATTGGCTCCATGGTGA

Probes IHr 2-5' and IHr 2-3' were employed in a PCR amplification using .sup.32 P-dCTP on the clone 19C template to generate a domain 2 specific probe for cDNA screening. Likewise, probes IHr 3-5' and IHr 3-3' were employed to generate a domain 3 specific probe. Finally, probes Icam 1-5 and Icam 1-3 were employed to amplify an ICAM-1 segment probe corresponding to bases 440 through 609 of the ICAM-1 cDNA sequence (GenBank, Accession No. 22634), i.e., the ICAM-1 second domain.

The cDNA libraries were plated, transferred on nylon membranes, hybridized with the domain 2 probe (derived from clone 19C) in 40% formamide, 5.times.SSC, 5.times.Denhardt, 0.5% SDS and washed as described above. All the plaques that hybridized with the domain 2 probe were derived from the monocyte and HL60 libraries. These phage plaques were purified by dilution, plating, transfer and hybridization with the domain 2 probe. To further characterize the cDNA clones, each plaque that had hybridized with the domain 2 probe was grown on an array in triplicate, transferred to a nylon membrane and hybridized under higher stringency conditions (50% formamide, 5.times.SSC, 5.times.Denhardt, 0.5% SDS) with three different probes: the domain 2 probe; the domain 3 probe, and the ICAM-1 second domain probe. Five clones were found in the HL60 library and 2 clones in the monocyte library which hybridized with both domain 2 and domain 3 probes and not with the ICAM-1 second domain probe. A sixth clone from the HL60 library hybridized only with domain 2 probe and did not hybridize with either domain 3 or with ICAM-1 second domain. The cDNAs of the 6 clones from the HL60 library were further analyzed. The phages were tested by PCR for the presence of properly spliced cDNA using oligonucleotide primers corresponding to the 5' extremity (IHr2-5') of domain 2 and to the 3' extremity (IHr3-3') of domain 3. The clones were also tested by PCR for length and location of the domains inside the clones. The cDNA plasmids were extracted and cyclized from phage lambda 345 by digestion with SfiI and self-ligation. To facilitate making single strand templates and sequencing in both orientations, each cDNA was also subcloned in Bluescript SK+ vector (Stratagene). Plasmid pVZ147 was determined to include the entire ICAM-R coding sequence in a single open reading frame.

Example 5

A. Characterization of the ICAM-R Polypeptide

FIG. 1(A through G) graphically illustrates the sequence of the human cDNA insert of the lambda phage clone pVZ 147 isolated in Example 4, above. The total of 1781 bp shown are as set out in SEQ ID NO: 2. The deduced amino acid sequence of the ICAM-R polypeptide as set out in SEQ ID NO: 1 is graphically subdivided in FIG. 1(A through G) into the following regions:

(1) A putative signal or leader sequence is illustrated preceding the sequence of the "mature" protein and spanning amino acids designated -29 through -1. Determination of whether the translation product is actually initiated at -29 or -26 will be provided by amino acid sequencing of intercellular expression products. The designation of the first residue of the mature protein was based on generalized analogy to amino acids (and corresponding bases) for residues of secreted human proteins in the region of the junction of the mature protein and leader sequences. Confirmation of the actual initial residue of the mature protein awaits sequencing of a secreted recombinant product or, e.g., an immunopurified natural product.

(2) Within the mature protein spanning residues +1 through 518, five putative immunoglobulin-like loop regions are shown (white on black) bounded by cysteines within the five putative immunoglobulin-like domains (shown in boxes). Note that in the first domain (residues 1 through 91), cysteine residues potentially significant to loop formation are present at positions 24, 28, 67 and 71. Each of the remaining putative loops has a single relevant cysteine at each of its ends.

(3) Also within the mature protein, a putative hydrophobic "transmembrane" region is illustrated with dashes connecting residues 457 through 481 which follow the fifth immunoglobulin-like domain. A putative carboxy terminal "cytoplasmic" region constitutes residues 482 through 518.

(4) Potential N-linked glycosylation sites characterized by the consensus sequence, Aspargine-X-(Serine or Threonine)! are indicated with an asterisk. Potential O-linked glycosylation sites occur at any serine or threonine residue.

A comparison was made between the amino acid sequence (SEQ ID NO: 1) of ICAM-R and the published 537 residue amino acid sequence of ICAM-1 (GenBank Accession No. 22634; cf, FIG. 8 of European Patent Application 0 289 949 published Nov. 11, 1988). This comparison revealed 249 matches within the aligned 537 residues, indicating an overall amino acid identity of 48% between the two polypeptides. The highest percentage of matches was noted to be present between domains 2 and 3 of ICAM-1 and putative domains 2 and 3 of ICAM-R. Likewise the alignment of SEQ ID NO: 1 with the published 295 residues of the amino acid sequence of ICAM-2 (GenBank accession No. 22635; cf, FIG. 2 of European Patent Application 0 387 668 published Sep. 19, 1990) revealed 78 matches among the 282 aligned residues, for a 27% overall identity of amino acids in one possible alignment. The cytoplasmic domain of ICAM-R was found to be 20% identical to the cytoplasmic domain of ICAM-1 and 34% identical to the cytoplasmic domain of ICAM-2 in one possible alignment.

B. Characterization of ICAM-R DNA

A comparative alignment of the human ICAM-R DNA sequence (SEQ ID NO: 2) was made with the published DNA sequences of ICAM-1 and ICAM-2, supra. A total of 677 matches were noted among the 1623 aligned bases of ICAM-R and ICAM-1 providing an overall identity of 41%. A 42% identity (484 matches) between the aligned 1136 bases of ICAM-R and ICAM-2 DNAs was noted.

Reference points in the FIG. 1 (A through G) DNA having "historical" significance to the isolation of the ICAM-R gene include the following:

(a) bases 420 through 567 correspond to the subclone 13-3C7 isolated in Example 2;

(b) bases 373 through 663 correspond to the immunoglobulin-like domain 2 localized in clone 19C of Example 3 (with bases 418 through 435 and 561 through 578, respectively corresponding to probes IHr2-5' and IHr2-3' employed for PCR amplification of domain 2 to provide one of the oligonucleotide probes for use in Example 4); and

(c) bases 664 through 957 correspond to the immunoglobulin-like domain 3 localized on clone 19C of Example 3 (with bases 699 through 717 and 800 through 819, respectively corresponding to probes IHr3-5' and IHr3-3' employed for PCR amplification of domain 3 to provide another oligonucleotide probe for use in Example 4.

C. Chromosomal Localization of Sequences Encoding Human ICAM-R

An ICAM-R specific DNA probe was utilized in the methods described in Cannizzaro et al., Cancer Res., 51: 3818-3820 (1991) to determine that the human ICAM-R encoding sequences are located on chromosome 19 with primary localization to the short (p) arm region.

D. Cloning of Genomic ICAM-R Sequences

Human ICAM-1 and -R have been mapped to the same region of chromosome 19. Therefore, the human P1 Genomic library of Genome Systems Inc. (St. Louis, Mo.) was screened with human ICAM-1 oligonucleotides:

H-1/D3(S) (SEQ ID NO: 23)

CCGGGTCCTAGAGGTGGACACGCA and

H-1/D3(AS) (SEQ ID NO: 24)

TGCAGTGTCTCCTGGCTCTGGTTC,

designed to amplify a 230 bp fragment of ICAM-1 domain 3. Two clones containing 75-95 kb genomic DNA inserts were analyzed. Plasmid DNA from each clone was digested with BamHI and blotted onto nylon membranes. Southern blots were hybridized under either low stringency (30% formamide) or high stringency (60% formamide) at 42.degree. C. with an ICAM-R domain 1 through 4 radiolabelled probe (other constituents of the hybridization solution were as described in Example 6A). The low stringency hybridization series was washed at room temperature in 2.times.SSPE, 0.1% SDS. The high stringency hybridization series was washed at 65.degree. C. in 0.2.times.SSPE, 0.1% SDS. The washed membranes were exposed to X ray film for 3.5 hours. ICAM-R genomic sequences were determined to be located on 4.0 kb and 1.5 kb BamHI fragments. The ICAM-R fragments were subcloned onto pBS+ (Stratagene) and their identity confirmed by limited sequence analysis. The genomic sequence information obtained for ICAM-R corresponds to the third domain of the protein.

Example 6

Human ICAM-R cDNA was transfected into L-M(TK.sup.-) mouse cells (ATCC CCL 1.3) and the cells were assayed for expression of ICAM-R by Northern blot and in situ hybridization.

A. Transfection of ICAM-R DNA

The full length human ICAM-R cDNA insert of pVZ-147 (Example 4) and a small portion of the phagmid vector 3' to the cDNA insert was excised using restriction enzymes NotI and XbaI and ligated into commercial plasmid pCDNA1-neo (Invitrogen Inc., San Diego, Calif.) cut with NotI and XbaI. The resulting plasmid, designated pCDNA1-neo-ICAM-R, was transfected into mouse L cells by the calcium phosphate precipitation method described in Chen et al., Molecular and Cellular Biology, 7: 2745-2748 (1987). ICAM-1 DNA (construct pCDNA-neo-ICAM-1) was also transfected into mouse L cells as a control. A cDNA fragment containing the complete ICAM-1 protein coding region was ligated into plasmid pCDNA1-neo and transfected into L cells by the calcium phosphate precipitation method. Following selection for neomycin resistance, individual ICAM-R or ICAM-1 transfectants were subcloned using cloning cylinders (Bellco Glass Inc., Vineland, N.J.). The clones expressing the highest level of ICAM-R and ICAM-1 protein were then sorted on a cell-sorter.

Constructs pCDNA-neo-ICAM-R and pCDNA-neo-ICAM-1 were also transfected into CV-1 cells by the calcium phosphate precipitation method. The clones expressing high levels of ICAM-R and ICAM-1 were selected as described above for L cell tranfectants. Based on FACs analysis with ICAM-R and ICAM-1 specific antibodies the level of protein expression was higher with CV-1 transfectants then with the mouse LTK transfectants.

B. Northern Blot Hybridizations

Following transfection of full length ICAM-R or ICAM-1 cDNAs into mouse L cells, specific expression of the corresponding mRNAs in transfected and untransfected L cells was determined by Northern blot hybridization with .sup.32 P-labelled ICAM-R or ICAM-1 DNA probes. Transfectants were grown in log phase, then centrifuged and washed two times with 150 mM NaCl. The pellet was resuspended in 3.5 ml GIT (guanidinium isothiocyanate) buffer, then sheared in a polytron mixer for 20 seconds. After adding 1.7 ml CsCl buffer to an ultracentrifuge tube, the GIT/RNA mix was layered on top. Samples were spun at 35 K (179,000.times.g), 20.degree. C., for 21 hours. All liquid was removed and the pelleted RNA was resuspended in 300 .mu.l 0.3M sodium acetate pH 5.2, then precipitated with 750 .mu.l EtOH at -20.degree. C. The precipitate was resuspended in H.sub.2 O, then treated with Proteinase K to remove any RNAses. After a phenol/chloroform extraction, the RNA was re-precipitated, resuspended in H.sub.2 O and the OD of the sample at 260 nm was measured.

The RNAs were electrophoresed in 1% formaldehyde agarose gels, prepared with diethyl pyrocarbonate (DEPC) treated solutions. Ten .mu.g of each total RNA sample was loaded per lane. RNA was electrophoresed at 30 V for approximately 18 hours with continuous circulation of buffers accomplished with a peristaltic pump. Each resulting gel was soaked two times in 20.times.SSPE for 20 minutes each at room temperature. Transfer of RNA to Hybond-C membranes (Amersham Corp., Arlington Heights, Ill.) was accomplished by capillary action overnight in 20.times.SSPE. Using a Stratagene stratalinker, RNA was stably crosslinked to each membrane by exposure to ultraviolet light.

To generate ICAM-1 DNA probes, 100-200 ng template DNA (a 1.8 kb Xba/Kpn fragment incorporating the entire ICAM-1 coding sequence) was mixed with H.sub.2 O and random hexamer, boiled for 5 minutes, and then incubated 5 minutes on ice. To the template DNA were added: .sup.32 P-dCTP and .sup.32 P-dTTP, 10.sup.-4 M dGTP/dATP, 10.times.Klenow Buffer (Boehringer Mannheim Biochemicals, Indianapolis, Ind.) and Klenow enzyme, and the mixture was left at room temperature for 1 hour. Samples were passed over a Quickspin G25 DNA column (Boehringer) to separate incorporated from unincorporated label.

To generate ICAM-R DNA probes, 200 pg of DNA template (a 1.4 kb fragment of clone pVZ-147 truncated to remove the poly-A tail) was amplified by PCR primed with oligonucleotides complimentary to the 5' and 3' extremities of domain 1. .sup.32 P-dCTP was added to the reaction mixture. Samples were held at 94.degree. C. for 4 minutes then run through 30 cycles of the temperature step sequence (94.degree. C., 1 minute; 50.degree. C., 2 minutes; 72.degree. C., 4 minutes) Samples were then run over a Quickspin column and incorporation of label was assessed by scintillation counting of 1 .mu.l aliquots.

The DNA probes were denatured with 5M NaOH, then neutralized with 1M Tris. The Hybond-C membranes were prehybridized at 50.degree. C. for 30 minutes in a 50% formamide pre-hybridization mix. Probe was added to each membrane to a concentration of 1.times.10.sup.6 cpm/ml hybridization mix (50% formamide, 5.times.Denhardt's solution, 5.times.SSPE, 1% SDS), and the membranes were incubated overnight at 42.degree. C. Each membrane was then washed 5 times in 2.times.SSPE/0.1% SDS at room temperature for 10 minutes each wash. One 10 minute wash was done at 50.degree. C. in 0.5.times.SSPE/0.1% SDS, with an additional rinse in 2.times.SSPE. Hybridization with the major RNA transcript was quantitated using a Molecular Dynamics (Sunnyvale, Calif.) Model 400A PhosphorImager.

Results of the northern blot hybridizations are presented in bar graph form in FIG. 2(A through B). FIG. 2A illustrates specific hybridization of the ICAM-R probe with RNA extracted from ICAM-R transfectants, but not with RNA from ICAM-1 transfectants or untransfected L cells. Reciprocally, FIG. 2B indicates hybridization of the ICAM-1 probe with RNA extracted from ICAM-1 transfectants, but not with RNA from ICAM-R transfectants or parental L cells.

C. In situ Hybridizations

L cells and L cells transfected as described above with either ICAM-R or ICAM-1 cDNAs were hybridized in situ with radiolabelled single-stranded RNA probes derived from ICAM-R or ICAM-1. Single-stranded RNA probes were generated from DNA templates corresponding to the first (i.e., N-terminal) immunoglobulin-like domain of ICAM-R or ICAM-1 by in vitro RNA transcription incorporating .sup.35 S-UTP. Probes were chemically hydrolyzed to approximately 200 bp.

Transfected and untransfected L cells were layered onto Vectabond (Vector Laboratories, Inc., Burlingame, Calif.) coated slides and stored at -70.degree. C. Prior to use, slides were removed from -70.degree. C. and placed at 55.degree. C. for 5 minutes. Sections were then fixed in 4% paraformaldehyde for 20 minutes at 4.degree. C., dehydrated in 70-95-100% EtOH for 10 minutes at room temperature, and allowed to air dry for 30 minutes. Sections were denatured for 2 minutes at 70.degree. C. in 70% formamide/2.times.SSC, rinsed in 2.times.SSC dehydrated and then air dried for 30 minutes. Prehybridization for 2 hours at 42.degree. C. with a mixture containing 50% formamide, 0.3M NaCl, 20 mM Tris pH 8.0, 10% dextran sulfate, 1.times.Denhardt's solution, 100 mM dithiothreitol (DTT) and 5 mM EDTA was performed. Hybridization was carried out overnight (12-16 hours) at 50.degree. C. in the same mixture additionally containing either .sup.35 S-labelled ICAM-1 or .sup.35 S-labelled ICAM-R RNA probes (6.times.10.sup.5 cpm/section). After hybridization, sections were washed for 1 hour at room temperature in 4.times.SSC/10 mM DTT, then for 40 minutes at 60.degree. C. in 50% formamide/1.times.SSC/10 mM DTT, 30 minutes at room temperature in 2.times.SSC, and 30 minutes at room temperature in 0.1.times.SSC. The sections were alcohol dehydrated, air dried for 30 minutes, developed (after storage at 4.degree. C. in complete darkness) and counterstained with hematoxylin/eosin.

Photomicrographs of the in situ hybridizations are set out in FIG. 3(A through F) wherein photomicrograph 3A is of parental L cells probed with ICAM-R RNA; 3B is of ICAM-R transfected L cells probed with ICAM-R RNA; 3C is of ICAM-1 transfected L cells probed with ICAM-R RNA; 3D is of parental L cells probed with ICAM-1 RNA; 3E is of ICAM-R transfected L cells probed with ICAM-1 RNA; and 3F is of ICAM-1 transfected L cells probed with ICAM-1 RNA. The photomicrographs demonstrate specific hybridization of each RNA probe only with L cells transfected with a homologous cDNA.

Example 7

Experiments testing the adhesion of leukocytes to transfected L cells expressing ICAM-R on their surface or to soluble ICAM-R (Example 10) indicate that ICAM-R is a ligand/receptor for an adhesion molecule or molecules on leukocytes.

A. CD18-Dependent Cell Adhesion

SKW3 cells (T lymphoblastoid cells) were pretreated with phorbol ester to activate LFA-1-dependent adhesion as described in Dustin et al., Nature, 341: 619-624 (1989) and were assayed for binding to ICAM-R and ICAM-1 transfectants.

Untransfected L cells or L cells transfected with either ICAM-R or ICAM-1 (see Example 7) were seeded in 24-well tissue culture plates (3.times.10.sup.5 cells per well) 24-48 hours prior to the adhesion assay. SKW3 cells were washed in serum-free RPMI (Gibco, Canada), labelled with Calcein-AM (Molecular Probes Inc., Eugene, Oreg.), and stimulated with 10 ng/ml phorbol myristylacetate (PMA) for 20 minutes at 37.degree. C. Selected stimulated SKW3 cells were then pretreated with anti-CD18 (TS1/18, ATCC HB203), anti-CD11a (TS1/22, ATCC HB202) hybridoma supernatant or control anti-CD2 (ATCC HB195) purified monoclonal antibody for 30 minutes at room temperature before incubation with adherent, transfected L cells. Antibody-treated and non-antibody-treated, calcein-labelled SKW-3 cells were added (5.times.10.sup.5 cells per well) to confluent monolayers of ICAM-R or ICAM-1 transfectants and incubated for 30 minutes at 37.degree. C. in RPMI/1% fetal calf serum (FCS, Hyclone Laboratories Inc., Logan, Utah) Unbound cells were aspirated and wells were filled with RPMI-FCS. Plates were sealed, centrifuged in an inverted position at 200 rpm for 4 minutes and aspirated. The plates were then washed with RPMI-FCS and scanned with an automatic fluorescence reader.

Adhesion of stimulated SKW3 cells to both the ICAM-R and the ICAM-1 transfectants was inhibited by monoclonal antibodies against either the .alpha. (CD11a) or .beta. (CD18) chains of LFA-1 indicating that ICAM-R may participate in intercellular adhesion events involving a .beta.2 integrin pathway. Intracellular adhesion was unaffected by the control anti-CD2 reagent.

B. CD18-Independent Cell Adhesion

CD18 negative lymphoblastoid cells from patients with leukocyte adhesion deficiency (LAD) bind to soluble ICAM-R described in Example 10. (See FIG. 4A wherein the experimental control was binding of cells to plates coated with 1% BSA.) In addition, the majority (80-90%) of binding of the Jurkat T lymphoblastoid cell line to ICAM-R is not inhibited by anti-CD18 monoclonal antibody 60.3 described in Beatty et al., J. Immunol., 131: 2913-2918 (1983)! or anti-CD11a monoclonal antibody (TS1/22) (FIG. 4B). These results suggest that binding of ICAM-R to these cell lines is CD 18-independent and that LAD and Jurkat cells express a counterreceptor for ICAM-R that is not a .beta..sub.2 integrin.

Example 8

Human sequence ICAM-R peptides were used to inhibit SKW3 and Jurkat cell binding to ICAM-R. The former type of adhesion is CD18-dependent while the latter is largely CD18-independent.

Based on amino acid sequence alignment with known .beta..sub.1 integrin binding domains in fibronectin and based on epitope mapping of anti-ICAM-R monoclonal antibodies that block cell adhesion (see Table 11 in Example 21), ICAM-R peptides corresponding to potential integrin binding sites were synthesized by Macromolecular Resources (Colorado State University, Fort Collins, Colo.). Four ICAM-R sequences which lie between or at the border of predicted beta strands in domains 1 and 3 of were chosen. Similar but not identical .beta.-strand predictions for ICAM-1 are set out in Staunton et al., Cell, 61: 243-254 (1990). Inhibition was assayed using a system involving cell adhesion to soluble ICAM-R coated plastic. Calcein-labeled cells (see Example 7 Section A above) were incubated with peptide at 1-2 mg/ml for 20 minutes at 25.degree. C. and the cells were transferred to wells of a 96-well plate previously coated with soluble ICAM-R (see Example 10) and containing 10 .mu.g/ml final concentration phorbol 12-myristate 13-acetate (PMA). After 50 minutes, the plate was inverted in PBS for 10 minutes to remove unbound cells. Bound cells were quantitated using a fluorescence concentration analyzer.

The results of the assay are presented below in Table 2 wherein numbering of peptide residues of ICAM-R corresponds to SEQ ID NO: 1 while numbering of peptide residues of ICAM-1 corresponds to the ICAM-1 amino acid sequence presented in Staunton et al., supra, and wherein the abbreviation "ND" stands for "not

ICAM-R peptide sequences from domain 3 inhibited binding of Jurkat cells to ICAM-R but not binding of SKW3 cells to ICAM-R. Domain 3 peptides were two-fold more efficient than domain 1 peptide sequences in inhibiting Jurkat cell binding, suggesting that Jurkat binding to ICAM-R may preferentially involve ICAM-R domain 3. The ICAM-R domain 1 peptide (NGSQI) corresponding to residues 72-76 of SEQ ID NO: 1 inhibited SKW3 binding to ICAM-R by 26%. The corresponding ICAM-1 peptide (DGQST, SEQ ID NO: 25) did not inhibit binding. In contrast, the ICAM-R domain 3 peptide (GDQML) corresponding to amino acids 230-234 of SEQ ID NO: 1 demonstrated the best inhibition (36%) of Jurkat binding to ICAM-R. The corresponding ICAM-1 peptide (GDQRL, SEQ ID NO: 26) inhibited Jurkat binding by 22%.

The tri-peptide RGD is a recognition sequence common to extracellular matrix components (e.g., fibronectin and vitronectin) that are ligands of the beta-1 integrins. Cyclizing RGD-containing peptides has resulted in a ten-fold increase in efficiency of blocking integrin binding to vitronectin Pierschbacher and Ruoslahti, J. Biol. Chem., 262(36): 17294-17298 (1987)!. ICAM-R peptide sequences corresponding to domain 1 residues 72-77 and domain 3 residues 230-234 are being cyclized using bromoacetic acid preparative to testing in the assay outlined above.

Example 9

A soluble variant of human ICAM-R was constructed and expressed as follows.

A. Construction of the Expression Vector Encoding Soluble ICAM-R

The human cDNA for ICAM-R was altered by standard procedures of site-directed mutagenesis see, e.g., Kunkel et al., Proc. Natl. Acad. Sci. USA, 82: 488-492 (1985)! in order to truncate the protein coding sequence at the predicted junction (amino acid 457) of its extracellular and transmembrane domains as determined by a computer algorithm that predicts hydropathy Kyte et al., J. Mol. Biol., 157: 105-132 (1982)!. The DNA sequence of ICAM-R was cut from pVZ147 (Example 4) with restriction enzymes SalI and NotI. The resulting fragment included the complete ICAM-R coding sequences beginning at the 5' end of the coding strand and also included at the 3' end a short segment of the multiple cloning sites. This fragment was subcloned into the M13 BM21 vector (Boehringer) linearized with SalI and NotI resulting in a molecule called M13 BM21ICAM-R.

A mutagenizing oligonucleotide was synthesized with the sequence below.

ICAM-Rt1 (SEQ ID NO: 27)

CTGCCCCTGAATCACCCTCGA

The oligonucleotide changes the phenylalanine at position 457 of ICAM-R to a stop codon. The oligonucleotide was utilized as described in Kunkel et al., supra, to generate from M13 BM21ICAM-R six M13 phage isolates encoding a stop codon at position 457. An isolate designated BM21ICAM-Rt1 was chosen for further study.

This single strand template was converted to a double strand DNA molecule by primer extension using Klenow DNA polymerase as follows. Ten .mu.g of purified single strand M13 BM21ICAM-Rt1 DNA was annealed to 50 ng Lac Z universal -20 primer (GTAAAACGACGGCCAGT, SEQ ID NO: 28) in 1.times.Klenow DNA polymerase buffer (10 mM Tris-Cl pH 7.5, 5 mM MgCl.sub.2, 7.5 mM dithiothreitol) by incubating the mix at 65.degree. C. for 5 minutes and then 25.degree. C. for 5 minutes. The following mixture was then added to the annealing reaction: 33 .mu.M final concentration dATP, dGTP, dCTP, dTTP; 4 units of Klenow DNA polymerase (Boehringer), and 1.times.Klenow buffer. The primer extension reaction was allowed to incubate at 37.degree. C. for 45 minutes prior to being stopped by a single phenol/chloroform (1:1) extraction and ethanol precipitation. A portion of the cDNA insert was released from the M13 BM21ICAM-Rt1 phage by restriction digest using restriction enzymes EcoRV and NcoI. The fragment of DNA released contained the complete coding sequence for the truncated ICAM-R protein, the 3' untranslated region and a small segment of polylinker sequence from the M13 BM21 phage. After agarose gel purification the fragment was ligated to linearized vector Bluebac III (Invitrogen Corp., San Diego, Calif.), a transfer vector containing genomic baculovirus sequences for homologous recombination that flank the ETL promoter driving expression of the E.coli beta-galactosidase gene and the polyhedron promoter driving expression of the gene of interest, in this case ICAM-Rt1.

The Bluebac III vector had been prepared in the following way prior to ligation. Three .mu.g of supercoiled plasmid DNA was digested with 20 units HinDIII endonuclease (Boehringer). After a phenol/chloroform extraction and ethanol precipitation the DNA pellet was resuspended in 1.times.Klenow DNA polymerase buffer; 33 .mu.M final concentration dATP, dGTP, dCTP, dTTP; 2 units of Klenow DNA polymerase (Boehringer) and incubated at 37.degree. C. for 60 minutes to fill in the termini of the molecule. The fill-in reaction was terminated by phenol/chloroform extraction and precipitation with ethanol. The blunt-ended DNA was resuspended in 1.times.NcoI buffer, 20 units of NcoI endonuclease were added and incubated at 37.degree. C. for 60 minutes.

A portion of the ligation reaction of the ICAM-Rt1 insert and linearized plasmid was used to transform electro-competent XL-1 E.coli (Stratagene) and individual colonies were selected on LB plates supplemented with 60 .mu.g/ml carbenicillin. Twelve individual isolates were analyzed by digestion of mini-prep DNA using PstI or EcoRI for diagnostic purposes. One isolate that exhibited the expected band pattern was designated pBBIII.ICAM-Rt1.

B. Expression of Soluble Human ICAM-R

Sf-9 cells (Invitrogen) to be transfected or infected with pBBIII ICAM-Rt1 DNA were maintained in spinner flasks in TNM-FH Grace's medium (Gibco, Grand Island, N.Y.) supplemented with 10% heat inactivated fetal bovine serum and gentamicin at 10 .mu.g/ml! at 27.degree. C. in a forced draft incubator. Spinner flask impellers were rotated at 60 rpm on an insulated five place stir plate. Log phase Sf-9 cells (1.5-2.5.times.10.sup.6/ ml) with greater than 90% viability were routinely subcultured twice weekly.

Sf-9 cells at log growth phase were plated (2.times.10.sup.6 cells/60 mm dish) in TNM-FH medium and allowed to attach for 1 hour at 27.degree. C. After this time the following mixture was made up in a sterile polystyrene tube and incubated at room temperature for 15 minutes: 1 ml TMN-FH medium, 1 .mu.g linear Autographa californica nuclear polyhidrosis virus (AcNPV, baculovirus) genomic DNA (Invitrogen), 3 .mu.g of pBBIII.ICAM-Rt1 DNA and 20 .mu.l of a stock cationic liposome solution (Invitrogen). Two other independent mixtures were made up with or without pBluebac III substituted for pBBIII.ICAM-Rt1 DNA as controls. The media was removed from the seeded plates, replaced with 2 ml of Grace's medium and allowed to incubate for 2 minutes. All media was removed from the plates and the DNA/liposome mixtures were added dropwise on the cells of individual plates. One plate received TNM-FH medium alone as a mock transfection control. The plates were then incubated at 27.degree. C. for 4 hours with occasional rocking. Following this incubation, 1 ml of TNM-FH medium was added to the plates. After further incubation for 48 hours, the transfection media containing virus was removed and these viral stocks were used to infect plates of Sf-9 cells for plaque identification.

Sf-9 cells were seeded at 2.times.10.sup.6 cells/60 mm dish in TNM-FH medium and allowed to attach for approximately 1 hour at 27.degree. C. The media was removed. Several 10-fold serial dilutions were made from each viral stock and 1 .mu.l of each dilution was added to a single dish of adherent Sf-9 cells and incubated for 1 hour at 27.degree. C. Following removal of the virus inoculum, each dish of cells was overlayed with 3 ml of a mixture of TNM-FH medium, 0.625% low melting point agarose (BRL, Gaithersburg, Md.) and 300 .mu.g/ml halogenated idolyl-beta-D-galactosidase (Bluo-gal, BRL) that had been previously equilibrated to about 30.degree. C. and allowed to solidify at room temperature for 1 hour. The plates were then incubated until blue color developed (typically 4-5 days). Twenty-four plaques of recombinant viruses (identified due to their expression of beta-galactosidase and conversion of the chromogenic substrate, Bluo-gal to a blue precipitate in infected cells) were transferred to individual wells of a 24-well cell culture plate that had been seeded with 1 ml of Sf-9 cells (2.times.10.sup.5 /ml) in TNM-FH. After 5 days at 27.degree. C. the media was harvested, microfuged at 1,000 rpm for 5 minutes at 4.degree. C. and the resulting supernatant was transferred to a fresh tube. These stocks were designated as BacR.P1 stocks with their respective isolate number.

BacR.P1 stocks were assayed for the production of ICAM-R by an antigen capture (ELISA) assay. Anti-ICAM-R monoclonal antibody ICR-4.2 (see Example 12) was biotinylated as follows. A tenth volume of 1M NaCO.sub.3 was added to monoclonal antibody ICR-4.2 at 1 mg/ml. NHS-biotin (Sigma Chemical Co., St. Louis, Mo.) was dissolved into dimethyl sulfoxide (DMSO, Mallinckrodt, Paris, Ky.) at 1 mg/ml. One hundred eighty .mu.l biotin solution was added to each 1 mg antibody and rotated at 4.degree. C. overnight. The biotinylation reaction was terminated by dialysis against PBS for 16 hours with 3 changes at 4.degree. C. For the assay of BacR.P1 stocks, each well of a ninety-six well plate was coated with monoclonal antibody ICR-1.1 (50 .mu.l at 10 .mu.g/ml) for either 2 hours at 37.degree. C. or 16 hours at 4.degree. C. The coating was then aspirated and the wells were rinsed 2 times with PBS. Wells were blocked with 200 .mu.l of 1% BSA in PBS for 30 minutes at 37.degree. C. Two ten-fold serial dilutions of BacR.P1 stocks were made in PBS. Fifty .mu.l from the BacR.P1 stocks (neat) or the dilutions were added to the wells and incubated for 30 minutes at 37.degree. C. After 2 washes with PBS, 50 .mu.l for a 1:250 dilution of biotinylated ICR-4.2 in 1% BSA/PBS was added to the wells and incubated for 30 minutes at 37.degree. C. After 3 washes with PBS, 50 .mu.l/well of horseradish peroxidase conjugated to streptavidin (Zymed Laboratories Inc., San Francisco, Calif.) diluted in 1% BSA/PBS to 1:4000 was added and incubated for 30 minutes at 37.degree. C. After 2 washes with PBS, 200 .mu.l/well substrate buffer with ABTS (Zymed) was added and incubated at room temperature until a color reaction developed. The plate was read in an automated plate reader at a wavelength of 410 nm.

Four of the highest expressors of soluble ICAM-R as determined by the above antigen capture assay were chosen for plaque purification and BacR.P1 stocks of those isolates were diluted by 10-fold serial dilutions and plated with an agar overlay. A single blue plaque from the highest dilution was isolated and placed in 1 ml of TNM-FH medium, vortexed vigorously and serially diluted for one more round of plaque isolation. A final plaque isolate was chosen that was clear of all wildtype baculovirus and removed to a T-25 flask that has been seeded with 2.times.10.sup.6 Sf-9 cells in TNM-FH media. After 5 days incubation at 27.degree. C., the media was harvested by centrifugation at 1200 rpm for 5 minutes and 4 ml of the supernatant (designated BAC-R.P2 stock) was transferred to a 1 liter spinner flask containing 500 ml of TNM-FH seeded with 2.times.10.sup.6 cells/ml. After another 5 days incubation at 27.degree. C., the infection media was harvested by centrifugation at 1000 rpm for 5 minutes. The supernatant was stored at 4.degree. C. and was designated BAC-R.P3 stock. The BAC-R.P3 stock was titered by plating aliquots of ten fold serial dilutions onto adherent Sf-9 cells and overlaying with 0.625% agarose in TNM-FH supplemented with 300 .mu.g/ml Bluo-gal (BRL). After 4 days incubation at 27.degree. C., the number of plaques was counted and a titer determined.

Infections for expression of soluble ICAM-R protein were carried out in 3 liter flasks containing 1.5 L of EX/Cell 401 medium (JRH Biosciences, Lenexa, Kans.). Sf-9 cells dividing at log phase (2.times.10.sup.6 /ml) were infected at a multiplicity of infection (moi) of 5 with BAC-R.P3 virus stock. After 4 days, the media was harvested and was separated from the cells by centrifugation. Soluble ICAM-R protein was purified from the insect cell media as follows. Four ml 1M Tris-Cl pH 7.5 was added to each 200 ml of insect cell supernatant and was pumped at about 35 ml/hour at 4.degree. C. onto an approximately 3.5 ml column of Lentil Lectin Sepharose (Pharmacia, Uppsala, Sweden) previously equilibrated with 20 mM Tris-Cl pH 7.5/0.1M NaCl (equilibration buffer). After loading, the column was washed with 25 ml equilibration buffer. The column was then eluted with 11 ml equilibration buffer containing 0.2M methyl .alpha.-D-mannopyranoside. The eluate contained soluble human ICAM-R (shICAM-R).

C. Binding of shICAM-R to Activated Lymphocytes

The partially purified shICAM-R protein was assayed for binding to SKW3 cells that were pretreated with phorbol ester as described in Example 7 to activate LFA-1-dependent adhesion. The ICAM-R protein was coated onto 96-well Immulon 4 (Dynatech) plates after adjusting the lectin eluate to 25 mM carbonate pH 9.6 and incubated overnight at 4.degree. C. The plates were washed two times with PBS, blocked for 30 minutes at 37.degree. C. with 200 ul/well PBS, 1% BSA, and washed again with PBS before adding cells. SKW3 cells were washed in serum-free RPMI (Gibco), labelled with Calcein-AM (Molecular Probes), and stimulated with PMA. Cells were then added to the plates and incubated for 1 hour at 37.degree. C. The plates were inverted in prewarmed PBS, 1% BSA and were incubated for 30 minutes. The plates were then removed and half of the contents of each well was aspriated. The plates were then scanned with a fluorescence microscope and an automated fluorescence reader. The results of the assay demonstrated adhesion of phorbol ester-activated lymphocytes to the plate bound shICAM-R protein.

D. Assays Utilizing shICAM-R

In vitro assays for identifying antibodies or other compounds which modulate the activity of ICAM-R may be developed that utilize shICAM-R. For example, such an assay may involve immobilizing ICAM-R or a natural ligand to which ICAM-R binds, detectably labelling the nonimmobilized binding partner, incubating the binding partners together and determining the effect of a test compound on the amount of label bound wherein a reduction in the label bound in the presence of the test compound compared to the amount of label bound in the absence of the test compound indicates that the test agent is an inhibitor of ICAM-R binding. Functional .beta..sub.2 leukointegrins that may be utilized in such assays are described in Dustin et al., CSH Symp. Qual., 54: 753-765 (1989).

The following preliminary experiment shows that purified shICAM-R can be bound to polystyrene beads and retain the ability to bind to purified leukointegrins coated on a plastic surface, thus providing the basis for development of an assay to identify modulators of ICAM-R binding. Purified shICAM-R was used to coat 6 .mu.m fluorescent polystyrene beads (Polysciences, Inc., Warrington, Pa.) overnight according to the manufacturer's instructions and then the beads were blocked with BSA. Replicate wells of a 96-well plate were coated with a diluted aliquot of purified LFA-1 (CD18/CD11a), Mac-1 (CD18/CD11b) or Gp 150,95 (CD18/CD11c). After blocking the wells with BSA, the plates were incubated in buffer alone or buffer including anti-CD18 antibody (60.3). The ICAM-R-coated beads were aliquoted into the well and incubated for one hour at room temperature followed by inversion in a tank of PBS-D to remove unbound beads from the wells. Fluorescence remaining in the wells was detected using a Cytofluor 2300 (Millipore, Inc., Bedford, Mass.). In parallel experiments, leukointegrin preparations of LFA-1 or Mac-1 were coated on the fluorescent polystyrene beads and ICAM-R was immobilized.

Specific modulators of binding between ICAM-R and its binding partners may also be identified by scintillation proximity assay techniques as generally described in U.S. Pat. No. 4,271,139; Hart et al., Mol. Immunol., 12:265-267 (1979), and Hart et al., J. Nuc. Med., 20:1062-1065 (1979), each of which is incorporated herein by reference, may also be utilized to identify modulators.

Briefly, one member of the ICAM-R/ligand pair is bound to a solid support. A fluorescent agent is also bound to the support. Alternatively, the fluorescent agent may be integrated into the solid support as described in U.S. Pat. No. 4,568,649, incorporated herein by reference. The non-support bound member of the ICAM-R/ligand pair is labelled with a radioactive compound that emits radiation capable of exciting the fluorescent agent. When, for example, ICAM-R binds the radiolabeled ligand, the label is brought sufficiently close to the support-bound fluorescer to excite the fluorescence and cause emission of light. When not bound, the label is generally too distant from the solid support to excite the fluorescent agent, and light emissions are low. The emitted light is measured and correlated with binding between ICAM-R and the labelled ligand. Addition of a putative modulator to the sample will decrease the fluorescent emission by keeping the radioactive label from being captured in the proximity of the solid support. Therefore, binding inhibitors may be identified by their effect on fluorescent emissions from the samples. Potential ligands to ICAM-R may also be identified by similar assays in which no modulator is included.

Example 10

To rapidly screen for the functional consequences (i.e., counter-receptor binding) of point mutations in ICAM-R extracellular immunoglobulin-like domains, a system was employed from which shICAM-R molecules having point mutations can be expressed and purified. The system relies on the specific binding properties of a poly-histidinyl tract fused to the amino or carboxyl terminus of a given protein Hochuli et al., Bio/Technology, 6: 1321-1325 (1988)!. The utility of the system in the purification of proteins under native conditions has been demonstrated Janknecht et al., Proc. Natl. Acad. Sci., USA, 88: 8972-8976 (1991)!.

Plasmids pCS57.1 and pCS65.10 both are pcDNAlamp (Invitrogen) with the full length human ICAM-R cDNA inserted between EcoRV and XhoI sites, but pCS65.10 includes point mutations that encode Ala.sub.37 and Ser.sub.38 rather than the wild type Glu.sub.37 and Thr.sub.38, respectively! were used for the initial studies. These DNAs were digested with SacI and EcoRI to release the entire extracellular domain of ICAM-R (amino acids -29 to +454) and the fragments were gel isolated.

Two complimentary oligonucleotides were synthesized that encoded wild type residues Ser.sub.454 and Ser.sub.455, and introduced a Gly.sub.456, Pro.sub.457 and Gly.sub.458 to encourage an alpha helical turn followed by a stretch of six histidine residues and a translational terminator codon. The sequences of the oligonucleotides were:

SEQ ID NO: 29

CAGGTCCCGGTCATCATCATCATCATCATTAAT

SEQ ID NO: 30

TAGATTAATGATGATGATGATGATGACCGGGACCTGAGCT

The oligonucleotides which contain a SacI site and an XbaI site at the ends were ligated to the extracellular domain of ICAM-R and pcDNA1amp cut with EcoRI and XbaI. One set of ligations contained 0.5 u polynucleotide kinase to phosphorylate the 5' ends of the synthetic DNAs thus increasing the efficiency of ligation. A second set of ligation reactions contained pre-phosphorylated oligonucleotides. Colonies were screened by either miniprep restriction enzyme digestion analysis and PCR with ICAM-R specific oligonucleotide primers or PCR alone. DNA sequence was obtained for several clones. The resulting plasmids were designated p57.1 wtHis6 and p65.10E37T His6.

COS cells were seeded in 10 cm dishes and grown to about 50% confluency at which time they were transiently transfected by the DEAE-dextran method in serum free DMEM using 10 ug of purified plasmid DNA per dish or mock transfected. After a brief DMSO shock, the cells were incubated in DMEM supplemented with fetal bovine serum. After 24 hours, the medium was replaced and the cells allowed to reach confluency over the course of the next four days. The final medium harvest was removed from the cell monolayer and spun at 1000 rpm to remove cells and stored at 4.degree. C. until ready for column chromatography.

Ni.sup.++ -nitrilotriacetic acid (Ni.sup.++ -NTA) agarose affinity column chromatography was performed essentially as described in Janknecht et al., supra, except that the purification was from medium rather than from lysed cells. To the medium was added an equal volume of buffer A (830 mM NaCl, 34% glycerol, 1.6 mM imidazole) and the mixture was clarified by centrifugation at 10,000.times.g for 10 minutes at 4.degree. C. One ml of an Ni.sup.++ -NTA agarose bead suspension (50%) (Qiagen) per 16 mls of buffered medium sample was preequilibrated in 3.3 ml of 0.5.times.buffer A by gentle rocking at 25.degree. C. for 30 minutes. The beads were then spun to a pellet at 600 rpm and most of the supernatant was removed. The beads were resuspended to a total volume of 3 ml in fresh 0.5.times.buffer A and 1 ml dispensed to each clarified and buffered medium sample. The remainder of the prep was carried out at 4.degree. C. After 60 minutes of constant agitation each medium sample was passed through a disposable 10 ml polypropylene column (Biorad) to pack the beads and the flow through collected. The beads were then washed with 9 column volumes (4.5 mls) of buffer D (10 mM HEPES pH 7.9, 5 mM MgCl.sub.2, 0.1 mM EDTA, 50 mM NaCl, 1 mM dithiothreitol, 17% glycerol) supplemented with 0.8 mM imidazole. The beads were then washed twice with 9 column volumes of buffer D supplemented with 8 mM imidazole, twice with 5 column volumes of buffer D supplemented with 40 mM imidazole and twice with 5 column volumes of buffer D supplemented with 80 mM imidazole.

Two hundred ul of each fraction were assayed for ICAM-R immunoreactivity by enzyme linked immunofiltration assay (ELIFA) in a 96-well format as described by the manufacturer (Pierce). Purified monoclonal antibody ICR-4.2 (5 ug/ml) (see Example 11) was used as the primary detection agent and a purified goat anti-mouse horseradish peroxidase conjugate (Boehringer Mannheim Biochemicals) (1:500) was used as the secondary antibody. The assay was developed with the soluble substrate ABTS (Zymed) as recommended by the supplier and read using a Dynatech plate reader with a 410 nm test filter. The results showed that ICAM-R immunoreactivity was predominantly found in the first 40 mM imidazole wash.

Peak fractions from wtHis6, E37His6 and mock transfectants were concentrated about 6.5 fold using Centricon 30 (Amicon) centrifugation units. The resultant concentrates were adjusted to equal vols. (0.34 ml) using PBS-D. Control soluble ICAM-R (15 ug/ml) (Example 9) in carbonate buffer pH 9.6 or in buffer D with 40 mM imidazole were made up. Fifty ul of a protein solution was aliquoted per well of a 96-well plate (Immulon 4, Dynatech) to coat the wells which were then assayed for binding of SKW3 cells as described in Example 9 using untreated, PMA-treated and anti-CD18 monoclonal antibody (60.3) treated cells.

Preliminary results indicate that wild type histidine tagged protein (wtHis6) functions as an adhesive ligand for SKW3 cells.

Example 11

Monoclonal antibodies specific for ICAM-R were generated from the fusion of NS-1 myeloma cells with spleen cells of Balb/c mice immunized with human cell lines that express ICAM-R. Monoclonal antibodies were generated from seven different fusions designated fusions 26, 42, 43, 46, 56, 63, and 81.

A. Immunization of Mice

For fusion 26, five 6 to 12-week old Balb/c mice (Charles River Biotechnical Services, Inc., Wilmington, Mass., IACUC #901103) were immunized with HL-60 cells to generate anti-ICAM-R monoclonal antibodies. Two Balb/c mice were bled retro-orbitally for the collection of pre-immune serum on day 0. On day 2, each animal received a total of 6.times.10.sup.6 HL-60 cells in 0.5 ml PBS (0.1 ml s.c. and 0.4 ml i.p.). A second immunization with 9.5.times.10.sup.6 HL-60 cells was administered on day 28 in the same manner. Immune serum was collected via retro-orbital bleeding on day 35 and tested by FACS (FACS screening is described in detail in Section C below) to determine its reactivity to ICAM-R transfectants. Based on these results, both animals were immunized a third time on day 51 with 6.5.times.10.sup.6 HL-60 cells and a fusion was performed with spleen cells sterilely removed from one animal (#764) on day 54.

For fusion 42, on day 0 each of five mice was prebled and then immunized i.p. with 5.times.10.sup.6 SKW3 cells in 0.5 ml PBS containing 50 .mu.g adjuvant peptide (Sigma). The mice were boosted in the same manner on days 21 and 42. Ten days after the third injection, the mice were bled and immune sera was tested by FACS. Mouse #843 was given a final boost of SKW3 cells on day 64. The spleen was sterilely removed three days later.

For fusion 43, on day 0 each of five mice was prebled and then immunized i.v. with 5.times.10.sup.6 cells from the erythroleukemic cell line K562. Each mouse was given a daily i.p. injection of 1.5 mg cyclophosphamide in 150 .mu.l for the next two days. On day 10, SKW3 cells plus adjuvant peptide were injected as in Fusion 42. On day 30, mice were given another cycle of K562 cells followed by cyclophosphamide. On day 42 mice were boosted with SKW3 cells with adjuvant peptide. Mice were bled on day 56 and immune sera was tested by FACS. Mouse #1021 was given a final boost of SKW3 cells and adjuvant peptide on day 78. The spleen was sterilely removed three days later.

For fusion 46, a mouse (#900) was immunized as described for fusion 42. On day 128, the mouse was given a final boost of approximately 4.times.10.sup.6 Macaca nemestrina spleen cells. The single cell suspension of monkey spleen was prepared as described below in the following paragraph. The monkey cells were pelleted and resuspended in erythrocyte lysis buffer: 0.15M NH.sub.4 Cl, 1M KHCO.sub.3, 0.1 mM Na.sub.2 EDTA, pH 7.2-7.4. After lysing the erythrocytes, the splenocytes were washed twice in RPMI and once in PBS. Finally, the cells were resuspended in 400 .mu.l PBS containing 50 .mu.g adjuvant peptide and injected. The mouse spleen was removed sterilely three days later.

For fusions 56 and 63, mice (#845 and #844) were immunized as described for fusion 42, except that no boost of SKW3 cells was given on day 64. Instead, these mice were given additional immunizations of SKW3 in PBS with adjuvant peptide on days 158 and 204 and were given i.p. injections of Macaca nemestrina spleen cells in 0.5 ml PBS containing 50 .mu.g adjuvant peptide on days 128 and 177. For fusion 56, mouse #845 was injected with 2.24 .mu.g soluble ICAM-R (Example 10) in 700 .mu.l PBS, 100 .mu.l was given i.v. with the remainder given i.p. The spleen was sterilely removed four days later. For fusion 63, mouse #844 was immunized on day 226 with Macaca nemestrina spleen cells as described for fusion 56 and on day 248 with 50 .mu.g soluble ICAM-R in 100 .mu.l complete Freuds adjuvant given s.c. The mouse received a final boost i.v. of 66 .mu.g soluble ICAM-R in 100 .mu.l PBS. The spleen was removed sterilely four days later.

For Fusion 81 each of 5 mice was prebled on day 0 and then immunized s.c. with 30 ug of soluble human shICAM-R (Example 9) in 0.2 ml complete Freund's adjuvant. On days 45 and 77, each mouse received 40 .mu.g of shICAM-R in 0.2 ml incomplete Freund's adjuvant. On day 136 mouse #1264 (Fusion 81) was given a final boost i.p. of 0.1 mg of shICAM-R in PBS. The spleen was sterilely removed three days later and a fusion was performed as described above.

B. Fusions

Briefly, a single-cell suspension was formed from each mouse spleen by grinding the spleen between the frosted ends of two glass microscope slides submerged in serum free RPMI 1640 (Gibco), supplemented with 2 mM L-glutamine, 1 mM sodium pyruvate, 100 units/ml penicillin, and 100 .mu.g/ml streptomycin (Gibco). The cell suspension was filtered through sterile 70 mesh Nitex cell strainer (Becton Dickinson, Parsippany, N.J.), and washed twice by centrifuging at 200 g for 5 minutes and resuspending the pellet in 20 ml serum free RPMI. Thymocytes taken from three naive Balb/c mice were prepared in a similar manner.

NS-1 myeloma cells, kept in log phase in RPMI with 11% fetal bovine serum (FBS) or Fetalclone (Hyclone) for three days prior to fusion, were centrifuged at 200 g for 5 minutes, and the pellet was washed twice as described in the foregoing paragraph. After washing, each cell suspension was brought to a final volume of 10 ml in serum free RPMI, and 10 .mu.l was diluted 1:100. Twenty .mu.l of each dilution was removed, mixed with 20 .mu.l 0.4% trypan blue stain in 0.85% saline (Gibco), loaded onto a hemacytometer (Baxter Healthcare Corp. Deerfield, Ill.) and counted.

A sample of 2.times.10.sup.8 spleen cells was combined with 4.times.10.sup.7 NS-1 cells, centrifuged and the supernatant was aspirated. The cell pellet was dislodged by tapping the tube and 2 ml of 37.degree. C. PEG 1500 (50% in 75 mM Hepes, pH 8.0) (Boehringer) was added with stirring over the course of 1 minute, followed by adding 14 ml of serum free RPMI over 7 minutes. An additional 16 ml RPMI was added and the cells were centrifuged at 200 g for 10 minutes. After discarding the supernatant, the pellet was resuspended in 200 ml RPMI containing 15% FBS or Fetalclone, 100 .mu.M sodium hypoxanthine, 0.4 .mu.M aminopterin, 16 .mu.M thymidine (HAT) (Gibco), 25 units/ml IL-6 (Boehringer) and 1.5.times.10.sup.6 thymocytes/ml. The suspension was dispensed into ten 96-well flat bottom tissue culture plates at 200 .mu.l/well. Cells in plates were fed three times typically on 2, 4, and 6 days post fusion by aspirating approximately 100 .mu.l from each well with an 18 G needle (Becton Dickinson), and adding 100 .mu.l/well plating medium described above except containing 10 units/ml IL-6 and lacking thymocytes.

C. Screening

When cell growth reached 60-80% confluency (day 8-10), culture supernatants were taken from each well of Fusions 26 and 42, pooled by column or row and analyzed by FACS on parental L cells (Fusion 26) or parental CV-1 cells (Fusion 42); (negative control) and on L cells (Fusion 26) or CV-1 cells (Fusion 42) transfected with ICAM-R DNA. Briefly, transfected and nontransfected L cells or CV-1 cells were collected from culture by EDTA (Versene) treatment and gentle scraping in order to remove the cells from the plastic tissue culture vessels. Cells were washed two times in Dulbecco's PBS with Ca.sup.2+ and Mg.sup.2+, one time in "FA Buffer" (either D-PBS or RPMI 1640, 1% BSA, 10 mM NaN.sub.3), and dispensed into 96-well round bottomed plates at 1.5-2.0.times.10.sup.5 cells/100 .mu.l FA Buffer per well. At this point, the assay was continued at 4.degree. C. Cells were pelleted by centrifugation in a clinical centrifuge at 4.degree. C. The supernatant from each well was carefully suctioned off, the pellets were broken up by gently tapping all sides of the assay plate. One hundred .mu.l of hybridoma supernatant pool was added per well using a 12-channel pipetman. Each monoclonal antibody-containing supernatant pool was incubated for 1 hour on both parental and transfected cells at 4.degree. C. Assay plates were then washed 2 times with FA Buffer as above. The last wash was replaced with a 50 .mu.l/well of a 1:100 dilution of a F(ab').sub.2 fragment of sheep anti-mouse IgG (whole molecule)-FITC conjugate (Sigma) prepared in FA Buffer. Assay plates were incubated at 4.degree. C. protected from light for 45 minutes. The assay plates were then washed 2 times with D-PBS containing NaN.sub.3 only (i.e., no BSA) in the same manner as before and the last wash was replaced with 200 .mu.l/well 1% paraformaldehyde in D-PBS. Samples were then transferred to polystyrene tubes with the aid of a multichannel pipet for flow cytometric analysis (FACS) with a Becton Dickinson FACscan analyzer.

Fusions 43 and 46 were screened initially by antibody capture ELISA, testing for the presence of mouse IgG in hybridoma supernatants. Immunlon 4 plates (Dynatech, Cambridge, Mass.) were coated at 4.degree. C. with 50 .mu.l/well goat anti-mouse IgA, IgG or IgM (Organon Teknika Corp., Durham, N.C.) diluted 1:5000 in 50 mM carbonate buffer, pH 9.6. Plates were washed 3 times with PBS with 0.05% Tween 20 (PBST) and 50 .mu.l culture supernatant was added. After incubation at 37.degree. C. for 30 minutes, and washing as above, 50 .mu.l of horseradish peroxidase conjugated goat anti-mouse IgG(fc) (Jackson ImmunoResearch, West Grove, Pa.) diluted 1:3500 in PBST was added. Plates were incubated as above, washed 4 times with PBST and 100 .mu.l substrate, consisting of 1 mg/ml o-phenylene diamine (Sigma) and 0.1 .mu.l/ml 30% H.sub.2 O.sub.2 in 100 mM Citrate, pH 4.5, was added. The color reaction was stopped in 5 minutes with the addition of 50 .mu.l of 15% sulfuric acid. A.sub.490 was read on an automatic plate reader.

Fusions 56 and 63 were screened initially by antigen capture ELISA. Immulon 4 plates (Dynatech) were coated at 4.degree. C. overnight with 100 ng 26E3D Fab' (see Section F below) per well, diluted in 50 mM carbonate buffer. The plates were blocked with 100 .mu.l/well 2% BSA in PBS for 1 hour at ambient temperature. After the plates were aspirated, culture supernatant containing soluble ICAM-R was diluted 1:8 in PBST and added at 50 .mu.l/well. After 1 hour incubation at ambient temperature, the wells were washed three times with PBST, hybridoma culture supernatant was added at 50 .mu.l/well, and the plates were again incubated as above. The plates were washed 3 times and 50 .mu.l/well peroxidase conjugated goat anti-mouse IgG diluted 1:3500 in PBST was added. The remainder of the assay was performed as described in the foregoing paragraph.

Fusion 81 was screened by ELISA on COS cells transiently transfected with either a domain 1 deleted ICAM-R construct Example 14.C.1! or with an ICAM-2 construct. The transfected cells were paraformaldehyde fixed on 96-well plates, and the remainder of the assay was performed as previously described, except no Tween 20 was used. The wells that were positive for domain 1 deleted ICAM-R but negative for ICAM-2, were tested on Cos cells transiently transfected with domain 1 deleted ICAM-R or domain 3 deleted ICAM-R by ELISA.

D. Subcloning

Supernatants from individual wells representing the intersection points of positive columns and rows (Fusions 26 and 42), individual wells producing IgG (Fusions 43 and 46), or individual wells reactive with soluble ICAM-R (Fusions 56 and 63) were rescreened by FACS the following day. L cells or L cells transfected with ICAM-R DNA were used for screening Fusion 26 antibodies and CV-1 cells or CV-1 cells transfected with ICAM-R DNA were used for screening antibodies from Fusions 42, 43, 46, 56 and 63. Twenty-nine wells (designated 26E3D-1, 26E3E, 26H3G, 26H11C-2, 26I8F-2, 26I10E-2, 26I10F, 42C5H, 42D9B, 43H7C, 46D7E, 56D3E, 56I4E, 63A10E, 63C3F, 63C11A, 63E9G, 63E12C, 63G3G, 63H6H, 63H9H, 63I1C, 63I6G, 63I12F, 63G4D, 63E11D, 63H4C, showed preferential staining of the ICAM-R transfectants versus the control cells. Fusion 81 was also rescreened by FACS. One well from fusion 81, designated 81K2F, was positive on ICAM-R domain 1 deletion cells but negative on domain 3 deletion cells (Example 14). The well was subcloned successively using RPMI, 15% FBS, 100 .mu.M sodium hypoxanthine, 16 .mu.M thymidine, and 10 units/ml IL-6. Subcloning was performed either by doubling dilution or by limiting dilution, by seeding 96 well plates at 0.5-1.0 cells/well. Wells of subclone plates were scored visually after 4 days and the number of colonies in the least dense wells were recorded. Selected wells of each cloning were tested, by FACS or ELISA as described, for reactivity observed in the original fusion well. Activity was retained in sixteen cell lines which were deposited with the ATCC 26E3D-1 (ATCC HB 11053), 26H11C-2 (HB 11056), 26I8F-2 (HB 11054), 26I10E-2 (ATCC HB 11055), 42C5H (ATCC HB 11235), 42D9B (ATCC HB 11236), 43H7C (ATCC HB 11221), 46D7E (ATCC HB 11232) and 46I12H (ATCC HB 11231), 63E11D (ATCC HB 11405), 63G4D (ATCC HB 11409), 63H4C (ATCC HB 11408), 63H6H (ATCC HB 11407), 63I1C (ATCC HB 11406), 63I6G (ATCC HB 11404), and 81K2F (ATCC HB 11692). In the final cloning, positive wells containing single colonies were expanded in RPMI with 11% FBS. Names assigned to the monoclonal antibodies produced by the hybridomas are presented in Table 4 in Example 12.

E. Characterization

The monoclonal antibodies produced by above hybridomas were isotyped in an ELISA assay. Immulon 4 plates (Dynatech) were coated at 4.degree. C. with 50 .mu.l/well goat anti-mouse IgA, IgG or IgM (Organon Teknika) diluted 1:5000 in 50 mM carbonate buffer, pH 9.6. Plates were blocked for 30 minutes at 37.degree. C. with 1% BSA in PBS, washed 3 times with PBS with 0.05% Tween 20 (PBST) and 50 .mu.l culture supernatant (diluted 1:10 in PBST) was added. After incubation and washing as above, 50 .mu.l of horseradish peroxidase conjugated rabbit anti-mouse IgG.sub.1, G.sub.2a, G.sub.2b or G.sub.3 (Zymed) diluted 1:1000 in PBST with 1% normal goat serum was added. Plates were incubated as above, washed 4 times with PBST and 100 .mu.l substrate, consisting of 1 mg/ml o-phenylene diamine (Sigma) and 0.1 .mu.l/ml 30% hydrogen peroxide in 100 mM Citrate, pH 4.5, was added. The color reaction was stopped in 5 minutes with the addition of 50 .mu.l of 15% sulfuric acid. A.sub.490 was read on a plate reader. The isotypes of the monoclonal antibodies are give in Table 11 in Example 21.

FACS analyses of indirect immunofluorescence staining of control cells and cells transfected with ICAM-R or ICAM-1 DNA using monoclonal antibodies against ICAM-R, ICAM-1 and ICAM-2 were performed. Staining was carried out as described for FACS analyses in Example 12C using either 0.1 ml hybridoma culture supernatant (anti-ICAM-R) or 1 .mu.g pure monoclonal antibody (anti-ICAM-1 or ICAM-2) per 5.times.10.sup.5 cells. Results of the analyses are presented as histograms (representing 10.sup.4 cells analyzed) in FIG. 5. Anti-ICAM-R antibodies specifically bound to L cells transfected with ICAM-R cDNA, but not to parental or ICAM-1 transfected L cells. ICAM-R transfectants did not react with antibodies against ICAM-1 (Mab LB2 from Edward Clark, University of Washington) or ICAM-2 (IC2/2, Biosource Genetics Corp., Vacaville, Calif.).

FACS analysis of indirect immunofluorescence of Macaca fascicularis, porcine or canine peripheral blood leukocytes was performed using the anti-ICAM-R monoclonal antibodies. Twenty ml of heparinized Macaca fascicularis blood or porcine blood was diluted with 280 ml of erythrocyte lysis buffer, incubated 3-5 minutes at room temperature, and centrifuged at 200 g for 5 minutes. The supernatant was discarded. The pellet was washed once in cold D-PBS containing 2% fetal bovine serum and the cells were counted by hemacytometer. Twenty ml of heparinized canine blood was diluted in two volumes of Waymouth's medium (Gibco) plus 2% nonessential amino acids (NEAA). Each 5 ml of blood solution was layered over 4 ml of Histopaque (Sigma) and centrifuged at 1000 g for 20 minutes at room temperature. Cells were collected from the interface, washed once in Waymouth's medium plus 2% NEAA, and counted as above. Each cell population was stained as described previously in Example 12C and analyzed by FACS. Anti-ICAM-R antibodies produced by hybridoma cell lines 26I10E, 46I12H, 63H4C, 56I4E and 63I12F specifically stained monkey PBL while the other antibodies did not. None of the antibodies specifically stained canine or porcine PBL. The monoclonal antibodies produced by the hybridoma cell lines 63A10E, 63E9G, 63E12C, 63G3G and 63H9H were not tested.

F. Purification

Hybridoma culture supernatants containing the anti-ICAM-R monoclonal antibodies listed in Table 11 in Example 21 were adjusted to 1.5M glycine, 3.0M NaCl, pH 8.9, and put over a 2 ml bed volume protein A column (Sigma). After washing with 1.5M glycine, 3M NaCl, pH 8.9, the column was eluted with 100 mM sodium citrate, pH 4.0. One ml fractions were collected into 100 .mu.l of 1.5M Tris, pH 8.8. Fractions containing antibody as determined by A.sub.280 were pooled and dialyzed against PBS.

G. Affinity

Nine of the purified anti-ICAM-R monoclonal antibodies were diluted serially and assayed in an ELISA format for binding to a fixed amount of soluble ICAM-R (Example 9) coated onto plastic. The results of the assay are presented in Table 3 below wherein high affinity binding was defined as 50% maximal binding at a monoclonal antibody concentration of less than 1 .mu.g/ml and low affinity binding was defined as 50% maximal binding at a monoclonal antibody concentration of greater than 1 .mu.g/ml.

F. Fab' Fragment Production

Fab' fragments were generated from the monoclonal antibodies produced by hybridomas 26E3D, 26I10E, 42D9B, 43H7C and 46D7E by the method described in Johnstone et al., p. 52 in Blackwell, Immunochemistry in Practice, Oxford Press (1982).

Example 12

ICAM-R specific monoclonal antibodies listed in Table 11 in Example 21 were tested for their ability to inhibit binding of JY cells (CD18.sup.+) to recombinant soluble human ICAM-R. Adhesion assays were performed as described in Example 9. Cells were treated with PMA and antibodies were then added at a final concentration of 10 .mu.g/ml. Data was collected from triplicate wells during three independent experiments. Total CD18-dependent binding was determined as the amount of adhesion blocked by a control anti-CD18 monoclonal antibody 60.3. The percentage of total CD18-dependent binding that was inhibited by each monoclonal antibody is shown below in Table 4 wherein the names assigned to monoclonal antibodies produced by each hybridoma are given and "ND" indicates the antibody was not tested. The monoclonal antibody names are used throughout the following examples instead of hybridoma designations.

Example 13

Monoclonal antibodies ICR-8.1 and ICR-1.1 were humanized as follows.

A. ICR-8.1 Humanization

1. RNA Isolation

ICR-8.1 hybridoma cells were grown in RPMI 1640 plus 10% FBS to about 4.times.10.sup.5 cells per ml. 4.times.10.sup.7 cells were harvested by centrifugation, washed twice in ice-cold PBS and lysed in 5 ml RNAStat (Tel-Test B Inc. Friendswood, Tex.). After extraction with chloroform, the RNA was precipitated with isopropanol, collected by centrifugation, washed in 70% ethanol, dried and dissolved in 600 .mu.l water. The yield was determined spectrophotometrically as 1.4 mg.

2. Isolation of ICR-8.1 V region cDNA

The heavy chain of the ICR-8.1 murine antibody is of the IgG.sub.1 subclass. V.sub.H cDNA was reverse transcribed from RNA primed with an oligonucleotide,

CG1FOR (SEQ ID NO: 31)

GGAAGCTTAGACAGATGGGGGTGTCGTTTTG,

which is based on amino acids 114-122 of the murine IgG.sub.1 constant region (Kabat et al., in Sequences of Immunological Interest, U.S. Department of Health and Human Services, NIH, 1991). The primer includes a HindIII site for directional cloning. The light chain of the murine antibody is of the kappa class. V.sub.K cDNA was reverse transcribed from RNA primed with an oligonucleotide,

LKC-1 (SEQ ID NO: 32)

GCTATCGGATCCACTGGATGGTGGGAAGATGGA,

which is based on amino acids 116-122 of the murine kappa constant region (Kabat et al., supra). The primer includes a BamHI site for directional cloning.

cDNA reactions in a volume of 50 .mu.l consisted of 5 .mu.g ICR-8.1 RNA, 50 mM Tris HCl pH 8.5, 8 mM MgCl.sub.2, 30 mM KCl, 1 mM DTT, 25 pmol CG1FOR or LKC-1, 250 .mu.M each of dATP, dCTP, dGTP and dTTP and 20 u RNase inhibitor (Boehringer Mannheim). Oligonucleotides were annealed to the RNA by heating at 70.degree. C. for 5 minutes and slowly cooling to 42.degree. C. Then, 11 u AMV reverse transcriptase (Boehringer Mannheim) was added and incubation at 42.degree. C. continued for 1 hour.

V.sub.H and V.sub.K cDNAs were amplified using a battery of primers based on the mature N-terminal regions of known murine V.sub.H and V.sub.K genes (Kabat et al., supra). For V.sub.H these oligonucleotides were:

HFR1-1 (SEQ ID NO: 33)

CGATACGAATTCSADGTRCAGCTKMAGGAGTCRGGA,

HFR1-2 (SEQ ID NO: 34)

CGATACGAATTCSAGGTYCARCTKCARCARYCTGG,

HFR1-3 (SEQ ID NO: 35)

CGATACGAATTCGARGTGAAGCTKSWSGAGWCTGG,

HFR1-4 (SEQ ID NO: 36)

CGATACGAATTCAGGTSMARCTGCAGSAGTCWG, and

HFR1-6 (SEQ ID NO: 37)

CGATACGAATTCSAGGTSMARCTGCAGSARHC.

These primers include an EcoRI restriction site (underlined) for directional cloning. For V.sub.K the primers were:

LFR1-1 (SEQ ID NO: 38)

CGATACGAATTCSAAAWTGTKCTSACCCAGTCTCCA,

LFR1-2 (SEQ ID NO: 39)

CGATACGAATTCGACATTGTGMTGWCMCARTCTCC,

LFR1-3 (SEQ ID NO: 40)

CGATACGAATTCGATRTTKTGATGACYCARRCTSCA, and

LFR1-4 (SEQ ID NO: 41)

CGATACGAATTCGAYATYSWGATGACMCAGWCTMC.

The N-terminal V region primers were used in concert with the CG1FOR and LKC-1 primers to amplify the V.sub.H and V.sub.K cDNAs by PCR. The mixtures for the PCR consisted of 5 .mu.l cDNA, 10 mM Tris HCl pH 8.3, 50 mM KCl, 1.5 mM MgCl.sub.2, 250 .mu.M each of dATP, dCTP, dGTP, dTTP, 0.01% (v/v) Tween 20, 0.01% (w/v) gelatin, 0.01% (v/v) NP-40, 25 pmol CG1FOR or LKC-1, 25 pmol HFR1-1, HFR1-2, HFR1-3, HFR1-4 or HFRH1-6, or LFR1-1, LFR1-2, LF1-3 or LFR1-4 and 2.5 u Thermalase (IBI, New Haven, Conn.) in a reaction volume of 50 .mu.l. Samples were subjected to 25 thermal cycles of 94.degree. C., 30 seconds; 50.degree. C., 30 seconds; 72.degree. C., 1 minute. Aliquots were analysed by agarose gel electrophoresis. Products of the expected size were found for all combinations of V.sub.H primers and all combinations of V.sub.K primers except for LKC-1 and LFR1-1.

The V.sub.H DNA was cut with EcoRI and HindIII, and the V.sub.K DNA cut with EcoRI and BamHI. Both DNA types were cloned into M13 BM21 (Boehringer Mannheim) and M13 tg 130 and the DNA sequence of the inserts determined. Full-length functional V region sequences were obtained. By comparison with other murine V region sequences the ICR-8.1 murine V.sub.H and V.sub.K genes were members of murine heavy chain subgroup IIA and murine kappa subgroup II (Kabat et al., supra). In order to determine the authentic N-terminal sequences of both heavy and kappa V regions, PCRs were done with oligonucleotides based on the known signal sequences of murine heavy chain subgroup IIA and kappa subgroup II. SEQ ID NOs: 42 and 43, and 44 and 45 respectively show the entire DNA and amino acid sequences of ICR-8.1 murine V.sub.H and V.sub.K.

3. Humanized ICR-8.1 V.sub.H

The murine V.sub.H sequence was aligned with consensus sequences of human V.sub.H subgroups (Kabat et al., supra) and was found to be most homologous to human subgroup I. Therefore, a consensus human subgroup I was chosen as the framework for receiving the murine complementarity determining regions (CDRs).

The template for humanization was single-stranded DNA encoding a consensus sequence of human subgroup I containing irrelevant CDRs cloned in M13. The M13 clone was grown in E. coli RZ1032 (dut.sup.- ung.sup.-) so that its DNA contained uracil instead of thymine. For grafting the CDR sequences onto the human framework region (FR), oligonucleotides were synthesised, each encoding a CDR and flanked at both ends by nucleotides corresponding to the human template DNA. The sequences of the mutagenic primers were:

CDR1 (SEQ ID NO: 46)

GGCCTGTCGCACCCAGAGTATGATGCAGTCAGTGAAGR TGTATCC,

CDR2 (SEQ ID NO: 47)

TGTGTCCRCGGTAATGGTCACTCTGCCCTTGAATTTCA GATTATAGGTAGTAGTACCAAAGTAAGGATTAATTTTTCCC ATCCATTCGAG,

CDR3 (SEQ ID NO: 48)

TCCTTGGCCCCCAGTAGTCCATAGCATCTGGGTAGGCC TCCTTTCTTGCACAGTAATACACGG

Ten pmol of each oligonucleotide was phosphorylated in 25 .mu.l 100 mM Tris HCl pH 8.0, 10 mM MgCl2, 7 mM DTT, 1 mM ATP and 5 u polynucleotide kinase (Boehringer Mannheim) for 1 hour at 37.degree. C. Primers were annealed to the template in a 20 .mu.l reaction mixture consisting of 0.2 pmol template, 2 pmol each phosphorylated oligonucleotide, 100 mM Tris HCl pH, 10 mM MgCl.sub.2 and heating to 90.degree. C. for 10 seconds, followed by rapid cooling to 70.degree. C. and slow cooling to room temperature. To the annealed DNA was added 2 .mu.l 0.1M DTT, 2 .mu.l 0.5M Tris HCl pH 8.0, 0.1M MgCl.sub.2, 2 .mu.l 0.1M ATP, 1 .mu.l 6.25 mM each of dATP, dCTP, dGTP, and dTTP, 2.5 u T7 DNA polymerase (United States Biochemicals, Cleveland, Ohio), 0.5 u T4 DNA ligase (Boehringer Mannheim) and 3.7 .mu.l water. Incubation was for 2 hours at 22.degree. C. The DNA was ethanol precipitated, washed, dried and dissolved in 50 .mu.l 60 mM Tris HCl pH 8.0, 1 mM EDTA, 1 mM DTT, 0.1 mg/ml BSA and 1 u uracil-DNA glycosylase (Boehringer Mannheim) and incubated at 37.degree. C. for 1 hour. Phosphodiester bonds at apyrimidinic sites were cleaved by making the sample 0.2M NaOH and incubating at 22.degree. C. for 5 minutes. The sample was neutralized by the addition of 0.5 vol 7.5M ammonium acetate and the DNA precipitated with ethanol. The washed and dried DNA was finally dissolved in 20 .mu.l 10 mM Tris HCl, 1 mM EDTA pH 8.0. The sample containing mutated DNA was amplified by PCR in a reaction mixture containing 2 .mu.l mutant DNA mix, 250 .mu.M each of dATP, dGTP, dCTP and dTTP 10 mM Tris HCl pH 8.3, 50 mM KCl, 1.5 mM MgCl.sub.2, 0.01% (v/v) Tween 20, 0.01% (w/v) gelatin, 0.01 (v/v) NP-40, 25 pmol M13 universal sequencing primer (GTAAAACGACGGCCAGT, SEQ ID NO: 49), 25 pmol M13 reverse sequencing primer (AACAGCTATGACCATG, SEQ ID NO: 50) and 2.5 u Thermalase (IBI). Samples were subjected to 15 thermal cycles of 94.degree. C., 30 seconds; 50.degree. C., 30 seconds; 72.degree. C., 45 seconds.

Because humanized V regions often exhibit lower affinities than their progenitor antibodies, the primers used for the mutagenesis had the potential to encode either threonine or serine at position 28 and alanine or valine at position 71, sites which can play a role in antigen binding. The DNA was cut with NotI and BamHI and cloned into M13 BM21 for sequence determination. An M13 clone (designated M13 HuVHV) containing the fully humanized V region with no spurious mutations and with Thr28 and Val71 was selected. A version containing Ala71 was made by PCR overlap and extension Ho et al., Gene, 77: 51-59 (1989)!. In the first PCRs the primers were oligonucleotide 42 (ACCATTACCGCGGACACATCCAC, SEQ ID NO: 51) with the M13 universal sequencing primer and mutagenic CDR2 oligonucleotide primer with the M13 reverse sequencing primer. Reaction conditions for the PCR were as in Section A.2 above except that 5 pmol of primer 42 and the CDR2 primer and 1 u Thermalase were used and there were 15 cycles of 94.degree. C., 30 seconds; 40.degree. C., 30 seconds; 72.degree. C., 30 seconds. The two product DNAs were joined in a second PCR which consisted of 1 .mu.l DNA from the first PCRs in 50 .mu.l 20 mM Tris HCl pH 8.2, 10 mM KCl, 6 mM (NH.sub.4).sub.2 SO.sub.4, 2 mM MgCl.sub.2, 0.1% (v/v) Triton X-100, 10 .mu.g/ml BSA, 250 mM each of dATP, dCTP, dGTP, dTTP 25 pmol each of M13 universal and reverse sequencing primers and 1 u Pfu DNA polymerase (Stratagene). The sample was subjected to 15 thermal cycles of 94.degree. C., 30 seconds; 50.degree. C., 30 seconds; 75.degree. C., 30 seconds. The DNA was cut with NotI and BamHI, cloned into M13 BM21 and the required clone containing the Ala 71 mutation selected by DNA sequencing. This DNA was designated ICR-8.1 HuVH and its sequence and deduced translation product are shown in SEQ ID NOS: 52 and 53. In addition, to reduce as far as possible the potential immunogenicity of the humanized antibody, a V.sub.H region was made with only the structural loop residues of murine origin together with CDR residures, which from cystallographic data of other antibodies, indicate a role in antigen-binding. Thus, residues at V.sub.H positions 60, 61 and 64 were changed to consensus residues of human subgroup I. The mutations were Asn60 to Ala, Leu61 to Gln and Lys64 to Gln. The mutations were introduced by PCR overlap and extension (Ho et al., supra) using oligonucleotides 51 (TACTACCTATGCTCAGAAATTCCAGGGCAGAG, SEQ ID NO: 54) and 52 (CTCTGCCCTGGAATTTCTGAGCATAGGTAGTAG, SEQ ID NO: 55), essentially as described for introduction of the Ala71 mutation. The amplified DNA was cut with NotI and BamHI, cloned into M13 BM21 and a clone containing the desired mutation was identified by DNA sequencing. This DNA and encoded protein were designated ICR-8.1 miHuVH.

Two additional versions based on miHuVH were made both containing the amino acid change Val2 to Ile2 (designated miHuVHI) and one additionally containing the amino acid change Ser7 to Thr7 (designated miHuVHIT). The miHuVHI version was made using the miHuVH DNA as a template for mutagenesis via PCR overlap and extension with oligonucleotides 81 (CACAGGTGTCCACTCCCAGATCCAGCTGG, SEQ ID NO: 56) and 82 (TGGGAGTGGACACCTGTGGAGAGAAAGGCAAAGTGG, SEQ ID NO: 57). The miHuVHIT version was similarly made using oligonucleotides 82 and 84 (CACAGGTGTCCACTCCCAGATCCAGCTGGTGCAGACTGGGGC, SEQ ID NO: 58).

4. Humanized ICR-8.1 V.sub.K

The murine ICR-8.1 murine V.sub.K amino acid sequence was aligned with consensus sequences of human V.sub.K groups and it showed greatest homology (79%) to human subgroup II. An initial humanized V.sub.K was made by taking an M13 clone containing the murine V.sub.K DNA and mutating the murine framework DNA to a consensus sequence of human subgroup II (Kabat et al. supra). Each framework region was mutated by PCR using:

FR1 primer 33 (SEQ ID NO: 59)

GCTCTCCAGGAGTGACAGGCAGGG,

FR1 primer 36 (SEQ ID NO: 60)

TTGCGGCCGCAGGTGTCCAGTCCGACATTGTAATGACCCAGTC TCCACTCTC,

FR1 primer 32 (SEQ ID NO: 61)

TCACTCCTGGAGAGCCAGCCTCCATCTCTTGCAGA,

FR3 primer 35 (SEQ ID NO: 62)

CCTCAGCCTCCACTCTGCTGATCTTGAGTGT,

FR3 primer 34 (SEQ ID NO: 63)

AGAGTGGAGGCTGAGGATGTGGGAGTTTATTACTGCTCTC, and

FR4 primer 37 (SEQ ID NO: 64)

TTGGATCCTAAGTACTTACGTTTTATTTCCACCTTGGTCCCCT GTCCG.

Mixtures for the PCR consisted of 0.2 .mu.l supernatant of an M13 clone containing the ICR-8.1 murine V.sub.K 25 pmol each oligonucleotide with other components as listed above in Section A.2. The samples were subjected to 15 thermal cycles of 94.degree. C., 30 seconds; 40.degree. C., 30 seconds; 72.degree. C., 30 seconds. Product DNAs were of the expected size and were joined in a second PCR to obtain full-length V.sub.K using oligonucleotides 36 and 37. The product DNA was cut with NotI and BamHI and cloned into M13 BM21. A fully humanized Vk with no spurious mutations was identified by DNA sequencing. The sequences of this DNA and encoded product designated ICR-8.1 HuVK are shown in SEQ ID NOs: 65 and 66.

5. Chimeric ICR-8.1 Antibodies

Chimeric antibodies consisting of murine V domains with human constant domains were made for use as controls in subsequent assays wherein binding affinities of CDR-grafted V regions were compared to parental antibodies using a common anti-human Ig Fc antibody as a detecting reagent. NotI and BamHI sites were introduced into an M13 clone containing ICR-8.1 murine VH cDNA obtained in Section A.2 using in PCR oligonucleotides 27 (TTGCGGCCGCAGGTGTCCAGTCCGAGGTGCAACTGCAGCAGTCTGGAC, SEQ ID NO: 67) and 28 (TGGATCCAAGGACTCACCTGAGGAGACGGTGACTGAGGTTCC, SEQ ID NO: 68). Conditions for the PCR were as described in Section A.2 except that there were 20 thermal cycles of 94.degree. C., 30 seconds; 30.degree. C., 30 seconds; 72.degree. C., 45 seconds. The amplified DNA was cut with NotI and BamHI, cloned into M13 BM21 and a clone containing the desired sequence identified by DNA sequencing. This V.sub.H differs from the murine V.sub.H in that amino acids 2 and 7 are valine and serine respectively, instead of the authentic residues Ile an Thr. This V.sub.H is designated ICR-8.1 MuVHVS.

Residue 28 in the murine V.sub.H is a serine and is part of the structural loop encompassing CDR1 and might be expected to play a role in antigen binding. The corresponding residue in the human subgroup I consensus FR is Thr. In order to minimize the reintroduction of murine residues into the humanized V.sub.H the role of Ser28 in binding was assessed by making a murine V.sub.H region containing threonine at this position. Thr28 was introduced into MuVHVS DNA in a PCR essentially as described above in Section A.3 using in the first PCRs oligonucleotides 40 (TTCTGGTTATACTTCACTGACT, SEQ ID NO: 69) with the M13 universal sequencing primer and oligonucleotide 41 (AGTCAGTGAAAGTATAACCAGAA, SEQ ID NO: 70) with the M13 reverse sequencing primer. The two amplified DNAs were joined in a second PCR and the product subsequently cut with NotI and BamHI, cloned into M13 BM21 and a clone containing the Thr28 mutation identified by DNA sequencing. This DNA is designated ICR-8.1 MuVHVST.

A true chimeric V.sub.H (designated MuVH) was constructed containing Ile2 and Thr7. This was accomplished by PCR overlap/extension mutagenesis using oligonucleotides 79 (CTCCGAGATCCAGCTGCAGCAGACTGGACC, SEQ ID NO: 71) and 91 (CAGCTGGATCTCGGAGTGGACACCTGTGGAGAGAAAGGCAAAGTGG ATG, SEQ ID NO: 72). V.sub.K for the chimeric kappa chain was constructed from an M13 clone containing ICR-8.1 V.sub.K cDNA obtained in Section A.3 above. Appropriate NotI and BamHI restriction sites were introduced by PCR using oligonucleotides 43 (TTGCGGCCCGCAGGTGTCCAGTCCGACGCTGTGACCCAAAC, SEQ ID NO: 73) and 44 (TTGGATCCTAAGTACTTACGTTTTCCAGCTTGGT, SEQ ID NO: 74) with conditions as described in Section A.2. Amplified DNA was cut with NotI and BamHI, cloned into M13 BM21 and a clone containing the correctly mutated DNA identified by DNA sequencing. This DNA was designated ICR-8.1 MuVK and encodes the authentic N-terminal amino acids.

6. Vectors for the Expression of Recombinant ICR-8.1 Antibodies

The heavy and light expression vectors were based on pSVgpt and pSVhygHuCK respectively Orlandi et al., Proc. Natl. Acad. Sci. USA, 86: 3833-3837 (1989)! modified to include a NotI site in the intron between the two exons encoding the signal peptide. Both vectors contain an immunoglobulin promoter and enhancer, signal sequence, appropriate splice sites, the SV40 promoter and the gpt or hygromycin resistance gene for selection in mammalian cells and genes for replication and selection in E. coli. For expression of the humanized or chimeric heavy chain the NotI-BamHI fragment containing the humanized V.sub.H (HuVH, HuVHV or miHuVH) or murine VH (MuVHVS or MuVHVST) was cloned into NotI-BamHI cut pSVgpt. For expression of MuVH, miHuVHI and miHuVHIT heavy chains the V regions were cloned in as HindIII-BamHI fragments into HindIII-BamHI cut pSVgpt. A human IgG.sub.4 constant region Flanagan et al., Nature, 300:709-713 (1982)! was then added as a BamHI fragment. For the expression of the humanized or chimeric kappa chain the NotI-BamHI fragment containing the humanized V.sub.K (HuVK) or murine V.sub.K (MuVK) was cloned into NotI-BamHI cut pSVhygHuCK which contains DNA encoding the human kappa chain constant region Hieter et al., Cell, 22:197-207 (1980)!.

7. Expression of Recombinant ICR-8.1 Antibodies

The host for the expression of recombinant antibodies was either the mouse myeloma NSO (ECACC 85110503) or the rat myeloma YB2/0 (ATCC CRL 1662) and were grown in RPMI 1640 or DMEM plus 10% FBS. Cells in the logarithmic phase of growth were harvested by centrifugation and resuspended in medium at 10.sup.6 cells per ml. 0.5 ml aliquots of cells were mixed with 10 .mu.g of PvuI cut heavy chain expression vector and 20 .mu.g of PvuI cut light chain expression vector in ice for 5 minutes in an electroporation cuvette. The cells were electroporated at 170 V, 960 .mu.F using a GenePulser apparatus (Biorad, Richmond, Calif.). After 20 minutes in ice the cells were added to 20 ml growth medium and allowed to recover for 24-48 hours. At this time the cells were put into 50 ml growth medium containing 0.8 .mu.g/ml mycophenolic acid and 200 .mu.g/ml xanthine and 250 .mu.l aliquots distributed into two 96-well plates. These were incubated for 10-14 days at which time gpt.sup.+ colonies were visible. Supernatant from the wells was then assayed for the presence of human IgG.sub.K antibodies. Micro-titer plates (Immulon 4, Dynatech) were coated with goat anti-human IgG (Fc) or goat anti-human IgG (H+L) antibodies (Jackson Immunoresearch) and culture supernatant applied for 1 hour at ambient temperature. After washing with PBST, captured human antibody was detected with peroxidase-conjugated goat anti-human kappa (Sigma) or peroxidase-conjugated goat anti-human IgG (Fc) (Jackson Immunoresearch) antibodies. The substrate for peroxidase activity was o-phenylenediamine at 0.4 mg/ml in 50 mM citrate buffer pH 5.0 and 0.003% (v/v) H.sub.2 O.sub.2. Reactions were stopped by the addition of 50 .mu.l 12.5% (v/v) sulphuric acid. The absorbance at 490 nm was then measured.

8. Purification of Recombinant ICR-8.1 Antibodies

Transfectants secreting recombinant ICR-8.1 antibody were expanded and grown to saturation in RPMI 1640 plus 2% FBS or DMEM plus 11% FBS in 175 cm.sup.2 flasks. Culture medium was made 0.1M Tris HCl pH 8.0 and was stirred overnight with protein A agarose (Boehringer Manngeim) at about 1 ml per liter medium. The protein A agarose beads were packed into a small column, washed with 10 ml 35 mM Tris HCl pH 8.0, 150 mM NaCl, 0.1% (v/v) Tween 20 and then with 4 ml 50 mM citrate pH 5.0. Antibody was eluted with 1 ml amounts of 50 mM citrate pH 3.0, 0.02% (v/v) Tween 20. Samples were immediately neutralized with 1M Tris HCl pH8.0 and the A.sub.280nm measured. Antibody containing fractions were pooled and dialysed against PBS. Concentrations were determined spectrophotometrically.

9. Binding of Humanized ICR-8.1 Antibodies to ICAM-R

The binding of recombinant antibodies to ICAM-R was assessed by ELISA. Wells of a micro-titer plate (Immulon 4, Dynatech) were coated with 50 ng per well baculovirus-produced soluble human ICAM-R in 50 or 100 .mu.l 35 mM sodium carbonate, 15 mM sodium bicarbonate pH 9.2 at 37.degree. C. for 1 hour or 16 hours at 4.degree. C. Purified antibody or culture supernatant containing recombinant antibody was added and its binding detected as described in section A.7 above.

Antibodies comprising the following combinations of humanized and chimerized chains showed equivalence in binding to ICAM-R: MuVHVS/MuVK, MuVHVST/MuVK, HuVH/MuVK, HuVHV/MuVK, HuVH/HuVK, miJuVH/MuVK and miHuVH/HuVK. However, antibodies containing the humanized heavy chain with the Ile2 mutation show an approximate 2-fold improvement in binding over the aforementioned antibodies. In experiments in blocking the binding of biotinylated murine ICR-8.1 to ICAM-R, the miHu/VHI/HuVK and miHuVHI/MuVK antibodies compete about 2-fold less well than the murine antibody itself.

B. Humanization of ICR-1.1

1. RNA Isolation

ICR-1.1 hybridoma cells were grown to subconfluency. One.times.10.sup.7 cells were lysed in 2.5 ml 1M guanidine thiocyanate, 80 mM sodium acetate pH4.0, 6.25M sodium citrate, 40% (v/v) phenol, 16% (v/v) chloroform, 0.28% (v/v) 2-mercaptethanol and 0.125% (v/v) Sarkosyl. After isopropanol precipitation the RNA was collected by centrifugation, dissolved in water, ethanol precipitated, washed in 70% ethanol, dried and dissolved in water. Yield of RNA measured spectrophotometrically was 0.16 mg.

2. Isolation of ICR-1.1 V Region cDNA

Heavy and light chain cDNA was made from random-primed ICR-1.1 RNA. The reaction mixture was as in Section A.2 except that 10 .mu.g RNA, 200 ng random primers pd(N).sub.6 ! and 44 u AMV reverse transcriptase were used and incubations were at 65.degree. C. for 2 minutes, and after addition of the enzyme, at 42.degree. C. for 1.5 hours.

V.sub.H cDNA was amplified by PCR using a primer,

HG2A-1 (SEQ ID NO: 75)

GCTATCGGATCCGGARCCAGTTGTAYCTCCACACAC based on amino acids 127-136 of the murine IgG2A constant region (Kabat et al., supra) and including a BamHI site for cloning purposes, together with primers HFR1-1, -2, -3, -4, and -6 which are set out above in Section A.2 and

HFR1-5 (SEQ ID NO: 76)

CGATACGAATTCSAGGTSMARCTGCAGSAGTCT.

V.sub.K cDNA was amplified using LKC-1 with either LFR1-1, LFR1-2, LFR1-3 or LFR1-4. The sequences of these primers are listed in Section A.2. The conditions for the PCR were essentially as in Section A.2 except that the samples were subjected to 25 thermal cycles of 95.degree. C., 1 minute; 55.degree. C., 2 minutes; 72.degree. C., 3 minutes. V.sub.H DNA of the expected size was obtained with HFR1-1, -2, -3, -4 with HG2A-1, V.sub.K DNA of the expected size was obtained using LKC-1 with LFR1-2, -3 and -4. Amplified DNAs were cut with BamHI and EcoRI, cloned into Bluescript SK+ and sequenced. Full length functional V region sequences were obtained. By comparison with other murine V regions, the ICR-1.1 murine V.sub.H and V.sub.K genes were members of murine heavy chain subgroup IIB and murine kappa subgroup VI (Kabat et al., supra). In order to determine the authentic N-terminal sequences of both heavy and kappa V regions, PCRs were performed with oligonucleotides based on known signal sequences of the murine heavy and kappa chain subgroups.

The murine ICR-1.1 V.sub.H sequence and its translation product are shown in SEQ ID NOs: 77 and 78, while the murine V.sub.K sequence and its translation product are shown in SEQ ID NOs: 79 and 80.

3. Humanized ICR-1.1 V.sub.H

The murine ICR-1.1 V.sub.H amino acid sequence was aligned with consensus sequences of human V.sub.H subgroups where it shows closest homology (66%) to human subgroup I (Kabat et al., supra). Therefore, a consensus FR amino acid sequence of human heavy subgroup I was used as the recipient for the ICR-1.1 CDRs. The template for the mutagenesis was uracil containing single-stranded DNA encoding a consensus sequence of human heavy chain subgroup I containing irrelevant CDRs and cloned in M13. Mutagenesis reactions were as described above in Section A.3 using mutagenic oligonucleotides for

CDR1 (SEQ ID NO: 81)

CCTGTCGCACCCAGTGCATCCAGTAAACAGTGAAGGTGTATCC;

CDR2 (SEQ ID NO: 82)

GTCCGCGGTAATGGTCACTCTGTCCTGGAACCTCTGATTGTACTC AGTATAATCAGTGTTAGGATTAATGTATCCMATCCACTCGAGCCC;

and

CDR3 (SEQ ID NO: 83)

GGCCCCAGTAGTCCAAACCATAGGAGTTACCCCCCCATCTGG CACAGTAATACACGG.

Amplified DNA was cut with NotI and BamHI, cloned into M13 BM21 and a humanized VH clone identified by DNA sequencing. This clone was designated M13 ICR-1.1 HuVHI and contains the murine framework amino acid Ile48. Another variant that has the human amino acid methionine at position 48 was made by the PCR overlap extension method as described in Section A.3 using in the first PCRs oligonucleotide 17 (CTCGAGTGGATGGGATACATTAA, SEQ ID NO: 84) and the CDR2 oligonucleotide with, respectively, the M13 universal and reverse sequencing primers. DNA after the second joining PCR was cut with NotI and BamHI, cloned into M13 and a clone containing DNA encoding Met48 was identified by DNA sequencing. This DNA was designated ICR-1.1 HuVH and its DNA and encoded amino acid sequence are shown in SEQ ID NOs: 85 and 86.

An additional humanized V.sub.H was made with a Ser73 to Lys73 mutation in framework region 3. This HuVHK was made by mutagenesis using PCR overlap/extension of HuVH DNA and oligonucleotides 8 (GTGGATCCAAGGACTCACCTGAGGAG, SEQ ID NO: 87) with 89 (ACCGCGGACAAATCCACGAG, SEQ ID NO: 88) and 90 (CTCGTGGATTTGTCCGCGGT, SEQ ID NO: 89) with 92 (CACAGGTGTGTCCACTCCCAAGTCCAGC, SEQ ID NO: 90). The two products were joined in a second PCR using oligonucleotides 8 and 92. The product of this PCR was cut with HindIII and BamHI, cloned into M13 and the desired clone identified by DNA sequencing.

4. Humanized ICR-1.1 V.sub.K

The murine ICR-1.1 V.sub.K amino acid sequence was compared with consensus sequences of human V.sub.K subgroups. It shows closest homology (62%) to human V.sub.K subgroup I (Kabat et al., supra). Therefore, a consensus FR amino acid sequence of human kappa subgroup I was chosen as acceptor onto which the murine ICR-1.1 CDRs were grafted. The template for the mutagenesis was uracil containing single-stranded DNA encoding a consensus sequence of human kappa subgroup I (Kabat et al., supra) containing irrelevant CDRs and cloned in M13. Mutagenesis reactions were as described in section using the following mutagenic primers:

CDR1 (SEQ ID NO: 91)

TTCTGTTGGTACCAGTAAATGTAACTTACACTTGAGCTGGCACTG CAAGTGATGGTGAC;

CDR2 (SEQ ID NO: 92)

TTGATGGGACCCCAGAAGCCAGGTTGGATGTAAGATAGATCA GGAGC;

CDR3 (SEQ ID NO: 93)

CCCCTGGCCGAACGTGAGTGGGATACTCTTCCACTGCTGACA GTAGTAAGTTG.

One mutagenesis reaction included a fourth oligonucleotide (GTGAGAGTGTAGTCTGTCC, SEQ ID NO: 94) which would mutate Phe71 to tyrosine. Amplified DNA was cut with NotI and BamHI, cloned into M13 BM21 and humanized V.sub.K DNAs identified by DNA sequencing. The clones obtained were designated M13 ICR-1.1 HuVK and M13 ICR-1.1 HuVKY (containing Tyr71). The DNA and deduced amino acid sequences of HuVK are shown in SEQ ID NOs: 95 and 96.

An additional humanized Vk was made containing the murine N-terminal amino acids Glu1 Val3 Leu4. Mutagenesis of HuVK to HuVHQVL was accomplished by PCR using oligonucleotide 110 (CACAGGTGTCCACTCCCAAATCGTGCTGACCCAGTCTCCATCCTCCC, SEQ ID NO: 97) and 68 (TTAAAGATCTAAGTACTTACGTTTGATCTC, SEQ ID NO: 98). The DNA product was joined to DNA amplified with oligonucleotides 12 and 82 containing the immunoglobulin promoter and signal sequence. The full-length DNA was cut with HindII and BamHI, cloned into M13 and the desired clone identified by DNA sequencing.

5. Chimeric ICR-1.1 Antibodies

Chimeric antibodies consisting of murine V domains with human constant domains were constructed to act as appropriate controls. For construction of the murine VH as a HindII-BamHI fragment containing the immunoglobulin promoter and signal sequence, an M13 clone with the murine VH was subject to PCR with oligonucleotides 92 (CACAGGTGTCCACTCCCAAGTCCAGC, SEQ ID NO: 99) and 93 (TTGGATCCAAGGACTCACCTGAGGAGACGGTGACTGAGGT, SEQ ID NO: 100). The PCR product was joined to that resulting from the amplification of the immunoglobulin promoter and signal sequence using oligonucleotides 12 and 82. The product DNA was cut with HindIII and BamHI and the correct clone identified by DNA sequencing.

For construction of the chimerized V.sub.K, an M13 clone containing MuVK was subject to a PCR using oligonucleotides 87 (TTGGATCCTAAGTACTTACGTTTCAGCTCCAGCTTGGTCCCAG 3', SEQ ID NO: 101) and 88 (CAGGTGTCCACTCCCAAATTGTTCTCACCCAGTCTCCAGCACTCATG, SEQ ID NO: 102). The product was also joined to DNA amplified with 12 and 82. The resulting DNA was cut with HindIII and BamHI, cloned into M13 and the desired clone identified by DNA sequencing.

6. Vectors for the Expression of Recombinant ICR-1.1 Antibodies

The CDR-grafted humanized heavy and kappa chain V region DNAs were cut with NotI and BamHI and cloned into the pSVgpt and pSVhyg HuCK expression vectors as described above in Section A.7. Where the recombinant V regions were constructed with the immunoglobulin promoter and signal sequence these DNAs were cut with HindIII and BamHI and cloned into HindIII and BamHI cut pSVgpt and pSVhygHuCK as appropriate. A human IgG4 constant region was added to the heavy chain vectors as described above in section A.6.

7. Expression of Humanized ICR-1.1 Antibodies

Transfection of vectors into YB2/0 and NS0 cells and the isolation of antibody-secreting clones was as described in above section A.7.

8. Purification of Humanized ICR-1.1 Antibodies

Recombinant ICR-1.1 antibodies were purified as described above in section A.8.

9. Binding of Recombinant ICR-1.1 Antibodies to ICAM-R

Binding of humanized or chimeric ICR-1.1 antibodies to soluble baculovirus produced human ICAM-R (Example 9) was determined by ELISA as described in Section A.9 above. Analysis of the binding of the chimeric, humanized and hybrid antibodies indicates that compared with their murine progenitors the HuVH and HuVK versions show 2-3 fold and 2-fold deficits respectively. The HuVH/HuVK antibody is thus 5-10 fold less effective in binding to ICAM-R than the chimeric (MuVH/MuVK) antibody. Antibodies containing HuVHI, HuVKY or HuVKQVL show no increase in affinity whereas those containing HuVHK show a possible 1.5-fold improvement.

Example 14

FACS-based competition assays utilizing human peripheral blood leukocytes or SKW3 cells (both ICAM-R expressing cells) indicate that monoclonal antibodies ICR-4.2 and ICR-1.1 are immunologically reactive with distinct epitopes of ICAM-R.

In the assays, human peripheral blood leukocytes (PBL) obtained by Ficoll Hypaque centrifugation of normal peripheral blood were washed twice in ice cold FACS buffer (PBS containing 0.1% sodium azide and 1% bovine serum albumin) and 2.times.10.sup.5 cells were incubated in triplicate polypropylene tubes with 5 .mu.g of each of the following antibodies ICR-1.1, ICR-4.2, and control isotype IgG (Sigma). All tubes containing the first stage antibodies were then incubated for 30 minutes at 4.degree. C. and washed twice in cold FACS buffer. To each triplicate tube, 5 .mu.g of each of the following second stage antibodies were added: biotinylated-ICR-1.1, biotinylated-ICR-4.2, biotinylated-anti-rat CD4 (negative control). All second stage antibodies were biotinylated according to standard procedures as described in Example 12 and all tubes were then incubated for an additional 30 minutes at 4.degree. C. before washing twice in FACS buffer. Five ul of a 1:10 dilution of Strepavidin-phycoerythrin (Southern Biotechnology, Birmingham, Ala.) was then added to each tube containing 50 ul FACS buffer and all tubes were incubated for 30 minutes at 4.degree. C. Finally, all tubes were washed twice in FACS buffer and analyzed by flow cytometry (FACScan, Becton-Dickinson).

While monoclonal antibody ICR-4.2 blocked binding of biotinylated- ICR-4.2 to ICAM-R on PBL, it did not block binding of monoclonal antibody ICR-1.1. Similarly, monoclonal antibody ICR-1.1 did block binding of biotinylated-ICR-1.1 but did not block binding of monoclonal antibody ICR-4.2. These results indicate that the two antibodies recognize distinct epitopes on ICAM-R. Equivalent results were obtained when using the human cell line SKW3 as follows. SKW3 cells were labelled with either 1 .mu.g of antibody ICR-1.1 or ICR-4.2, washed in FACS buffer and incubated with 1 .mu.g biotinylated-ICR-1.1 or biotinylated ICR-4.2. All tubes were then washed in FACS buffer, incubated with Strepavidin-phycoerythrin for an additional 30 minutes at 4.degree. C. and analyzed by FACScan.

In the assays, if an unlabelled antibody (the "blocking" antibody) prevented the labelled antibody from binding to ICAM-R, it indicates that the unlabelled antibody "competes" with the labelled antibody for binding to ICAM-R and that the two antibodies recognize the same, sequential or sterically overlapping epitopes on ICAM-R. A variation of the competition assay in which unlabelled antibody is used to "compete away" binding of a labelled antibody may also be utilized to determine if two antibodies recognize the same, sequential or sterically overlapping epitopes.

The specific ICAM-R epitopes recognized by the various monoclonal antibodies of the invention can be mapped by four different methods.

A. Epitope Mapping Using The Multipin Peptide Synthesis System

The first method for mapping linear epitopes recognized by the ICAM-R specific antibodies of the invention utilized the Multipin Peptide Synthesis System (Chiron Mimotopes Pty. Ltd., Victoria, Australia) which places ten amino acid peptides representing overlapping segments of the protein of interest on the surface of a series of plastic pins. A modified ELISA test is performed to determine binding of a monoclonal antibody to each peptide.

The ELISA to determine binding of the monoclonal antibodies to ICAM-R peptides was run as follows. The pins were placed in five 96-well plates containing 200 .mu.l per well blocking buffer (2% weight/volume BSA, 0.1% volume/volume Tween 20, 0.01M PBS, pH 7.2) and incubated for one hour at 20.degree. C. with agitation. The pins were transferred to plates with 175 .mu.l per well of undiluted anti-ICAM-R monoclonal antibody supernatant and incubated overnight at 4.degree. C. with agitation. The pins were then washed four times with 0.01M PBS, pH 7.2 (10 minutes/wash at 20.degree. C. with agitation) and placed in plates containing 175 .mu.l per well HRP-Goat anti-mouse IgG (H+L) (Kirkegaard and Perry Laboratory Inc., Gaithersburg, Md.) diluted to an appropriate concentration in conjugate diluent (1% volume/volume sheep serum, 0.1% volume/volume Tween 20, 0.1% weight/volume sodium caseinate and 0.01M PBS). The plates were agitated for one hour at 20.degree. C., and washed four times with 0.01M PBS. The pins were transferred to plates containing ABTS substrate solution 0.5 mg/ml ABTS, 0.01% weight/volume H.sub.2 O.sub.2 in substrate buffer (17.9 g/L Na.sub.2 HPO.sub.4 H.sub.2 O, 16.8 g/L citric acid monohydrate, pH 4.0)! for 45 minutes at 20.degree. C. with agitation and then the plates were read at 410/495 nm.

Relative reactivity with individual pins was determined after normalizing results for differences in immunoglobulin concentrations in anti-ICAM-R and control hybridoma supernatants and reactivities of positive controls between assays. Mouse IgG levels for each supernatant had been determined by antibody capture ELISA as follows. Immulon 4 plates were coated and washed as described in Example 10C. Fifty .mu.l/well of culture supernatant diluted in PBST or known concentrations in doubling dilutions in PBST of mouse IgG.sub.1 and IgG.sub.2a (MOPC-21, and UPC-10) (Sigma)! was added to the plate. After incubating for 1 hour at room temperature and washing 3 times with PBST, horseradish peroxidase conjugated goat anti-mouse IgG(fc) (Jackson ImmunoResearch, West Grove, Pa.) was diluted 1:2000 for mouse IgG.sub.1 and 1:1000 for IgG.sub.2a, and added 50 .mu.l/well. After the plate was incubated for 1 hour at room temperature and washed 4 times in PBST, the remainder of the assay was conducted as described in Example 10C. Antibody concentrations of culture supernatant were determined by fitting measured optical densities to the standard curve of the isotype matched control.

Strong reactivity of monoclonal antibody ICR-1.1 was noted with two overlapping peptides spanning amino acids 13-23, as illustrated below:

SEQ ID NO: 103

VLSAGGSLFV

SEQ ID NO: 104

LSAGGSLFVN

Regions reactive with anti-ICAM-R antibodies can also be defined and/or verified using the following methodologies.

B. Epitope Mapping Using A Library of Bacterial Clones

Epitope mapping with the anti-ICAM-R antibodies was also performed using the Novatope Library Construction and Screening System (Novagen, Madison, Wis.). Using this method, a library of bacterial clones is generated wherein each clone expresses a polypeptide including a small peptide derived from the protein being examined. The library is then screened by standard colony lift methods using monoclonal antibodies as probes.

Double-stranded DNA encoding the external domain of ICAM-R (amino acids 1 to 487) from pVZ147 (See Example 4) was cut with different amounts of DNAseI in the presence of 10 mM manganese for 10 minutes at 21.degree. C. The reaction was stopped with EDTA and 1/10 of the reaction was electrophoresed on a 2% agarose gel with ethidium bromide and appropriate markers. Those reactions containing fragments in the 50-150 bp range were pooled and electrophoresed on another 2% gel. The area of the gel between 50-150 bp was excised, the fragments contained therein were electroeluted into dialysis tubing (SP Brand Spectra/Por 2, MWCO 12-14,000), and then phenol/chloroform extracted and ethanol precipitated.

One .mu.g DNA was blunted according to the manufacturer's protocol, using T4 DNA polymerase and all four dNTPs. The reaction was stopped by heating to 75.degree. C. for 10 minutes, then a single 3' dA residue was added by using Tth DNA polymerase (Novagen). The reaction was stopped by heating to 70.degree. C. for 15 minutes and extracted with chloroform. When starting with 1 .mu.g of DNA, the final concentration was 11.8 ng/.mu.l in 85 .mu.l. The dA tailed fragments are ligated into the pTOPE T-vector (Novagen) which is designed for the expression of inserts as stable fusion proteins driven by T7 RNA polymerase (the structural

gene for which is carried on a replicon in the host cell). Using 6 ng of 100 bp DNA (0.2 pmol), the ligation reaction was run at 16.degree. C. for 5 hours. NovaBlue(DE3) (Novagen) cells were transformed with 1 .mu.l (1/10) of the reaction mix, and spread on LB agar (carbenicillin/tetracycline) plates to obtain an initial count of transformants. The remainder of the ligation reaction was put at 16.degree. C. for an additional 16 hours. Based on the initial plating, 2 .mu.l of the ligation reaction was used to transform 40 .mu.l of competent NovaBlue(DE3) cells, then 8 plates were spread at a density of approximately 1250 colones/plate for screening with antibody.

Colonies were screened using standard colony lift methods onto nitrocellulose membranes, lysed in a chloroform vapor chamber and denatured. Using anti-ICAM-R monoclonal antibody ICR-1.1 at a 1:10 dilution in TBST (Tris-buffered saline/Tween) as a primary antibody, the assay was developed using an alkaline phosphatase-coupled secondary reagent. The substrate mix was incubated for 30 minutes. One isolated colony gave a strong positive reaction. Three others areas (not isolated colonies) gave weak positive reactions. Streaks were made from a stab of the isolated colony or colony areas for re-screening. Upon re-probing with ICR-1.1, the streak from the isolated colony had positive reactive areas after a 20 minute incubation with substrate. The other three colony area samples were negative. A stab from the ICAM-R reactive area was re-streaked, incubated overnight at 37.degree. C. and re-probed incubating with substrate for 10 minutes. Many ICR-1.1. reactive colonies resulted. Plasmid DNA recovered from these colonies can be sequenced and the amino acid sequence corresponding to the ICR-1.1 reactive epitope can be determined.

C. Epitope Mapping by Domain Substitution--Construction of Chimeric ICAM-R Molecules and Deletion Mutants

Conformational epitopes of ICAM-R recognized by the monoclonal antibodies of the invention may be mapped by domain substitution experiments. In these experiments, chimeric variants of ICAM-R are generated in which selected immunoglobulin-like domains of ICAM-R are fused to portions of ICAM-1 and assayed for binding to the monoclonal antibodies of the invention by FACS.

FIG. 7 is a diagram of the chimeric proteins whose construction is outlined below. Protein number 1 contains the amino-terminal immunoglobulin-like domain of ICAM-R (residues 1 to 93) fused to ICAM-1 (residue 117 to 532). Protein number 2 contains the first two amino terminal immunoglobulin-like domains of ICAM-R (residues 1 to 190) fused to ICAM-1 (residues 216 to 532). Protein number 3 contains the first three immunoglobulin-like domains of ICAM-R (residues 1 to 291) fused to ICAM-1 (residues 317 to 532).

1. Chimeric Protein 1

Protein number 1 was made by engineering a unique NheI site into the coding sequences of ICAM-R and ICAM-1 at the junction of immunoglobulin-like domains 1 and 2 of each. The DNA sequence of ICAM-R was subcloned into the M13 BM21 vector (Boehringer) as described in Example 9 resulting in a molecule called M13 BM21ICAM-R. The entire coding sequence of ICAM-1 Simmons et al., Nature, 331: 624-627 (1988)! was subcloned into the plasmid pBSSK(+) (Stratagene). The resulting plasmid, pBSSK(+)ICAM-1 was cut with SalI and KpnI to release the ICAM-1 coding sequence along with a short segment of the multiple cloning sites and ligated to M13 BM21 cut with restriction enzymes SalI and KpnI resulting in a molecule called M13 BM21ICAM-1. M13 phage isolates were verified by DNA sequence analysis.

Mutagenizing oligonucleotides ICAM1.D1.Nhe 1 (corresponding to nucleotides 426 to 393 of ICAM-1) and ICAMR.D1Nhe 1 (corresponding to nucleotides 367 to 393 of ICAM-R) having the following sequences were synthesized by routine laboratory methods:

ICAM1.D1.NheI (SEQ ID NO: 105)

AGAGGGGAGGGGTGCTAGCTCCACCCGTTCTGG

ICAMR.D1.NheI (SEQ ID NO: 106)

GAGCGTGTGGAGCTAGCACCCCTGCCT

Nucleotides 16 and 19 of ICAM1.D1.NheI and nucleotide 15 of ICAMR.D1.NheI form mismatch base pairs when the oligos are annealed to their respective complementary DNA sequences. Both oligonucleotides introduce a recognition site for endonuclease NheI. Site-directed mutagenesis with the oligonucleotides was employed to introduce the sequences of these oligos into the respective ICAM-1 and ICAM-R target DNA sequences M13 BM21ICAM-1 and M13 BM21ICAM-R. Several phage isolates from each mutagenesis reaction were sequenced to verify that the correct DNA sequence was present. These isolates were designated M13 BM21ICAM-R.NheI and M13 BM21ICAM-1.NheI.

The coding region for the ICAM-R signal peptide and immunoglobulin-like domain 1 was isolated from M13 BM21ICAM-R.NheI by the following method. Ten .mu.g of purified single strand M13 BM21ICAM-R.NheI phage DNA was annealed to 50 ng Lac Z universal -20 primer (SEQ ID NO: 28 in 1.times.Klenow DNA polymerase buffer (10 mM Tris-Cl pH 7.5, 5 mM MgCl.sub.2, 7.5 mM dithiothreitol) by incubating the mix at 65.degree. C. for 5 minutes and then 25.degree. C. for 5 minutes. The following mixture was then added to the annealing reaction: 33 .mu.M final concentration dATP, dGTP, dCTP, dTTP; 4 units of Klenow DNA polymerase (Boehringer), and 1.times.Klenow buffer. The primer extension reaction was allowed to incubate at 37.degree. C. for 45 minutes prior to being stopped by a single phenol/chloroform (1: 1) extraction and ethanol precipitation. The dried pellet was resuspended in 1.times.EcoRI buffer and 20 units each of EcoRI and NheI endonucleases were added prior to a 60 minute incubation at 37.degree. C. A 412 bp fragment containing the coding sequence for ICAM-R signal peptide and immunoglobulin-like domain 1 was agarose gel purified.

The DNA sequence of ICAM-1 containing the coding region for immunoglobulin-like domains 2 through 5, the transmembrane and cytoplasmic domains was isolated by restriction enzyme digest. Ten .mu.g of primer extended M13.BM21ICAM-1.NheI were cut with NheI and NotI. This resulted in the release of a DNA fragment of 1476 bp which was agarose gel purified.

Five .mu.g of the mammalian expression plasmid pcDNAI/Amp (Invitrogen) was digested with EcoRI and NotI and purified by spin column chromatography. A 20 .mu.l ligation mix was assembled containing the following components: 50 ng linear pCDNA1Amp with EcoRI and NotI termini, 100 ng of the 412 bp ICAM-R fragment, 100 ng of the 1476 bp ICAM-1 fragment, 1.times.ligase buffer and 1 unit of T4 DNA ligase (Boehringer). The reaction was incubated at 25.degree. C. for 16 hours and used to transform competent XL-1 cells (Biorad). Transformants were selected on LB plates supplemented with carbenicillin at a final concentration of 100 .mu.g/ml. Transformants were analyzed using a standard mini DNA prep procedure and digestion with diagnostic endonucleases. Isolates designated pCDNA1Amp.RD1.ID2-5 were chosen for expression studies.

A chimeric gene encoding protein number 1 was also generated by an alternative method as follows. An approximately 375 bp EcoRI-NheI fragment of ICAM-R containing domain 1 and an approximately 1500 bp NheI-NotI fragment of ICAM-1 containing the extracellular domains 2-5, the transmembrane domain and the cytoplasmic tail were gel purified after restriction enzyme digestion of the double stranded RF (replicative form) DNA from the M13BM21ICAM-R and M13 BM21ICAM-1 clones and agarose gel electrophoresis of the corresponding double stranded plasmid DNAs. The resulting two DNA fragments were cloned by a three way ligation into an EcoRI and NotI digested and calf intestinal phosphatase-treated expression vector pcDNAI/Amp (Invitrogen). E. coli XL1 blue (Stratagene) strain was transformed with the ligation mixture and the transformants were selected on carbenicillin containing plate. Clones with the desired inserts were identified by restriction enzyme digestion of the plasmid DNA minipreps.

2. Chimeric Proteins 2 and 3

To construct coding sequences for proteins 2 and 3, engineered versions of M13 BM21ICAM-1 and M13 BM21ICAM-R in which a unique NheI site was created between immunoglobulin-like domains 2 and 3 or a unique AflII site was created between immunoglobulin-like domains 3 and 4 were generated by methods similar to those described in the foregoing paragraphs. Four oligonucleotides (ICAM-1.D2.NheI corresponding to nucleotides 686 to 713 of ICAM-1, ICAM-R.D2.NheI corresponding to nucleotides 655 to 690 of ICAM-R, ICAM-1.D3.AflII corresponding to nucleotides 987 to 1026 of ICAM-1, and ICAM-R.D3.AflII corresponding to nucleotides 962 to 993 of ICAM-R) with the sequences set out below were synthesized for this purpose.

ICAM-1.D2.NheI (SEQ ID NO: 107)

GGGGGAGTCGCTAGCAGGACAAAGGTC

ICAM-R.D2.NheI (SEQ ID NO: 108)

CGAACCTTTGTCCTGCTAGCGACCCCCCCGCGCCTC

ICAM-1.D3.AflII (SEQ ID NO: 109)

TGAGACCTCTGGCTTCCTTAAGATCACGTTGGGCGCCGG

ICAM-R.D3.AflII (SEQ ID NO: 110)

GACCCATTGTGAACTTAAGCGAGCCCACC

Nucleotide 13 of ICAM1.D2NheI; nucleotides 17, 18 and 20 of ICAMR.D2.NheI; nucleotides 17, 18, 20 and 22 of ICAM-1.D3. AflII; and nucleotides 15 and 17 of ICAM-R.D3.AflII form mismatch base pairs when the oligonucleotides are annealed to their respective complementary DNA sequences. The appropriate coding sequences of ICAM-R and ICAM-1 (sequences encoding the first two amino terminal immunoglobulin-like domains of ICAM-R fused to sequences encoding ICAM-1 residues 118 to 532 for protein 2 and sequences encoding the first three immunoglobulin-like domains of ICAM-R fused to sequences encoding ICAM-1 residues 317 to 532 for protein 3) were then subcloned into expression plasmid pCDNA1Amp (Invitrogen) to generate isolates pCDNA1Amp.RD1-2.1D3-5 and pCDNAAmp.RD1-3.1D4-5 respectively encoding ICAM-R variant proteins 2 and 3.

Gene fusions encoding protein numbers 2 and 3 were also constructed by alternative methods as follows.

For the generation of protein 2 encoding sequences, an NheI was introduced by oligonucleotide directed in vitro mutagenesis in between domains 2 and 3 in both ICAM-R and ICAM-1. An approximately 700 bp EcoRI-NheI fragment of ICAM-R containing the domains 1 and 2, and an approximately 1100 bp NheI-NotI fragment of ICAM-I containing the domains 3-5, the transmembrane domain and the cytoplasmic tail were subcloned by a three-way ligation into the EcoRI and NotI digested and calf intestinal phosphatase-treated pcDNAI/Amp plasmid DNA. For the generation of protein 3 encoding sequences an approximately 1000 bp NotI-Aflll fragment of ICAM-R containing domains 1 through 3, and an approximately 850 bp Aflll-NotI fragment of ICAM-1 containing domains 4-5, the transmembrane domain and the cytoplasmic tail were purified by restriction enzyme digestion of the plasmid DNAs and agarose gel electrophoresis. These two fragments were cloned by a three way ligation into the NotI digested and phosphatase treated pcDNAI/Amp plasmid DNA. Clones containing the insert with the desired orientation were identified by restriction enzyme digestion of plasmid DNA mini preparations.

3. Domain Deletion Proteins

ICAM-R domain deletion variants were generated by similar oligonucleotide directed mutagenesis protocols as described above for chimeric protein numbers 1, 2 and 3. A domain 1 deletion variant which lacks amino acids 2-90 of ICAM-R (SEQ ID NO: 1), a domain 1 and 2 deletion variant which lacks amino acids 2-203, and a domain 3 deletion variant lacking amino acids 188-285 were constructed.

Control plasmids containing the full length ICAM-R or ICAM-1 cDNA sequences were generated by ligating gel-purified cDNA fragments to plasmid pCDNA1Amp. The two plasmids pCDNA1AmpICAM-1 and pCDNA1AmpICAM-R express the full length ICAM-1 and ICAM-R proteins, respectively, so that monoclonal antibody binding to native protein in equivalent cellular contexts can be assessed.

COS cells were transfected with the plasmid DNA encoding the ICAM-R chimeric or deletion mutant proteins or with the plasmid DNA pCDNA1AmpICAM-1, pCDNA1AmpICAM-R or pCDNA1Amp by the DEAE-dextran method. Typically, the COS cells were seeded at a density of about 7.0 .times.10.sup.5 cells on a 10 cm diameter plate and grown overnight in Dulbecco's modified Eagles medium (DMEM) containing 10% fetal bovine serum (FBS). The next day the cell monolayer was rinsed with DMEM and exposed to 10 ml of transfection mixture containing 10 ug of the desired plasmid DNA, 0.1M chloroquine and 5.0 mg DEAE-dextran in DMEM for 2.5 hours at 37.degree. C. After the incubation, the transfection mixture was aspirated and the monolayer was treated with 10% DMSO in PBS for one minute. The cells were washed once with DMEM and incubated with DMEM containing 10% FBS. The next day the medium was replaced with fresh medium and the incubation was continued for two more days.

Expression of all the chimeric and deletion ICAM-R proteins was obtained. The domain 1 and domain 3 deletion variants expressed at a level of 50-60% compared to the wild type ICAM-R protein.

D. Epitope Mapping by Domain Substitution--Monoclonal Antibody Binding Assay

For the anti-ICAM-R monoclonal antibody binding assay, COS cells transfected with constructs encoding the ICAM-R chimeric proteins or control constructs were removed from the plates by EDTA treatment and aliquoted at 2.5.times.10.sup.5 cells per well in a 96-well round bottom plate. Cells were washed 3 times with ice cold washing buffer (PBS containing 1% BSA and 0.05% sodium azide). Anti-ICAM-R monoclonal antibody was applied at 5.0 ug/ml in 50 ul final volume and incubated on ice for 30 minutes. Cells were then washed three times with cold washing buffer and incubated with the. FITC labeled secondary antibody (sheep anti-mouse IgG F(ab').sub.2) at a 1:100 dilution on ice for 30 minutes in dark in 50 ul final volume. After the incubation, cells were washed again for three times in the ice cold washing buffer and resuspended in 200 ul of 1% paraformaldehyde. The samples were analyzed on a Becton-Dickinson FACScan instrument. Results of the assay are given below in Table 5 as percent positive COS cell transfectants, wherein MOPC 21 (IgG1) and UPC 10 (IgG2a) are isotype matched controls, 18E3D is an ICAM-1 specific monoclonal antibody and ICR-1.1 to ICR-9.2 are ICAM-R specific monoclonal antibodies. The reactivities of monoclonal antibodies ICR-1.1 through ICR-9.2 were assayed in a different experiment than monoclonal antibodies ICR-12.1 through ICR-17.1.

The results presented above show that the antibodies ICR-1.1, 2.1, 3.1, 5.1, 7.1, 8.1, 12.1, 13.1, 14.1, 15.1, 16.1 and 17.1 recognize the hybrid molecule in which only the ICAM-1 domain 1 has been replaced with the ICAM-R domain 1. The antibodies ICR-4.2, 6.2 and 9.2 recognize the molecule in which a minimum of 2 domains (domain 1 and 2) of ICAM-1 was replaced with the corresponding domains of ICAM-R. Based on these results the antibodies have been categorized as either domain 1 or domain 2 specific.

The ICAM-R chimeric and deletion mutant protein constructs can also be used to transfect rat L cells by a calcium phosphate co-precipitate protocol using 10 .mu.g of 2.times.CsCl-banded plasmid DNA. In this protocol, forty-eight hours post-transfection the cells are released from the dishes by mild trypsinization. The cells are divided and incubated on ice with anti-ICAM-R monoclonal antibodies or a control isotype matched monoclonal antibody at a concentration of 10 .mu.g/ml or no monoclonal antibody for 1 hour. The cells are then processed for FACS analysis as previously described in Example 12C.

E. Epitope Mapping by Amino Acid Substitution

Differential reactivity of an anti-ICAM-R antibody of the invention with the ICAM-R variant proteins as described above thus is indicative of reactivity with a specific domain of ICAM-R. Once particular domains are identified that reacted with specific anti-ICAM-R monoclonal antibodies, individual residues within those domains are changed by oligo-directed site specific mutagenesis to determine their relative effects on monoclonal antibody binding. Based on computer algorithms that predict protein hydropathy and secondary structure (Kyte et al., supra), particular residues that have the potential for antibody interactions are targeted for mutagenesis.

Mutagenesis of ICAM-R was carried out according to the procedure of Kunkel et al., supra. E. coli strain Cj236 (dut ung) was transformed with the plasmid pcDNA1/AmpICAM-R (see Section C above) by electroporation. The transformants were selected on carbenicillin containing plate. One of the transformants was infected with the helper phage M13K07 and grown overnight. Uracil-containing single stranded DNA was prepared from the culture supernatant and used for mutagenesis. Mutagenic oligonucleotides were hybridized to the uracil containing single stranded DNA of pcDNA1/Amp-ICAM-R. Using the mutagenic oligonucleotides as primers, DNA synthesis and ligation reactions were carried out using T7 DNA polymerase and T4 DNA ligase, respectively. An aliquote of the synthesis reaction was used to transform E. coli XL1 blue (Stratagene) strain and transformants were selected on carbenicillin containing plates. Growth of the uracil containing plasmid DNA in this strain markedly reduces the propagation of the uracil containing DNA (wild type) strand. Mutants were selected by plasmid DNA minipreps and diagnostic restriction enzyme digestion. Sequences were further verified by DNA sequence analysis. The mutations made were: F21V/AS, E32K/AS, K33I/AL, E37T/AS, T38/A, L40/A, K42E/AS, E43/A, L44V/AL, W51A/AS, R64/Q, S68/A, Y70/A, N72/Q, Q75I/AS, N81/Q. Mutation "F21V/AS" indicates, for example, that the phenylalanine at position 21 of ICAM-R (SEQ ID NO: 1) and the valine at position 22 were respectively changed to an alanine and a serine, while mutation "T38/A" indicates that the threonine at position 38 of ICAM-R (SEQ ID NO: 1) was changed to an alanine. Effects of each mutation on anti-ICAM-R monoclonal antibody binding were tested according to the procedure described in Section C above. Table 6 below summarizes the results obtained, wherein a mutation with a "critical" effect was defined as 0-20% binding of an antibody in comparison to binding to wild type ICAM-R, an "important" effect was defined as about 50% binding in comparison binding to wild type ICAM-R, and a minor effect was defined as about 75% binding in comparison to binding to wild type ICAM-R. Mutations that did not effect binding of an antibody are not listed in Table 6.

The mutations T38/A, S68/A and R64/Q had no effect on binding to any of the nine antibodies tested, while mutation E43/A resulted in 50% increase in binding of the above nine monoclonal antibodies in comparison to binding to wild type ICAM-R. Mutation F21V/AS abolished binding of all antibodies and appears to grossly affect the conformation of ICAM-R.

Example 15

The distribution of ICAM-R protein and the expression of ICAM-R RNA in various cells and cell lines were respectively assayed by FACS analysis and Northern blot hybridization.

A. FACS Analyses of ICAM-R Protein Distribution In Leukocytic Cell Lines and Normal Leukocytes

FACS analyses carried out as described in Example 12C on leukocyte cell lines using anti-ICAM-R monoclonal antibody ICR-2.1, anti-ICAM-1 antibody (LB2) and anti-CD18 antibody (TS1/18, ATCC HB203) illustrated that ICAM-R is expressed on a wide variety of in vitro propagated cells lines representative of the major leukocyte lineages: T lymphocytes, B lymphocytes, and myeloid cells. Surface expression of ICAM-R was not detected on the primitive erythroleukemic line, K562. Further, ICAM-R was not expressed detectably by cultured human umbilical vein endothelial cells (HUVECS) either before or after stimulation with tumor necrosis factor which did upregulate expression of ICAM-1. This pattern of expression is also distinct from that observed for ICAM-2 which is expressed on endothelium. Table 7 below provides the mean fluorescence of each cell sample and the percent positive cells relative to a control in each cell sample (e.g., mean fluorescence of 13/11% positive cells).

B. FACS Analyses of ICAM-R Distribution On Human and Macague Leukocytes

FACS analyses performed as described in Example 10C on normal human and macaque peripheral blood leukocytes showed that the anti-ICAM-R monoclonal antibody ICR-2.1 reacted with the three major human leukocyte lineages: lymphoid, monocytoid and granulocytoid. See the final six entries of Table 7. In addition, monoclonal antibodies ICR-4.2 and ICR-9.2 cross-reacted with macaque leukocytes (Table 2 and Example 11E) indicating that these monoclonal antibodies may be useful in monitoring the expression of ICAM-R in disease models executed in this animal.

Human bronchiolar aveolar lavage cells (primarily macrophages) were stained with five anti-ICAM-R antibodies (ICR-2.1, ICR-3.1, ICR-4.2, ICR-6.2 and ICR-7.1) as described in Example 11E and analyzed by FACS. None of the antibodies specifically stained these cells. Other data obtained via immunohistological tests suggests that ICAM-R can be expressed on macrophages on interstitial spaces in the lung. (See Example 18, where the expression of ICAM-R in lung tissue is described.)

C. Northern Blot Analyses of ICAM-R RNA Expression in Leukocytic Cell Lines and HUVECS

RNA was extracted from human leukocyte cell lines and from HUVECS as described in Example 6, and was analyzed by Northern blot hybridization (also as described in Example 6) by probing with either ICAM-R or ICAM-1 cDNA. After phosphorimaging of the initial hybridization, blots were stripped and reanalyzed using a human actin probe. The results of the actin normalized Northerns of ICAM-R and ICAM-1 probed blots are presented in FIG. 7(A through B) as bar graphs. At the RNA level, ICAM-R was expressed in a variety of leukocytic cell types. ICAM-R RNA expression was not necessarily concomitant with the expression of ICAM-1 RNA. For example, unstimulated HUVECS express low levels of ICAM-1 and expression is upregulated following TNF stimulation (FIG. 7B). In contrast, detectable levels of ICAM-R message were not observed in unstimulated or stimulated HUVECS (FIG. 7A).

Example 16

The expression of ICAM-R transcript in endothelial cells was examined. Poly A+ mRNA was obtained from human skin angiomas and analyzed by Northern blot using ICAM-R .sup.32 P-labeled riboprobes to determine whether splice variants of human ICAM-R were present in endothelial cells.

The angiomas analyzed were benign human vascular tumors. These surgical samples were frozen and first examined by immunohisto-chemistry, using the anti-ICAM-R monoclonal antibodies ICR-3.1 and ICR-4.2. Angiomas were separated into two groups based on the level of expression of ICAM-R, one group expressing high levels of ICAM-3, another group expressing low or no detectable levels of ICAM-R. These benign tumors did not show signs of inflammation and ICAM-R expression was almost completely restricted to endothelial cells.

For Northern blot analysis, other human tissues (cerebellum, liver, lung carcinoma, abnormal small intestine and spleen) were used as controls. All tissue samples were frozen and stored at -80.degree. C. PolyA mRNA was extracted from angiomas and control tissue blocks using RNA STAT60 mRNA isolation reagents (Tel-test "B", Inc., Friendswood, Tex.). The tissue was ground and homogenized, then total mRNA was extracted from the homogenate with chloroform and precipitated. Poly A+ mRNA were purified from total mRNA by chromatography on oligo dT cellulose columns. Five ug of each polyA+ mRNA were loaded per lane on a 1% formaldehyde agarose gel, then transferred to Hybond-C membranes (as described in Example 7).

To generate human ICAM-R riboprobe, a subclone of pVZ-147 plasmid (Example 4) encoding ICAM-R domain 1 was utilized. The plasmid was digested with Asp718 and the complementary strand synthesized via T3 primer and RNA polymerase using .sup.32 P UTP. The resultant RNA transcript was used as a probe.

The nylon membranes were pre-hybridized and hybridized in 50% formamide, 5.times.SSC, 1.times.PE (50 mM Tris-HCL pH 7.5, 0.1% sodium pyrophosphate, 0.2% polyvinylpyrolidone, 0.2% ficoll, 5 mM EDTA, 1% SDS) and 150 ug/ml denatured salmon sperm. RNA probes were denatured at 100.degree. C. for 5 minutes then added to each membrane at a concentration of 1.times.10.sup.6 dpm/ml hybridization mix. Membranes were hybridized overnight at 65.degree. C., then washed at 65.degree. C. twice in 2.times.SSC, 0.1% SDS and twice in 0.1.times.SSC, 0.1% SDS for 15 minutes each. They were exposed on film for three hours to three days.

Analysis of the Northern blot demonstrated that the size of the transcripts present in both the high and low expression groups of angiomas was identical. Two transcripts were detectable, one migrating at about 2.2 kb identical to the one present in hematopoietic cells. Another transcript migrating at 3 kb was also present in both angiomas and control tissues that could correspond either to a variant of ICAM-R or to cross-hybridization with another molecule. The level of hybridization of both transcripts were high in the group of angiomas expressing high levels of ICAM-R protein, while it was low in the low expressing group. The regulation of ICAM-R expression on endothelial and hematopoietic cells is apparently distinct; expression is constitutive on hematopoietic cells, while it is induced on endothelial cells at neovascularizing sites. However, based on immunohistochemistry with anti-ICAM-R antibodies and on this Northern analysis, it seems likely that a significant fraction of the ICAM-R molecules expressed on endothelial cells are likely to be identical in primary structure to that of the molecule expressed on cells of hematopoietic origin.

Example 17

Immunoprecipitations of detergent solubilized lysates of surface biotinylated human cell lines KG1a, K562 and CEM were performed using the four anti-ICAM-R monoclonal antibodies: ICR-2.1, ICR-1.1, ICR-4.2, and ICR-3.1.

Cell surface proteins on human leukocyte cell lines KG1, K562, and CEM were labelled by reaction with sulfo-NHS-biotin (Pierce Chemical Company, Rockford, Ill.) as follows. For each reaction 0.5-1.times.10.sup.7 cells were washed twice in phosphate buffered saline (PBS), resuspended in 1 ml PBS and 10 .mu.l of 100 mM sulfo-NHS-biotin diluted in PBS was added. Following incubation for 10 minutes at 37.degree. C. the cells were washed once with PBS, and 4 ml of 10 mM Tris pH 8.4, 0.25M sucrose was added and the cells were then incubated for 30 minutes at 4.degree. C. with gentle mixing. The cells were pelleted by centrifugation, the supernatant was aspirated and the pellet was solubilized with 300 .mu.l of 10 mM Tris pH 8, 50 mM NaCl, 1% Triton X-100, 1 mM phenylmethylsulfonyl fluoride, 1 mM EDTA by incubating on ice for 15 minutes. The lysate was clarified by centrifugation and the supernatant was precleared by addition of 25 .mu.l normal mouse serum and incubation for 1 hour at 4.degree. C. This step was followed by the addition of 20 .mu.l of a 50/50 (v/v) solution of protein-A sepharose beads (Sigma) that had been preincubated with 20 .mu.g of affinity purified rabbit anti-mouse Immunoglobulin (Zymed). After incubation for 30 minutes at 4.degree. C., the sepharose beads were removed by centrifugation.

Specific immunoprecipitations were then performed by addition of 20 .mu.l of sepharose beads that had been prearmed by sequential incubation with rabbit anti-mouse immunoglobulin and either anti-ICAM-R or control IgG.sub.1 or IgG.sub.2a monoclonal antibodies. Following overnight incubation at 4.degree. C. with agitation, sepharose beads were pelleted in a microcentrifuge and washed sequentially 2 times with 1 ml 10 mM Hepes pH 7.3, 150 mM NaCl, 1% Triton X-100; 1.times.with 0.1M Tris pH 8, 0.5M LiCl, 1% beta mercaptoethanol; and 1.times.with 20 mM Tris pH 7.5, 50 mM NaCl, 0.5% NP-40. Beads were then eluted with 50 .mu.l 150 mM Tris pH 6.8, bromphenol blue, 20% beta mercaptoethanol, 4% SDS and 20% glycerol; boiled for 5 minutes; and pelleted by centrifugation. Thirty-five .mu.l of the resulting eluate was then analyzed by SDS-PAGE (10% acrylamide). After electrophoresis, proteins were electroblotted onto Immobilon-P membranes (Millipore, Bedford, Mass.) and incubated in 2% bovine serum albumin diluted in Tris-buffered saline containing 0.2% Tween-20 for 20 minutes at 4.degree. C. Blots were then incubated with horseradish peroxidase coupled to streptavidin (Vector) in TBS-Tween at room temperature for 20 minutes. Following 3 rinses in TBS-Tween, ECL western blotting detection reagents (Amersham) were added and chemiluminescent bands were visualized on Kodak X-OMAT-AR film.

FIG. 8(A through B) shows the resulting Western blots. A single specifically precipitated species of 120 kD was observed in immunoprecipitates with monoclonal antibody ICR-2.1 from KG1 cells, but not from K562 cells (See FIG. 8A).

A 120 kD band was also resolved in immunoprecipitates of the T cell line CEM (FIG. 8B, wherein Lane A was reacted with monoclonal antibody ICR-2.1; Lane B, monoclonal antibody ICR-4.2; Lane C, monoclonal antibody ICR-3.1; Lane D, monoclonal antibody ICR-1.1; and Lane E, a negative control antibody). The size of the ICAM-R species resolved in other immunoprecipitations varied slightly depending on the cellular source. Species ranging from about 116 kD on some lymphoid cells to about 140 kD on some myeloid cells were observed. Given the predicted size (about 52 kD) of the core peptide based on the nucleotide sequence of the ICAM-R gene, these results imply that ICAM-R is heavily modified post-translationally to yield the mature cell surface form of the protein.

Example 18

Immunohistologic staining with anti-ICAM-R monoclonal antibodies ICR-4.2, ICR-1.1, and ICR-2.1 and control antibodies was carried out on various human tissues including tonsil, spleen, liver, lung, kidney, heart, digestive tract, skin, synovium, and brain (both normal and multiple sclerosis-afflicted brain tissue). Similar staining patterns were obtained using the different anti-ICAM-R antibodies as well as when using purified anti-ICAM-R monoclonal antibody ICR-1.1 or hybridoma supernatant.

Sections (6 .mu.m) of various tissues were layered onto Vectabond (Vector) coated slides and stored at -70.degree. C. (some sections were stored at -20.degree. C.). Prior to use, slides were removed from -70.degree. C. and placed at 55.degree. C. for 5 minutes. Sections were then fixed in cold acetone for 10 minutes and air dried. Sections were blocked in a solution containing 1% BSA, 60% normal human sera, and 6% normal horse sera for 30 minutes at room temperature. Primary antibody directed against ICAM-R, a negative control antibody, anti-ICAM-1 monoclonal antibody or anti-ICAM-2 monoclonal antibody was applied to each section for 1 hour at room temperature. Unbound antibody was washed off by immersing the slides 3 times in 1.times.PBST for 5 minutes each time. Biotinylated anti-mouse immunoglobulin (Vector) was then applied to each section in the same fashion. ABC-HPO (Avidin-Biotin Complex-HPO) was used to detect the second antibody. A solution of reagent A (9 .mu.l) (Vector) combined with reagent B (9 .mu.l) (Vector) in 1 ml of 1% BSA/PBST was applied to each section for 30 minutes at room temperature. Slides were then washed 3 times in 1.times.PBST. DAB substrate (3'3 diaminobenzidine-tetrahydrochloride, Sigma) (stock: 600 mg/ml DAB diluted 1:10 in 0.05M Tris Buffer, pH 7.6, with 3% H.sub.2 O.sub.2 added to a final concentration of 1%) was applied to each slide for 8 minutes at room temperature. Slides were washed in water for 5-10 minutes at room temperature and then 1% osmic acid was added (to enhance color development) for one minute at room temperature. Slides were then washed in tap water for 5-10 minutes and counterstained in 1% Nuclear Fast Red (NFR) for 30 seconds at room temperature. Lastly, slides were alcohol dehydrated, treated with Histroclear and mounted with coverslips using histomount.

A selection of results of staining with the monoclonal antibodies is presented in FIG. 9(A through G) as photomicrographs wherein the tissue in 9A, 9B and 9E is human tonsil; in 9C and 9D is human liver; in 9F is brain from a human patient afflicted with multiple sclerosis; and in 9G is normal human brain. Sections shown in 9A, 9C, 9F and 9G were stained with anti-ICAM-R monoclonal antibody ICR-4.2. Sections shown in 9B and 9D were stained with the negative control antibody, while the section shown in 9E was stained with the anti-ICAM-1 antibody. Staining revealed high level expression of ICAM-R in lymphoid tissues such as tonsil (9A). Expression was also detected on tissue leukocytes in other nonlymphoid organs such as the liver wherein Kupfer cells (liver macrophages) were positively stained (9C). Evidence that ICAM-1 and ICAM-R expression are regulated distinctly in vivo is given by the staining pattern observed in tonsil and lymph node: ICAM-1 is strongly expressed on B cells in the germinal centers of secondary follicles and not expressed in primary follicles, whereas ICAM-R is expressed strongly in the primary follicles and weakly in the germinal centers (10A and 10E). Significantly, ICAM-R expression was also detected on leukocytes infiltrating sites of inflammation. For example, ICAM-R expression was observed on perivascular infiltrating leukocytes in the brain tissue of individuals afflicted with multiple sclerosis (9F). Similar staining was not observed in anatomically equivalent locations of brain tissue from normal individuals (9G). ICAM-R expression was also detected on leukocytes infiltrating synovia of arthritic joints. Also, whereas expression of ICAM-1 and ICAM-2 was detected on endothelia lining vessels, ICAM-R was not typically observed on vascular endothelium. Expression of ICAM-R was detected on cells in the aveoli of the lung.

More generally, cells expressing ICAM-R were detected in all normal and pathological tissues. These ICAM-R expressing cells could be identified morphologically and by comparison of serial immunological staining as leucocytes and antigen-presenting cells. All CD3.sup.+ T cells present in various tissues expressed high levels of ICAM-R. In contrast, only a subset of B cells (IgD+) present in primary follicles and in the mantle zone of germinal centers expressed high levels of ICAM-R. Amongst antigen-presenting cells, Langerhans cells in the epithelium expressed high levels of ICAM-R while only a subset of other tissue macrophages expressed ICAM-R.

ICAM-R monoclonal antibodies ICR-1.1 and ICR-4.2 were also used in procedures similar to those described above to stain biopsy tissue sections of both human mammary carcinoma (ductal and lobular) and melanomas. In both tumor types some sections exhibited specific patchy staining of the endothelia in a range of blood vessels (venular, arterioles and capillaries). Corresponding normal tissue showed no expression of ICAM-R on endothelium.

Thus, while ICAM-R is typically not expressed on endothelium of the general vasculature, it is apparently expressed on a subset of vessels associated with two types of solid tumors. Given this distribution, reagents (e.g., monoclonal antibodies) directed against ICAM-R may provide therapeutic vehicles which selectively target tumor versus normal vasculature.

In summary, the contrasts in the patterns of expression of ICAM-R versus ICAM-1 and ICAM-2 are significant. Constitutive expression of ICAM-2 was observed on both leukocytes and endothelium. Basal expression of ICAM-1 on leukocytes, endothelia and epithelia was low or absent but was induced in pathologic tissues or in vitro. ICAM-R was expressed at high levels on most leukocytes and, notwithstanding rare expression on tumor associated endothelia, was generally not expressed on vascular endothelia.

Example 19

In order to determine whether ICAM-R is involved in homotypic cell adhesion, aggregation assays were performed with a panel of cell lines which express ICAM-R including T lymphoblastoid cell lines (SupT1, CEM, Molt 4, Hut 78, Jurkat, SKW3), B lymphoblastoid cells lines (Jijoye, Raji), monocytic cell lines (U937, HL60), a myelogenous cell line (KG-1) and the erythroleukemia cell line K562. To determine the function of the ICAM-R molecule, the cells were incubated with various antibodies before aggregation was assayed. Anti-ICAM-R supernatants produced by hybridomas ICR-2.1, ICR-1.1, ICR-4.2, and ICR-3.1 were used as well as antibody preparations known to block aggregation through a .beta.2 integrin pathway: TS1/18 (ATCC HB203) specific for the CD18 molecule, the .beta.-subunit of LFA-1; TS1/22 (ATCC HB202) specific for the CD 11a molecule, the .alpha.-chain of LFA-1; and LM2/1 (ATCC HB204) specific for the CD11b molecule, the .alpha.-subunit of MAC-1. Purified anti-ICAM-1 antibody and hybridoma supernatant directed against the .alpha.-chain of the VLA-4 molecule (hybridoma clone 163H, Michael Longenecker, Alberta, Canada) were used as controls.

Aggregation assays were done in duplicate, with and without addition of PMA (50 ng/ml). 3.times.10.sup.5 cells in RPMI 1640 medium with 10% fetal calf serum were added in a flat-bottomed 96-well microtest plate. When one antibody was tested in an experiment, 50 .mu.l of purified antibody or hybridoma supernatant were added to the wells (PMA was added at the same time to selected wells). When two antibodies were tested in the same experiment, the antibodies were added sequentially to the cells at room temperature and incubated for 30 minutes each (incubation for 15 minutes at 37.degree. C. produced the same results), and then the cells were incubated at 37.degree. C. Incubating the antibodies with the cells before addition of PMA or at the same time as the PMA did not cause any significant change in the aggregation results. After incubation with the antibody or antibodies, cells were uniformly resuspended and then incubated at 37.degree. C. for 4 to 24 hours. Aggregation scoring was done with an inverted microscope. In each experiment, the efficacy of the PMA stimulation was checked in parallel by stimulating Raji cells with an equal amount of PMA and determining the amount of aggregation blockable by monoclonal antibodies to CD18, CD11a, and ICAM-1 molecules.

Table 8, below, sets out the results of one representative aggregation experiment wherein PMA was added. Aggregation scores are reported on a range from 0 to 5, wherein 0 indicates that no cells were in clusters; 1 indicates that less than 10% of the cells were in clusters; 2 indicates that 10 to 50% cells were aggregated; 3 indicates that 50 to 100% cells were in loose clusters; and 4 indicates that almost 100% of the cells were in compact aggregates.

Interestingly, treatment with three of the antibodies specific for ICAM-R (ICR-2.1, ICR-1.1, and ICR-3.1) stimulated homotypic cell-cell aggregation (data for ICR-1.1 and ICR-3.1 not shown). Stimulation occurred in both the presence and absence of co-stimulatory agents such as a phorbol ester (PMA). The fourth anti-ICAM-R monoclonal antibody (ICR-4.2) did not stimulate cell aggregation but blocked the aggregation stimulated by the other anti-ICAM-R antibodies. At least a portion of the aggregation stimulated by anti-ICAM-R antibodies in PMA treated cells was blocked by pretreatment with monoclonal antibodies against CD18 or CD11a indicating that one or more leukointegrins may participate in this type of adhesion.

To confirm that aggregation was induced by the anti-ICAM-R antibodies ICR-2.1, ICR-1.1, and ICR-3.1 the aggregation assays were performed using both the whole immunoglobulin (ICR-1.1-Ig) and Fab' fragments (ICR-1.1-Fab') purified from the same anti-ICAM-R monoclonal antibody (ICR-1.1). The assays were performed with SKW3 T cells as described above using ICR-1.1-Ig and ICR-1.1-Fab' at a concentration of 1 .mu.gm/l. Supernatants of anti-CD18 and anti-ICAM-R (CIR-1.1-sup and ICR-4.2-sup) hybridomas were used as controls. After four hours, the same increase in cell aggregation was found for whole immunoglobulin as for the Fab' fragments or the ICR-1.1 supernatant (See Table 9 below).

No increase in aggregation was found with anti-CD18 supernatant or anti-ICAM-R ICR-4.2 supernatant. These results rule out the trivial explanation that enhanced aggregation was due to antibody mediated cross-linking of the cells. The engagement of ICAM-R protein, in this case by selected antibodies, may transduce a signal which alters the adhesive potential of the bound cells.

Example 20

The process of activation and proliferation of cells of the immune system is marked by a continuum of cellular events. The upregulation of certain cell surface molecules (e.g., CD69 and the transferrin receptor) is an early marker of cell activation. Similarly, cell agglutination occurs early in the process of activation. The upregulation of the IL-2 receptor occurs at an intermediate to late stage and cell proliferation is a late event. Three types of experiments were performed to determine the extent to which ICAM-R is involved in immune cell activation/proliferation. In the first type, the capacity of ICAM-R presented on the surface of a transfected cell to stimulate proliferation of lymphocytes was examined. In the second type, antibodies of the invention recognizing distinct epitopes on ICAM-R were used as probes to engage the external domain of ICAM-R to determine the effects of antibody binding either alone or in combination with other stimuli on lymphocyte or monocyte activation and proliferation. In the third type of experiment, the effects of recombinant ICAM-R protein on T cell proliferation were determined. In the fourth type, variant ICAM-R proteins were expressed in lymphoblastoid cells and effects of the mutations on T cell receptor-dependent stimulation were measured. In the fifth type of experiment, the downstream intracellular biochemical consequences (e.g., effects on PKC translocation) of ICAM-R engagement were examined.

A. Stimulation of PBMC Proliferation by ICAM-R Transfectants

Mouse L cells transfected with either ICAM-R cDNA or ICAM-1 cDNA (Example 7) were assayed for their ability to stimulate human peripheral blood mononuclear cell (PBMC) proliferation as measured by .sup.3 H-thymidine incorporation assays which indicate changes in the rate of DNA replication. Nontransfected mouse L cells or transfected L cells were obtained by trypsinization from tissue culture flasks and washed in RPMI-1640 containing 10% fetal bovine serum. Five.times.10.sup.4 L cells in 120 .mu.l tissue culture media (RPMI-1640 with 10% fetal bovine serum) were added to individual wells of a sterile 96-well flat bottom tissue culture plate and the plates were incubated for 24-36 hours at 37.degree. C. in a 5% CO.sub.2 incubator. The media was then removed in a sterile manner and 2.times.10.sup.5 freshly isolated PBMC in a total volume of 200 .mu.l tissue culture media were added to individual wells containing either transfected or non-transfected mouse L cells. PBMC were also added to control wells containing no L cells. The PBMC were previously isolated from healthy donors by centrifugation on Histopaque gradients (Sigma). Fresh peripheral blood was mixed with an equal volume of PBS, layered onto Histopaque and centrifuged at 450 g for 20 minutes with no brake applied. PBMC-containing fractions were collected, washed in PBS and adjusted to 1.times.10.sup.6 viable cells/ml prior to addition into wells. The tissue culture plates were then incubated for a total of 4 days either in the presence or absence of PMA at a final concentration of 5 ng/ml. Lymphocyte proliferation was then assessed after the addition of 1 uCi .sup.3 H-thymidine (NEN, Boston, Mass.) to individual wells for the last 18-24 hours of culture. All cultures were then terminated by harvesting the contents of each well onto glass fiber filter strips using a PHD model plate harvester (Costar, Cambridge, Mass.). Individual filter mats were then placed in 3 ml Ecolume scintillation cocktail (ICN Biomedicals, Costa Mesa, Calif.) and counted using a beta-scintillation counter. LTK cells expressing ICAM-R stimulated proliferation of PBMC (as indicated by increased DNA replication) in comparison to nontransfected control LTK cells or in the absence of any stimulus. LTK cells expressing ICAM-1 induced the proliferation of PBMC to approximately an equal extent. By binding to its receptor(s) on PBMC, ICAM-R transmits an intercellular signal to the PBMC which in this cellular context results in cell proliferation.

B. PMBC Activation by ICAM-R Specific Monoclonal Antibodies

Anti-ICAM-R antibodies of the invention were also tested to determine their effect on immune cell activation and proliferation.

Anti-ICAM-R monoclonal antibodies were preliminarily tested for the ability to affect early events in cell activation including upregulation of the cell surface molecules CD69, the transferrin receptor and the IL-2 receptor on the target cells as measured by flow cytometry analysis. Unstimulated lymphocytes express low levels of the transferrin and IL-2 receptors. Expression of the receptors increases dramatically when lymphocytes are activated.

Anti-ICAM-R monoclonal antibodies ICR-1.1 and ICR-4.2 were each tested for the ability to induce PMBC activation in the absence of other inducing stimuli. Monoclonal antibodies ICR-1.1 or ICR-4.2 (or control monoclonal antibodies) were added (10 .mu.g/well in PBS) to individual wells of a 96-well flat bottom tissue culture plate and incubated for 3 hours at 37.degree. C. in a 5% CO.sub.2 incubator. The plates were washed 3 times with sterile PBS to remove unbound antibody and freshly isolated PBMC were immediately added to a final concentration of 2.times.10.sup.5 cells/well in a volume of 200 .mu.l media. The plates were then incubated for either 1 or 3 days at which time the cells cultured in the presence of different antibodies were removed, washed as described above in PBS containing 0.01% sodium azide and 1% BSA (FACS buffer) and stained with either FITC (Becton Dickinson)--conjugated negative control antibodies or a panel of FITC-conjugated anti-CD69, anti-transferrin receptor and anti-IL-2 receptor antibodies. Results were obtained by FACScan analysis. Expression of CD69 and the transferrin receptor but not the IL-2 receptor increased after 1 day when PBMC were cultured on immobilized (i.e., cross-linked) antibody ICR-1.1 but not when cultured on immobilized antibody ICR-4.2 PBMC incubated for 3 days on immobilized ICR-1.1 or ICR-4.2 had increased levels of cell surface expression of both the transferrin receptor and IL-2 receptor but not CD69. However, while increased expression of these lymphocyte activation markers was observed after 1 and 3 days this increased expression was unaccompanied by increased cell size. These results suggest that the anti-ICAM-R monoclonals ICR-1.1 and ICR-4.2 are able to directly induce early events in PMBC activation in the absence of additional exogenous stimuli but this activation does not result in blast transformation and associated increases in cell size.

C. Effect of ICAM-R Specific Monoclonal Antibodies on Stimulation of PMBC Activation by Anti-CD3 Antibody

Anti-ICAM-R monoclonal antibodies were also tested for their ability to alter early events in PMBC activation stimulated by immobilized anti-CD3 monoclonal antibody G19 Ledbetter et al., J. Immunol., 135(4): 2331-2336 (1985)!. Monoclonal antibody G19 binds to the CD3 complex on T cells (the T cell receptor) and activates T cells. When PBMC were cultured in wells precoated with anti-CD3 antibody (0.05 .mu.g/well) alone, only CD69 expression was elevated after one day. After three days, cell surface expression of CD69, the transferrin receptor and the IL-2 receptor was dramatically elevated. Upregulation of these activation markers was correlated with increases in cell size.

Ten .mu.g of anti-ICAM-R monoclonal antibodies ICR-1.1 or ICR-4.2 (or control monoclonal antibodies to HLA Class I; Serotec, Oxford, England) were added per well of 96-well flat bottom tissue culture plates either in the presence or absence of anti-CD3 antibody initially added at 0.025 .mu.g/well and washed to remove unbound antibody. Freshly obtained PBMC were immediately added (2.times.10.sup.5 cells/well). The cells were then incubated for a total of either 16 hours or 3 days at which time the cells were removed and washed 2 times in ice cold FACS buffer. Two.times.10.sup.5 cells were then resuspended in 50 .mu.l ice cold FACS buffer, and 5 .mu.l of FITC-conjugated anti-CD69, anti-transferrin receptor, anti-IL-2 receptor antibody or anti-FITC conjugated control Ig was added. The cells were incubated at 4.degree. C. for 30 minutes and then washed 2 times in 0.5 ml ice cold FACS buffer. After the final wash the cells were resuspended in 0.5 ml FACS buffer and fluorescence determined by FACScan analysis. When PBMC were cultured for 3 days on 0.025 .mu.g/well immobilized anti-CD3 either alone or in the presence of immobilized antibody to HLA Class I, expression of the transferrin and IL-2 receptors is not upregulated at this low does of immobilized anti-CD3. In contrast, culturing of PBMC in the presence of 0.025 .mu.g/well immunobilized anti-CD3 and either immobilized anti-ICAM-R antibodies ICR-1.1 or ICR-4.2 antibodies resulted in significant upregulation of both the transferrin and IL-2 receptors. The effect was more pronounced with antibody ICR-1.1. Similar results were also obtained after 16 hours in culture. Low dose anti-CD3 in the presence of immobilized ICR-1.1 or ICR-4.2 antibody induced expression of CD69, but not the transferrin receptor, while low dose anti-CD3 (0.025 .mu.g/well) in the presence of immobilized anti-HLA-I did not induce increased expression of either CD69 or the transferrin receptor. These results indicate that these anti-ICAM-R antibodies may serve as costimulatory molecules in early immune cell activation events.

D. Stimulation of PMBC Proliferation in the Presence of IL-2

Preliminary experiments were performed to determine if anti-ICAM-R monoclonal antibodies could affect the late event of cell proliferation again as measured by .sup.3 H-thymidine incorporation assays.

Monoclonal antibodies to ICAM-R were tested for their ability to directly stimulate PMBC proliferation in either the presence or absence of human recombinant IL-2 which potentiates but does not induce cell proliferation. Ten .mu.g of ICAM-R monoclonal antibodies ICR-1.1 or ICR-4.2 (or control IgG.sub.1 and IgG.sub.2) antibodies) in PBS were added per well of 96-well flat bottom tissue culture plates and the plates were incubated for 3-4 hours at 37.degree. C. in a 5% CO.sub.2 incubator. After incubation, each well was rinsed 3 times with PBS and freshly obtained PBL were added to a final concentration of 2.times.10.sup.5 cells/well in a volume of 200 .mu.l . Ten units/ml human recombinant IL-2 (Genzyme, Boston, Mass.) was then added to selected wells. The plates were incubated for a total of 3 days at 37.degree. C. in a 5% CO.sub.2 incubator. .sup.3 H-thymidine incorporation by the PMBC was determined as described earlier in this example. The anti-ICAM-R antibodies ICR-1.1 and ICR-4.2 did not induce PMBC proliferation even in the presence of rIL2. Positive controls for lymphocyte proliferation included immobilized anti-CD3 and anti-LFA-1 (60.3) monclonal antibodies. These results indicate that while the immobilized anti-ICAM-R antibodies stimulate expression of activation markers such as CD69, etc., by themselves they do not directly stimulate the entry of large numbers of PBMC into S phase of the cell cycle.

E. Costimulation of Lymphocyte Proliferation by ICAM-R Specific Antibodies

Because anti-ICAM-R antibodies with anti-CD3 antibodies costimulated early PBMC activation events, anti-ICAM-R antibodies were tested for the ability to costimulate lymphocyte proliferation induced by immobilized anti-CD3 antibody. In addition, to determine whether anti-ICAM-R antibodies costimulate T-lymphocytes in the absence of accessory cells, anti-ICAM-R antibodies were tested for their ability to costimulate proliferation of pure CD4.sup.+ T-lymphocytes, isolated using negative selection. To isolate CD4.sup.+ cells PBMC were suspended in tissue culture medium, added to 75 ml tissue culture flasks (Corning) and incubated for 1 hour at 37.degree. C., 5% CO.sub.2. Plastic nonadherent cells were then removed from the flask by gently rinsing once with PBS. The nonadherent cell fraction was suspended (10.sup.7 cells/ml) in an antibody cocktail containing 1 .mu.g/ml anti-CD8 antibody (Pharmingen, San Diego, Calif.), 1 .mu.g/ml anti-CD19 (Becton Dickinson), 1 .mu.g/ml anti-CD11b (Becton Dickinson) in 10% FBS-PBS (coating medium), and incubated for 1 hour at 4.degree. C. Unbound antibody was removed by washing twice in coating medium. Cells were then resuspended (10.sup.7 cells/ml) in coating medium containing Goat-anti-mouse Ig coated magnetic beads (45 .mu.l/10.sup.6 cells)(Advanced Magnetics, Cambridge, Mass.) and incubated for 1 hour at 4.degree. C. Cells bound to magnetic beads were then removed from suspension using a strong magnet. CD4.sup.+ populations obtained using this method were found to be >90% pure by flow cytometric analysis. PBMC or CD4.sup.+ cells were adjusted to a concentration of 1.times.10.sup.6 viable cells/ml in tissue culture medium. Individual wells of a 96-well flat bottom tissue culture plate were precoated with 0.001 .mu.g anti-CD3 monoclonal antibody G19 per well. The plates were incubated for 3 hours at 37.degree. C. in a 5% CO.sub.2 incubator and unbound antibody was removed by rinsing the wells 3 times in PBS. After the final PBS wash, monoclonal antibodies to ICAM-R (ICR-4.2 or ICR-1.1) or control antibodies were immediately added to a final concentration of 10 .mu.g/well. The plates were then reincubated for an additional 3 hours at 37.degree. C. The wells were again washed three times with PBS to remove unbound antibody and freshly isolated PBMC were immediately added to the wells (2.times.10.sup.5 cells in a volume of 200 .mu.l/well). The plates were then incubated for 3 days. Lymphocyte proliferation was measured by .sup.3 H-thymidine incorporation by the PMBC or CD4.sup.+ cells. As shown in FIG. 10 immobilized anti-ICAM-R monoclonal antibodies ICR-1.1 and ICR-4.2 increased the PBMC and purified CD4.sup.+ cell response to anti-CD3. Effects of the immobilized anti-ICAM-R antibodies on PBMC aggregation (an earlier event than PBMC proliferation) induced by anti-CD3 monoclonal antibody were also examined in this experiment. Anti-CD3 stimulated aggregation was inhibited almost 100% by antibody ICR-1.1 but was unaffected by immobilized ICR-4.2 and minimally inhibited by antibodies ICR-2.1 and ICR-4.1.

The results of the assays for the ability of anti-ICAM-R antibodies to affect the proliferation of cells on which ICAM-R is expressed indicate that binding of the antibodies of the invention to ICAM-R transmits a direct intracellular signal to T lymphocytes which modulates cell proliferation.

F. Co-Stimulation of Lymphocytes by Soluble ICAM-R

shICAM-R (Example 9) was assayed for the ability to costimulate human lymphocyte activation. Human peripheral blood lymphocytes (PBL) were obtained by Ficoll-Hypaque centrifugation and 2.times.10.sup.5 cells per well were incubated in the presence of either media, plate bound shICAM-R, plate bound anti-CD3 (OKT3) or a combination of plate bound anti-CD3 and shICAM-R. At 17 hours and 4 days after initiation of culture cells were removed, stained with monoclonal antibodies to human lymphocyte activation antigens and analyzed by flow cytometry.

Human lymphocytes cultured for 4 days in the presence of plate bound anti-CD3 (0.5 ug/well) and shICAM-R (100 ng/well) express elevated levels of the activation antigens ICAM-1, IL-2 receptor and transferrin receptor compared to lymphocytes cultured in the presence of anti-CD3 alone. In contrast, lymphocytes cultured in the presence of soluble ICAM-R (100 ng/well) alone expressed no increased levels of these activation antigens compared to cells cultured in media alone.

Experiments were also performed to determine if ICAM-R is involved in early events of qualitatively distinct types of cell-cell contact dependent T-lymphocyte activation (e.g., responses to staph enterotoxin A and alloantigen).

G. Effect of ICAM-R Specific Antibodies on Superantigen-Induced Proliferation of PBL

Superantigen-induced proliferation and aggregation of human PBL were assessed in the presence of the ICAM-R specific antibodies of the invention. The effect of soluble and plate-bound anti-ICAM-R antibodies and anti-HLA class I control B-H9 (Serotec) antibodies on proliferation and cell aggregation was measured three days after stimulation of human PBL with Staphylococcus Enterotoxin A (SEA) (Toxin Technology, Sarasota, Fla.). Plate-bound antibodies were prepared on the day of culture as follows. Purified antibody (10 .mu.g in 0.1 ml PBS) was added to individual wells of 96-well flat bottom plates. Plates were then incubated for 4 hours at 37.degree. C. Following incubation, unbound antibody was removed by aspirating each well and rinsing 4 times with fresh PBS. Human PBL were isolated from healthy donors on Histopaque (Sigma) gradients. Fresh peripheral blood was mixed with an equal volume of phosphate buffered saline (PBS), layered onto Histopaque and centrifuged at 450.times.g for 20 minutes with no brake applied. Lymphocyte fractions were collected and washed twice by adding a fresh volume of RPMI supplemented with 10% fetal bovine serum and centrifuging at 200.times.g for 8 minutes. PBL were suspended in a final volume of 10 ml of RPMI-FBS. Viable PBL were counted using the method of vital dye exclusion. Twenty .mu.l of a dilution of cell suspension in 0.4% trypan blue stain (Gibco) was added to a hemacytometer chamber and dye-excluding cells were then counted using an inverted microscope. Two-hundred thousand viable PBL were then added to 96-well flat-bottom tissue culture plates containing 100, 10 or 1 .mu.g soluble or plate-bound ICR-1.1, ICR-2.1, ICR-3.1, ICR-4.2, ICR-5.1, ICR-6.2, ICR-7.1, ICR-8.1, ICR-9.2, ICR-12.1, ICR-13.1, ICR-14.1, ICR-15.1, ICR-16.1, ICR-17.1, B-H9 or IOT2 (AMAC, Inc., Westbrooke, Me.) antibodies. Finally, each culture was stimulated with SEA (1000 or 10 pg/ml in triplicate) and cultured at 37.degree. C. in 5% CO.sub.2. After 3 days, proliferation was measured as .sup.3 H-thymidine incorporation.

Treatment with soluble anti-ICAM-R antibodies failed to alter proliferation in comparison to soluble control antibodies. Plate-bound (i.e., cross-linked) antibodies ICR-1.1, ICR-2.1, ICR-5.1, ICR-6.2 ICR-8.1 and ICR-17.1 however, significantly inhibited proliferation in response to SEA (p<0.05) while antibodies ICR-3.1, ICR-4.2, ICR-7.1, ICR-9.2, ICR-13.1, ICR-14.1 and ICR-15.1 did not (FIG. 11A and FIG. 11B). Antibodies ICR-12.1 and ICR-16.1 inhibited proliferation slightly, while antibodies ICR-12.1, ICR-13.1, ICR-14.1, ICR-15.1 and ICR-16.1 exhibited enhancing effects at the lowest concentration. Antibodies ICR-1.1 and ICR-8.1 were the most effective at inhibiting proliferation. FIG. 11C presents logistic dose response curves for monoclonal antibodies ICR-1.1, ICR-2.1, ICR-5.1, ICR-6.2 and ICR-8.1 in terms of the percentage of proliferation observed compared to proliferation in the presence of control antibodies and Table 10 below sets out the IC.sub.50 values obtained from the curves.

Concomitant to inducing entry into the cell cycle, SEA induces cell aggregation. Effects of the monoclonal antibodies ICR-1.1 and ICR-4.2 on cell aggregation were measured using an inverted microscope. Plate-bound ICR-1.1 also significantly inhibited cell aggregation at both SEA concentrations in comparison to plate-bound B-H9 and ICR-4.2 antibodies. Inhibition of aggregation by plate-bound ICR-1.1 was almost complete. In contrast, plate-bound ICR-4.2 antibody only slightly inhibited aggregation in comparison to plate-bound B-H9. Aggregation of PBL induced by SEA was not affected by soluble anti-ICAM-R antibodies ICR-1.1 or ICR-4.2 in comparison to soluble B-H9 antibody.

The minimum time required for plate-bound anti-ICAM-R to inhibit SEA-induced proliferation was also determined. PBL were pre-incubated on plate-bound ICR-4.2, ICR-1.1 or isotype-matched anti-HLA-I control antibodies B-H9 (IgG.sub.1) and IOT2 (IgG.sub.2) with or without SEA (10 .rho.g/ml) for 3, 5 and 7 hours. PBL were then transferred to clean wells and cultured in the presence of SEA (10 pg/ml) for 3 days. The results of .sup.3 H-thymidine incorporation (proliferation) assays are summarized in FIG. 12. Immobilized ICR-1.1 antibody and, to a lesser extent ICR-4.2 antibody, significantly reduced proliferation in comparison to isotype-matched controls after only 3 hours of incubation. This result indicates that binding of plate-bound ICR-1.1 or ICR-4.2 to ICAM-R transmits an intracellular signal capable of inhibiting proliferation even after cells have been removed from the immobilized antibodies. These results suggest that therapeutically efficacious engagement of ICAM-R may be achieved without maintaining saturating levels of an ICAM-R specific agent (e.g., a monoclonal antibody) over long periods of time.

Because both T cells and accessory cells express high levels of ICAM-R, the inhibition of cell-cell contact dependent T cell activation during the response to SEA by ICR-1.1 could be mediated by ICR-1.1 binding to T cells, accessory cells or both. Additionally, because ICAM-R and ICAM-1 differ markedly in their expression on nonactivated T cells, it is possible that anti-ICAM-1 and anti-ICAM-R may inhibit the SEA response by targeting T cell subsets in different states of activation. Because the role of ICAM-R may differ in naive and memory cells, the ability of anti-ICAM-R antibodies to inhibit SEA induced proliferation of CD4.sup.+ CD45RO.sup.+ ("memory") cells, or CD4.sup.+ CD45RA.sup.+ ("resting") cells was tested. Plasmatic nonadherent PBMC (10.sup.7 cells/ml) were incubated for 1 hour at 4.degree. C. with a cocktail of antibodies (1 .mu.g/ml each) containing anti-CD8, anti-CD19, anti-CD11b, anti-HLA-DR (Becton Dickinson) and either anti-CD45RO (Amac) (to obtain CD45RA.sup.+ CD4.sup.+ cells), or anti-CD45RA (Amac) (to obtain CD45RO.sup.+ CD4.sup.+ cells) in coating medium. The cell suspension was washed twice with coating medium to remove unbound antibody and incubated with goat anti-mouse IgG coated magnetic beads. Cells bound to magnetic beads were then removed from the suspension using a strong magnet. CD45RO.sup.+ and CD45RA.sup.+ populations obtained using this method were found to be >95% pure as determined by flow cytometric analysis. Two hundred thousand purified memory T cells, resting T cells or plastic adherent cells were incubated on immobilized ICR-1.1, anti-ICAM-1 antibody LB-2 or anti-HLA-I antibody p10.1 (10 .mu.g/ml) (Gerald Nepom, Virginia Mason Research Center, Seattle, Wash.) for 3 hours. The antibody treated memory or resting T cells were removed to clean wells and admixed with 2.times.10.sup.4 plastic adherent cells. Antibody treated accessory cells were admixed with either untreated memory T cells or untreated resting T-cells. Each reconstituted culture was then stimulated with SEA (10 pg/ml). The results of .sup.3 H-thymidine incorporation (proliferation) assays are summarized in FIG. 13 wherein the abbreviation "APC" stands for "antigen presenting cells," which are the accessory cells in this assay, and wherein the asterisks indicate the population of cells pretreated with antibody. Pretreatment of CD45RO.sup.+ T cells or accessory cells with ICR-1.1 blocked proliferative responses to SEA in comparison to p10.1 control antibody. When both cell populations were treated with ICR-1.1, the inhibitory effect was additive. Inhibition of proliferation by the anti-ICAM-1 antibody LB-2, occurred only when adherent cells were pretreated and was not further enhanced when the admixed cells were also pretreated. As shown in FIG. 14 pretreatment of CD45RA.sup.+ T cells with ICR-1.1 did not affect SEA responses. ICR-1.1 or LB2 pretreatment of adherent cells resulted in modest inhibition of CD45RA.sup.+ cell proliferation.

H. Inhibition of Lymphocyte Proliferation in Response to Allogenic Irradiated Stimulator Cells

Monoclonal antibodies to ICAM-R were also tested for the ability to alter lymphocyte proliferation (as measured by .sup.3 H-thymidine incorporation) in response to alloantigenic irradiated stimulator cells. Responder cells were prepared by obtaining PBMC from a normal donor using Histopaque centrifugation as described above. To prepare stimulator cells, PBMC from a second, unrelated donor were concurrently isolated and irradiated at 1500 R by exposure to a gamma emitting cesium source. Two hundred thousand responder cells and 2.times.10.sup.5 irradiated stimulator cells (suspended in culture medium) were then added to wells containing soluble or immobilized ICR-1.1, ICR-2.1, ICR-3.1, ICR-4.2, ICR-1.5, ICR-6.2, ICR-7.1, ICR-8.1, ICR-9.2, immobilized B-H9, immobilized p10.1, or soluble 515F (anti-rat CD18) antibody and incubated for 6 days at 37.degree. C., 5% CO.sub.2. Lymphocyte proliferation (.sup.3 H-thymidine incorporation) was assessed in the last 18-24 hours of culture.

Immobilized monoclonal antibodies ICR-1.1, 2.1, 6.2 and 8.1 consistently reduced proliferation in comparison to control antibodies. ICR-8.1 also inhibited alloantigen-stimulated proliferation when administered in soluble form.

I. Inhibition of IL-2 Production by T Lymphocytes

Human PBL were obtained by Ficoll-Hypaque centrifugation of whole peripheral blood. Adherent cells were depleted by incubation on plastic and nonadherent cells were subjected to discontinuous centrifugation on Percoll gradients to further separate subsets of lymphocytes into medium buoyant density (fraction B) and high buoyant density (fraction C) cell populations. Prior to cell addition, wells were coated with monoclonal antibodies by addition of 0.1 ml each antibody at 5 ug/ml in PBS per microtiter well. Following antibody addition, all wells were incubated overnight at 4 degrees and each well was washed free of unbound antibody by PBS rinsing prior to addition of cells. Each cell fraction was then incubated for 18 hours on either ICR-1.1, ICR-4.2 or control antibodies to human major Histocompatibility Complex Class I (MHC Class I). For these experiments 2.times.10.sup.5 cells per well were added in a volume of 0.2 ml RPMI-1640 containing 10% FCS to individual wells of a 96 well flat bottom microtiter plate (Costar, Cambridge, Mass.). After 18-20 hours incubation at 37.degree. C., the cells were collected from the microtiter wells and washed twice in RPMI-1640 media containing 10% FCS and adjusted to 1.times.10.sup.6 per ml in RPMI-1640 containing 10% FCS. Two.times.10.sup.5 prepulsed cells were then added to wells previously coated with 0.1 ml anti-CD3 antibody (clone OKT3 at 5 ug/ml in PBS) and the cells were incubated for 20-24 hours at 37 degrees. After incubation supernatants were obtained from each well and replicate supernatants were pooled, frozen at -80 degrees and assayed for IL-2 content by ELISA (Biosource).

Fraction C cells, composed largely of quiescent CD3 positive cells, produced ample IL-2 when prepulsed for 18 hours in wells containing either no antibody or a variety of control antibodies to MHC Class I. Cells prepulsed on ICR-1.1, however, produced less than 50% of IL-2 produced following prepulse on negative control antibodies while cells prepulsed on ICR-4.2 exhibited no decreased IL-2 production. Thus, not all ICAM-R specific antibodies were efficacious in inhibiting IL-2 release. Engagement of ICAM-R in an epitope specific manner is required for this effect to be achieved. It is anticipated that ICAM-R specific antibodies whose binding sites on ICAM-R overlap significantly or are identical to that bound by ICR-1.1 (e.g., ICR-8.1, see Example 14) would manifest similar effects.

J. Restoration of Anti-CD3 Mediated Proliferation by Addition of IL-2

Human PBL were fractionated and incubated on immobilized monoclonal antibodies as described in Section I above. The cells were collected, washed and replated on anti-CD3 either in the presence or absence of human rIL-2 at 30 U/ml. Addition of IL-2 completely restored proliferative responses to anti-CD3 by resting lymphocytes prepulsed on ICR-1.1, indicating that the inhibitory effect of ICR-1.1 was not due to irreversible toxicity to the cells.

K. Induction of IL-8 Release

Monocytes were isolated by elutriation from peripheral blood of normal donors. Plastic wells were coated with ICR-1.1, 2.1, 8.1 or albumin alone (10 .mu.g/ml). After blocking free sites, the cells were placed into the wells in medium. After 1 hour in culture, the cells that were in the ICR Mab treated wells had flattened onto the substratum. Those cells plated in wells treated with BSA plus or minus LPS were rounded and not spread. After 8 or 18 hours of incubation at 37.degree. C., the medium was removed and assayed for IL-8 immunoreactivity. At the 8 hour time point, all ICR Mab tested showed enhanced levels of IL-8 (four times over albumin control). By 18 hours IL-8 levels from ICR Mab were much elevated over control levels. ICR-1.1 induced levels increased ten times, ICR-2.1 induced levels were four times and ICR-8.1 induced levels were two times over the levels seen at the 8 hour time point. Similar experiments with monocytic cell lines (U937 and HL60) to determine if ICR Mab can induce IL-8 release were conducted. U937 cells responded to each of the ICR Mab treated wells by releasing IL-8 into the conditioned medium. ICR-1.1 elicited the most robust response which was 3-fold greater than the release from ICR-4.2, 6.2 or 8.1 treated wells, each of which showed levels twice that of the BSA control alone. HL60 cells did not respond to the ICR Mab treated wells by releasing IL-8 into the medium. LPS did induce a marked release from HL60 cells. No detectable morphological changes were detected with the U937 or HL60 cells.

Monocytes isolated and treated as described above were tested for release of MCP-1 at 8 and 18 hour time points. Mab ICR-1.1, 2.1 and 8.1 each induced release of MCP-1 into the conditioned medium although the kinetics of release differed. MCP-1 release elicited by ICR-1.1 peaked at 8 hours. Release from Mab ICR-2.1 and ICR-8.1 was not detected until 18 hours when it was 4-fold greater than the peak 1.1 induction level.

These results imply a potentially significant role of ICAM-R in the human disease atherosclerosis since engagement of ICAM-R in the presence of a pro-atherosclerotic compound (e.g., oxidated phospholipid) promotes synthesis/secretion of MCP-1 which has recently been implicated as a pro-atherosclerotic chemokine Edgington, BIO/TECHNOLOGY, 11:676-681 (1993)!.

L. Upregulation of the Activation Antigens CD69 and CD25

Resting PBL isolated as described in Section I above were prepulsed on ICR-1.1 or negative control antibodies, washed and incubated on immobilized anti-CD3 for 24 hours. The cells were then collected, labeled with monoclonal antibodies to the lymphocyte activation antigens CD69, CD25 and CD80 and examined by flow cytometry. Cells prepulsed on ICR-1.1 do not exhibit decreased ability to upregulate the activation antigens CD69 and CD25 in response to immobilized anti-CD3 compared to the level of CD69 and CD25 expressed by cells prepulsed on control antibodies.

M. Increased Tyrosine Phosphorylation in Human PBL

Human PBL was obtained by Ficoll-Hypaque centrifugation and were incubated for 5 minutes with soluble ICR-4.2, OKT3 or anti-HLA-I (1.times.10.sup.7 cells were incubated with each antibody at 30 ug/ml). The cells were then washed and goat anti-mouse IgG (Cappell) was added to a final concentration of 100 ug/ml. After varying periods of time, the cells were lysed in detergent and lysates were electrophoresed on a 10% acrylamide gel, transferred to blotting paper. Blots were then probed with the anti-phosphotryosine antibody 4G10.

ICR-4.2 induced phosphorylation on tyrosine of numerous substrates rapidly after crosslinking compared to the negative control antibody to MHC Class I. Phosphorylation of substrates were also observed in response to the positive control (crosslinked OKT3 antibody).

N. Effect of ICAM-R Binding on Early Signalling Events in PBL

Given the ability of antibodies to domain 1 of ICAM-R to inhibit subsequent T cell activation, intracellular T cell signalling pathways were examined in order to understand which signalling events are affected by ICAM-R. More specifically, the ability of cross-linked T cell antigen receptor to induce tyrosine phosphorylation was examined. T cells are known to show a rapid induction of tyrosine phosphorylation in response to antigen presentation or various cellular substrates. This induction is known to be essential to subsequent proliferation and IL-2 production.

Briefly, resting PBLs were isolated and treated with ICR-1.1 or isotype matched control antibody as described above in Section I. After the cells were removed from antibody, they were washed one time in PBS, resuspended in PBS to a concentration of 1.times.10.sup.6 cells/ml, and either treated or not treated with antisera to the T cell receptor (G19-4) for 2 minutes at 37.degree. C. The cells were then spun down and lysed in boiling 2.times.SDS-sample buffer, boiled a further 5 minutes and resolved on 10% SDS-PAGE. Proteins were transferred to nitrocellulose and blotted with the 4G10 anti-phosphotyrosine antibody as described. Regardless of whether the cells had been pretreated with ICR-1.1 or matched isotype antibody, there was an induction of tyrosine phosphorylation in response to G19-4 treatment. This suggests that, at least on a gross level, signalling through tyrosine kinases is normal in these cells. This is consistent with the result that CD69 upregulation in these cells is also normal (Section L above), since CD69 expression requires PLC-gamma stimulated PKC activity and PLC-gamma activation requires tyrosine phosphorylation. The notion that ICR-1.1 pretreatment inhibits a signalling event parallel or subsequent to PLC-gamma activation is also supported by the result that PMA/ionomycin treatment does not induce normal T cell activation in the pretreated cells.

O. ICAM-R Engagement and Translocation of NFAT

Translocation of the transcription factor NFAT from the cytoplasm to the nucleus is essential for IL-2 gene transcription. The presence of NFAT complexes in the nuclei of cells in response to ionomycin (a member of a class of compounds which cause calcium transport across cell membranes and which can signal the Ca.sup.2+ dependent pathway associated with TCR stimulation) may be assayed as follows. Briefly, an oligonucleotide corresponding to the IL-2 promoter of the human IL-2 gene is end-labelled with .sup.32 P and purified. Proteins in the nuclear fraction of ionomycin-treated Jurkat cells are isolated and then incubated with the oligonucleotide. Resulting complexes are resolved on a non-denaturing PAGE gel. In this assay, a Jurkat cell line which produced IL-2 at normal levels contained a transcription complex which formed in the presence of ionomycin but, as expected, was not formed when cells were pretreated with cyclosporin A. A Jurkat cell line which did not produce IL-2 at normal levels failed to form this transcription complex in response to ionomycin.

This result is significant since the NFAT family of transcription factors are thought to be the proximal targets for calcineurin. Calcineurin, in turn, is the intracellular target for cyclosporin A and FK506, two drugs which have been utilized to support tissue transplantation. Since ICAM-R may engage the same pathway as calcineurin but is expressed more selectively (e.g., on leukocytes) than calcineurin, engagement of ICAM-R may have a more selective therapeutic effect.

P. ICAM-R Specific Antibody Inhibits of Normal Alloantigen Presentation to CD4.sup.+ T Cells

The expression of ICAM-R in normal skin, psoriasis, atopic eczema and cutaneous T cell lymphoma was examined. Five .mu.m cryostat sections of skin were stained using monoclonal antibodies to ICAM-R (ICR-1.1 and ICR-8.1) and a well characterized immunoperoxidase technique. In normal skin, ICAM-R was expressed by all cutaneous leucocytes but most striking was the strong expression of ICAM-R by Langerhans cells (Lcs) within both epidermis and dermis. This observation was confirmed by double-labeling with CD1a (a Langerhans cell marker) and negative staining with an IgG.sub.1 isotype control. In psoriasis, atopic eczema, and cutaneous T cell lymphoma ICAM-R was co-expressed in all CD1a.sup.+ cells.

Blocking experiments were performed to determine whether the observed ICAM-R expression on Lcs was functionally important in antigen presentation. CD4.sup.+ T cells were prepared from peripheral blood and 10.sup.5 CD4.sup.+ T cells were combined with 4.times.10.sup.4 epidermal cells harvested from keratome biopsies of normal skin of an individual allogenic to the T cell donor. Proliferation was measured by .sup.3 H thymidine uptake. Alloantigen presentation was unaffected by addition of 50 .mu.g IgG.sub.1 isotype control. Addition of 50 .mu.g anti-ICAM-R antibody ICR-8.1 to the co-culture resulted in a marked (47%) reduction in degree of Lcs alloantigen-driven proliferative response of the T cells. Inhibition was 73% of that produced by addition of anti-LFA-1 (anti-CD11a) antibody.

Example 21

Table 11 below is a summary of certain characteristics of ICAM-R specific monoclonal antibodies of the invention which have been specifically described in the foregoing examples. In Table 11, the abbreviation "NC" stands for "not conclusive" and the abbreviation "ND" stands for "not determined." The antibodies marked with an asterisk in Table 11 enhanced activation at low concentrations.

Example 22

One inference from the aforementioned examples that antibodies specific for ICAM-R modulate the response of lymphocytes to a variety of stimuli (e.g., SEA and allogeneic cells) is that engagement of ICAM-R by either its natural counter-receptors or by antibodies of the invention transduces a signal to the ICAM-R expressing cell. ICAM-R specific signalling events are likely to involve the interaction of the cytoplasmic domain of ICAM-R with cellular enzymatic components (e.g., kinases, phosphatases) of one or more second messenger pathways and/or with cytoskeletal components in a pattern unique to ICAM-R.

Preliminary experiments are consistent with this concept and with the idea that ICAM-R is distinct from ICAM-1 in its linkages with second messenger systems. Extracts from unstimulated Raji cells were prepared, fractionated and assayed for kinase activity as follows. Seven.times.10.sup.7 cells were washed once in PBS and lysed in buffer containing 20 mM Tris pH 7.5, 0.5 mM EDTA, 1% Triton X-100 (Pierce), 10 ug/ml pepstatin and leupeptin (Boehringer), 2 mM PMSF for 1 hour on ice. Lysates were pelleted in a refrigerated microfuge at 14,000 rpm for 15 minutes and the resulting supernatant was applied to a DEAE sephacel column (Pharmacia) equilibrated in 20 mM Tris pH 7.5, 0.5 mM EDTA (Buffer A). The column was run at a rate of 0.25 ml/minute and developed with a gradient of 0 to 0.35M NaCl in buffer A over 60 minutes. In these initial experiments, only those fractions enriched in protein kinase C (PKC) activity (as determined using an Amersham assay kit and following manufacturers instructions) were examined. Fractions enriched in PKC activity were pooled and used as a source of kinase(s) to test for differential phosphorylation of synthetic peptides of the complete cytoplasmic domains of ICAM-1, ICAM-2 and ICAM-R (amino acids 481 to 518 of SEQ ID NO: 1). Assays were performed according to manufacture's instructions with peptides at 75 uM final concentration. Ten ul of the reaction mixture was boiled in 30 ul Laemmli sample buffer and resolved on a 12.5% SDS-PAGE gel. Following a 1.5 hour exposure of the gel on X-ray film phosphorylation of ICAM-R and ICAM-2 but not ICAM-1 was detected. Whether the phosphorylation was due to PKC or another co-fractionated kinase was not determined.

Further assays involved reacting fractions derived either from a column chromatography step or from solubilized cell fractions in the presence of Ca.sup.++, Mg.sup.++, cAMP, phosphatidylserine, cytoplasmic tail peptide and .sup.32 P!ATP. Phosphorylation of specific peptides was assessed following resolution by gel electrophoresis. Jurkat cells were separated into subcellular fractions and each fraction was assayed for kinase activity on the cytoplasmic tail peptides. In these experiments, phosphorylation of ICAM-1 and ICAM-R was detected. However, kinases which phosphorylated ICAM-1 associated with cell membrane fractions, whereas kinases which phosphorylated ICAM-R were primarily cytosolic although also present in membranes. Additional support for different kinases acting on these two ICAM's comes from preliminary purification studies of these kinases. Jurkat cytosol fractionated on a MonoQ column (Pharmacia) equilibrated in 50 mM Tris pH 8, 5 mM EDTA and developed with a gradient to 0.6M NaCl over 30 minutes gives a very broad activity profile for kinases acting on ICAM-R. Only a subset of these fractions also have activity towards ICAM-1. This provides additional evidence that cellular kinases exist which differentially phosphorylate ICAM-R but not ICAM-1. Two dimensional phosphoamino acid analysis on these phosphorylated peptides shows only serine phosphorylation on ICAM-R and threonine phosphorylation on ICAM-1.

The ability of the protein tyrosine kinases p56.sup.lck (UBI, Lake Placid, N.Y.), and p59.sup.fyn (UBI, Lake Placid, N.Y.) to phosphorylate a cytoplasmic tail peptide (amino acids 482 to 518) of ICAM-R was measured in vitro.

The assay was performed in triplicate as follows. Five .mu.l assay buffer stock solution (250 mM Tris pH 6.8, 125 mM MgCl.sub.2, 25 mM MnCl.sub.2, 0.25 mM Na.sub.3 VO.sub.4), 5 .mu.l cdc 2 (UBI, positive control) or ICAM-R cytoplasmic tail peptide or scrambled ICAM-R cytoplasmic tail peptide or H.sub.2 O, 10 .mu.l (1 u) p56.sup.lck or p59.sup.fyn, and 5 .mu.l ATP stock solution (0.25 .mu.Ci .gamma..sup.32 PATP in 500 .mu.M cold ATP) were mixed in a microfuge tube and incubated 30 minutes at 30.degree. C. Ten .mu.l 50% acetic acid were then added. Samples of 25 .mu.l were spotted on P81 phosphocellulose paper and washed four times with 150 mM phosphoric acid. The papers were then dried and Cerenkov counted for one minute.

p59.sup.fyn phosphorylated the cytoplasmic tail of ICAM-R (but not the scrambled cytoplasmic tail peptide) and did so approximately 1.8 times better than it phosphorylated an equimolar amount of cdc 2 peptide. p56.sup.lck failed to phosphorylate any of the substrates at a concentration of 1 u (unit), but was able to phosphorylate ICAM-R cytoplasmic tail peptide and scrambled peptide at concentrations of 10 u.

Preliminary experiments also indicate that the cytoplasmic domain of ICAM-R differentially associates with cytoskeletal components. Binding of the non-competing monoclonal antibodies ICR-1.1 and ICR-4.2 to ICAM-R was examined to assess the potential influence of each antibody on the association of lymphocyte ICAM-R with the cytoskeleton. The antibodies may mimic distinct natural ICAM-R ligands which employ ICAM-R as a cell surface receptor through which regulated cellular responses may be elicited.

Other investigators have previously observed that numerous human T lymphocyte surface antigens which occur as cell surface transmembrane glycoproteins can be induced to associate with the cytoskeleton if cell surface-bound antibody specific for these antigens is crosslinked with secondary antibodies Geppert et al., J. Immunol., 146: 3298 (1990)!. Many of these cell surface molecules are defined components of lymphocyte adhesion and/or activation pathways. The phenomenon of inducible association with the cytoskeleton is operationally defined as the resistance of cell-surface immune complexes to detergent extraction under defined conditions. Inducible detergent resistance does not require metabolic energy and can be observed in cells maintained at 0.degree.-4.degree. C. throughout the experiment.

Experiments were conducted using freshly prepared human PBL or the human T lymphoblastoid cell line CEM-CCRF (ATCC CCL119). Briefly, freshly drawn human blood from healthy volunteer donors was diluted 1:1 with PBS and layered onto Sigma HistoPaque density separation medium. The gradients were centrifuged for 30 minutes at 1500 rpm (600.times.g) and the mononuclear cell fraction at the interphase was collected and washed three times with PBS. The cell pellet was resuspended in complete RPMI-1640 medium (Gibco, supplemented with L-glutamine, penicillin/streptomycin, sodium pyruvate, 2-mercaptoethanol, and 10% FBS) and plated onto tissue culture-treated petri dishes for adherent cell depletion. Plates were incubated 1-2 hours at 37.degree. C., 5% CO.sub.2 after which nonadherent PBL were harvested and washed twice with ice-cold PBS. Conjugation of monoclonal antibodies to fluorescein using fluorescein isothiocyanate (FITC) was performed according to published procedures see, e.g., Goding, J. Immunol. Meth., 13: 215 (1976)! and, in brief, involves incubation of purified antibody with an excess of FITC (Sigma) in 0.1M bicarbonate buffer pH 8.1 for 90 minutes at 37.degree. C. followed by exhaustive dialysis against PBS to remove unreacted FITC.

PBL or washed CEM cell suspensions (1.times.10.sup.6 cells) were dispensed into Falcon 12.times.75 mm tubes in ice-cold PBS-5% FBS, pelleted, and resuspended in 50 .mu.l of FITC-conjugated anti-ICAM-R monoclonal antibody 26E3D-1 or 26I10E-2 adjusted to saturating concentration in the same buffer. Antibody binding was permitted to proceed for 30 minutes on ice, afterwhich unbound antibody was removed by pelleting cells which had first been resuspended in 1 ml of PBS-5% FBS through an underlaid cushion (0.7 ml) of neat (undiluted) FBS.

For groups stained with FITC-conjugated monoclonal antibody only, the 1 ml suspension was divided into two equal parts, each of which was separately underlaid with FBS, centrifuged, and the supernatant removed by aspiration. Cell pellets were then resuspended in 200 ul of control buffer (13 mM Tris pH 8.0, 150 mM NaCl, 2 mM MgCl.sub.2, 2 mM EGTA, 2% FBS, 2.5 ug/ml aprotinin, 1 mM PMSF, 10 mM iodoacetamide) or detergent buffer 0.5% NP-40 (v/v) (US Biochemical, Cleveland, Ohio) in control buffer! and held for 20 minutes at room temperature, or overnight at 4.degree. C., prior to FACS analysis. For groups in which cell surface-bound monoclonal antibody was crosslinked with secondary antibodies, following the first antibody staining step, washed cell pellets were resuspended in 50 ul of FITC-goat anti-mouse IgG (Sigma) diluted 1:100 in PBS-5% FCS and incubated for 30 minutes on ice. The cells were then resuspended, divided into two tubes as described above, pelleted, and buffer-treated in the presence or absence of detergent. FACS analysis was then performed on the cells.

Results (see FIG. 15) obtained for CEM cells were similar to those seen with PBL. ICAM-R association with the cytoskeleton as assessed by the detergent resistance assay was negligible when FITC-conjugated ICR-4.2 or ICR-1.1 antibodies alone were permitted to bind to cell surface ICAM-R. However, when cell surface-bound ICR-4.2 antibody was crosslinked with secondary antibodies, a modest increase in detergent resistance was detected. If secondary antibodies were used to crosslink cell surface-bound ICR-1.1, which recognizes a distinct ICAM-R epitope from that seen by ICR-4.2, a much greater (approximately 2-fold in PBL and 2-3 fold in CEM) increase in detergent resistance was reproducibly observed. Interaction of ICAM-R ligands with different structural regions of ICAM-R thus appears to differentially influence association of ICAM-R with the cytoskeleton.

Example 23

Characterization of ICAM-R interaction with specific cytoplasmic proteins was conducted.

A. Dihybrid Screen

The two-hybrid system developed in yeast Chien et al., Proc. Natl. Acad. Sci. USA, 88: 9578-9582 (1991)! was used to screen for products of a human lymphocyte cDNA library capable of interacting with the carboxy-terminal cytoplasmic tail of ICAM-R. This yeast dihybrid screen is based on functional in vivo reconstitution of the GAL4 transcription factor. The separable DNA-binding and transcription-activating domains of GAL4 were engineered into distinct plasmids as portions of novel fusion proteins. Under defined conditions GAL4 activity is measureable by assay of the beta-galactosidase reporter gene.

One plasmid, the "bait" vector (pAS1), contained sequences encoding the GALA DNA-binding domain amino acids 10-147, Keegan et al., Science 231:699-704 (1986)!, a -trp requirement, the HA epitope tag, and a polylinker region into which the ICAM-R cytoplasmic domain sequence was ligated at the BamHI site. The ICAM-R cytoplasmic domain was amplified by PCR from pVZ-147 ICAM-R DNA (Example 4) using the oligonucleotide primers:

DH3 (SEQ ID NO: 111)

CAGTGGGATCCTGTTAATGTACGTCTTCAGGG and

DH4 (SEQ ID NO: 112)

TGGGAGTTTGAAGGCTTT.

and then inserted at the BamHI site. The resulting construct, termed plasmid 9.4, was sequenced to confirm orientation and rule out PCR errors. Yeast strain Y190 (genotype MAT.alpha. gal4 gal80 his3 trp1-901 ade2-101 ura3-52 leu-3,-112 +URA3::GAL.fwdarw.lacZ, LYS2::GAL.fwdarw.HIS3 cyh.sup.r) was transformed with plasmid 9.4 by standard methods and grown in selective (-trp) media to mid-log phase. Cells were lysed with glass beads in lysis buffer and 50 ug of protein was loaded onto a 10% polyacrylamide gel which was electrophoresed and blot-transferred to a PVDF (Millipore) membrane by standard procedures. Control lanes of the gel contained lysate material from pAS1-transformed Y190. Blots were developed using anti-HA monoclonal antibody 12Ca5 (BAbCo, Berkeley, Calif.), rabbit anti-mouse IgG, and .sup.125 I-labeled protein A to confirm that the chimeric fusion protein (ICAM-R cytoplasmic tail/HA/GAL4 DNA-binding domain) was expressed at readily detectable levels.

The second expression plasmid, or "prey" vector (pACT), consists of sequences encoding the GAL4 activation domain II amino acids 768-881, Ma et al., Cell, 48:847-853 (1987)! fused to a human B cell cDNA library inserted at the Xho site of the vector, and a-leu selection requirement.

The 9.4-transformed Y190 cells were transformed by standard methods with pACT library DNA and grown under selective conditions (-leu/-trp/-his/3-aminotriazole). Only cells in which an interaction occurred between the ICAM-R cytoplasmic tail domain of the 9.4 chimeric protein product and an unknown (B cell cDNA library-derived) protein sequence fused to pACT GAL4 domains survived. This interaction was required to reconstitute GAL4 activity.

Fifty colonies grew using this selection method and were tested for beta-galactosidase activity. Specificity of the ICAM-R cytoplasmic tail interaction with pACT fusion proteins was verified by inability of the latter to complement recombinant pAS1 vector expressing distinct "bait" proteins p53, ICAM-1, ICAM-2, a kinase (surose non-fermemtor 1, snfl), and casein kinase inhibitor (CKI.DELTA. or CKIa)! using the dihybrid selection conditions described above. Sequence analysis of B cell cDNA-derived, pACT inserts obtained by this method revealed twenty novel sequences and thirty sequences encoding known proteins out of the fifty inserts.

Two of the known proteins, alpha-tubulin and protein kinase C inhibitor protein (PKCIP), were further investigated for their ability to interact with ICAM-R cytoplasmic tail. Alpha-tubulin, along with beta-tubulin, is a principal component of cytoplasmic microtubules, one major class of polymeric cytoskeletal proteins. The PKCIP which interacted with ICAM-R cytoplasmic has a sequence identical to that of human HS1-beta (GenePro Accession No. gp x57346), a known phospholipase and member of the highly conserved 14.3.3 family of PKC regulatory proteins.

Mutagenesis was employed to map the ICAM-R cytoplasmic tail sites responsible for interaction with the alpha-tubulin and PKCIP pACT plasmid products. Four ICAM-R cytoplasmic tail mutant sequences were produced in pAS1 by in vitro mutagenesis of plasmid 9.4 using the following mutagenic oligonucleotides wherein, for example, the E495D oligonucleotide introduces an aspartic acid at position 495 which is a glutamic acid in wild-type ICAM-R:

E495D (SEQ ID NO: 113)

TACATGTTAGGGAGGACAGCACCTAT,

EA94D (SEQ ID NO: 114)

TACCATGTTAGGGACGAGAGCACCTAT,

E495A (SEQ ID NO: 115)

TACCATGTTAGGGAGGCCAGCACCTAT,

E494 (SEQ ID NO: 116)

TACCATGTTAGGGCCGAGAGCACCTAT.

The resulting plasmids were transformed into Y190 cells, and cotransformed with either the alpha-tubulin or the PKCIP pACT plasmid. All contransformants grew in the selective medium described above and tested positive for beta-galactosidase activity, indicating that these mutations did not disrupt interactions between the GAL4 fusion proteins encoded by each plasmid.

Additional mutant 9.4 plasmids shown in Table 12 below were constructed by similar methods and were cotransformed into Y190 cells with the positive bait vectors tubulin (GenePro gp K00558), EF-8 (GenePro gp M27364), EF-11 (GenePro gp 29548), HS1-beta, HS1-theta (SWISS-PROT sp 27348), actin (PIR-Protein pir 505430), triose-6-isomerase (SWISS-PROT sp P00939), and proteosome (SWISS-PROT sp P25786)! that had been identified in the original screening of the cDNA library. Transformants were tested for .beta. galactosidase activity in order to map residues in the ICAM-R cytoplasmic domain contributing to each interaction. The results are illustrated in Table 12 wherein "blue" indicates no effect, "white" indicates a complete disruption and "Lbal" indicates a minimal disruption in an interaction and mutations R482-R493 and R482-Q506 represent the amino acid residues remaining after gross deletions of other amino acids of the ICAM-R cytoplasmic tail.

For single and double point mutations, the interaction of the 14.3.3 proteins HS1-beta and HS1-theta, are only disrupted (i.e., white) by the change in mutant QP505/506 while the interaction with tubulin was disrupted by a variety of changes to the cytoplasmic tail. Mutations R482-R493 and R482-Q506 are gross deletions of the cytoplasmic tail. The deletion of amino acids 506 through 518 appears to not disrupt the interactions tested, while deletion of amino acids 493 to 518 appears to disrupt all interactions. Two additional bait vectors respectively encoding the ICAM-1 and ICAM-2 cytoplasmic tails were constructed by similar methods to that described for ICAM-R to further test the specificity of the alpha-tubulin and PKCIP interactions with the ICAM-R C-tail. When Y190 transformants expressing either the ICAM-1 or ICAM-2 cytoplasmic tail fusion proteins were cotransformed with the alpha-tubulin or PKCIP pACT plasmids described above, there was no evidence for interactions among fusion proteins, indicating specificity of the original ICAM-R interactive proteins.

B. ICAM-R C-tail affinity chromatography

Additional direct evidence for binding interactions between ICAM-R cytoplasmic tail and alpha-tubulin was obtained by a variety of methods. An ICAM-R cytoplasmic tail (C-tail) peptide corresponding to ICAM-R amino acids 482-518 of SEQ ID NO: 1 was synthesized by Macromolecular Resources, (Colorado State Univ., Boulder, Colo.). The peptide was immobilized (19 mg/ml suspension) on agarose beads (AffiGel 10, BioRad) according to manufacturer's instructions. Detergent lysate (20 ml, lysis buffer: 50 mM octyl glucoside, 50 mM Tris pH 7.5, 0.15M NaCl, 1 mM MgCl.sub.2, 1 mM CaCl.sub.2 and a protease inhibitor cocktail) of freshly isolated human tonsils (12 gm) was applied first to an ethanolamine-blocked AffiGel 10 precolumn and then to the ICAM-R C-tail beads (0.5 ml) for 3 hours at 4.degree. C. Beads were batch-washed with over 50 bed volumes of lysis buffer and packed into a glass column which was then eluted with 1 ml of soluble ICAM-R C-tail peptide (2.5 mg/ml lysis buffer). Aliquots of the eluate were analyzed by standard western immunoblot analysis following SDS-PAGE using monoclonal antibodies specific for alpha-or beta tubulin (Sigma). Only alpha-tubulin was detected in the C-tail peptide eluate using the ECL detection system (Amersham). Silver-staining of the eluate fraction proteins resolved in SDS-PAGE revealed that additional proteins were present.

In another series of experiments, Jijoye cells were lysed at 30.times.10.sup.6 /ml in a buffer (HL) containing 1% Triton X-100, 10 mM HEPES, pH7.5, 42 mM KCl, 5 mM MgCl.sub.2, 20 .mu.M NaF, 1 mM Na.sub.3 Vo.sub.4 and a protease inhibitor cocktail for 15 minutes at 4.degree. C. The lysates were then centrifuged at 45K rpm for 30 minutes in a TL100 BECKMAN table top ultracentrifuge. The high speed supernatant was then rotated for 2 hours at 4.degree. C. with 100 .mu.l of ethanolamine-blocked AFFIGEL 10 beads equilibrated in HL buffer. The beads were then spun onto SPIN-X 0.22 .mu.m centrifuge filter units (Costar; 6K rpm for 2 minutes). The flow through was split equally and each half rotated for 3 hours at 4.degree. C. with 100 .mu.l of either C-tail beads (13 mg/ml) or scrambled C-tail beads (12.2 mg/ml). The beads were then collected on SPINEX filters and sequentially eluted with 0.6 mg/ml of soluble scrambled C-tail peptide, 0.6 mg/ml and 1.2 mg/ml of soluble C-tail peptide. Beads were finally eluted with 200 .mu.l of two times concentrated SDS-sample prep buffer. Aliquots of each eluate were analysed by standard Western immunoblot analysis following SDS-PAGE using a monoclonal antibody specific to .alpha.-tubulin (Sigma) .alpha.-tubulin was bound only to the C-tail beads and could be eluted only with soluble C-tail peptide.

C. ICAM-R C-tail Affinity Precipitation of Purified Cytoskeletal Protein

Purified alpha-beta dimer tubulin, alpha-actinin, and vinculin were radioiodinated to specific activities of 1.9.times.10.sup.8 CPM/nMol, 0.3.times.10.sup.6 CPM/nMol, and 4.times.10.sup.7 CPM/nMol, respectively, with .sup.125 1 radionuclide (Dupont NEN). Affinity interaction of each of these radiolabeled proteins with ICAM-R C-tail beads was assayed using conditions (0.5% Tween-20, 50 mM Tris pH 7.5, 0.15M NaCl, 1 mM MgCl.sub.2, 1 mM CaCl.sub.2) previously employed to demonstrate specific ICAM-1 cytoplasmic tail association with alpha-actinin Carpen et al, J. Cell Biol., 118: 1223-1234 (1992)!. Briefly, 20 .mu.l of C-tail beads (19 mg peptide/ml resin) were incubated for 4 hours at 4.degree. C. with 60.times.10.sup.3 cpm of .sup.125 I-protein in the above-described Tween-20 buffer. After incubation, the beads were spun onto 0.45 .mu.m filtration units (Milliport; 6K rpm for 2 minutes), the flow through collected and the beads washed three times with 50 mM Tris pH 7.5, 1 mM MgCl.sub.2, 1 mM CaCl.sub.2, 0.15M NaCl and 1% Triton X-100. Finally, the beads were boiled in 60 .mu.l of 2 times concentrated SDS-sample prep buffer. Five .mu.l aliquots of column flow through and SDS elutions were counted in a Beckman gamma counter.

When input CPM of radiolabeled proteins were standardized, only tubulin exhibited specific binding to ICAM-R C-tail.

Example 24

ICAM-R cytoplasmic domain over-expression studies were performed to elucidate the functional consequences of ICAM-R C-tail molecular interactions occurring inside cells which undergo phenomena such as T cell receptor activation and cell-cell adhesion. The effect of overexpression of ICAM-R C-tail within a cell type expressing endogenous levels of wild type ICAM-R on its surface was tested. In contrast to effects on proliferation and IL-2 secretions observed when resting PBLs were pretreated with ICR-1.1 (Example 21) concomitant treatment of a lymphoblastoid cell line with ICR-1.1 and anti-CD3 antibody resulted in co-stimulation of IL-2 production. Additional studies (see Example 21) utilizing the same combination of monoclonal antibodies have demonstrated a co-stimulatory role (probably via a second messenger cascade) for ICAM-R in normal human PBL. The specific interaction of the ICAM-R C-tail with alpha tubulin (see Example 23) as determined by biochemical criteria, may serve to anchor the membrane phase with the cytoskeleton or to co-localize signaling molecules inside the cell with those transmembrane proteins involved in cell-cell interactions, as is known to occur in focal adhesive interactions with the extra cellular matrix. Thus, ICAM-R appears to mediate second messenger signaling, probably via C-tail interactions with kinases/phosphatases and/or the cytoskeleton.

A. Cytoplasmic Domain Constructs

The following DNA constructs were made using the PCR-based method called synthesis by overlap extension Horton et al., Gene, 77: 61-68 (1989)!: haWT.sub.453-518, haWT.sub.286-518, and haCT.DELTA..sub.286-484. The following nomenclature has been used for the constructs. "Ha" denotes an epitope tag sequence from the influenza hemagglutinin protein, which has been repeated three times in tandem to increase binding affinity with a commercially available monoclonal (12CA5, Boehringer Mannheim, Indianapolis, Ind.) antibody. "WT" refers to the native ICAM-R amino acid sequences. "CT.DELTA." refers to a deletion of the cytoplasmic tail. The numbers in subscript denote the starting and ending amino acids from ICAM-R included in the respective protein. Restriction enzyme cloning sites (HindIII and NotI) were engineered into the DNA constructs for subcloning into the expression vector pRC/CMV (Invitrogen Corp., San Diego, Calif.) under control of the cytomegalovirus immediate early enhancer/promoter. The pRC/CMV plasmid also contains the bacterial neomycin resistance gene from Tn5, thus allowing selection and maintenance of stable DNA integration into the cellular genome. The DNA constructs for haWT.sub.286-518 and haCT.DELTA..sub.286-484 plus other C-tail deletions were also subcloned into a second expression vector pMHneo Hahn et al., Gene 127: 267-268 (1993)! which drives expression from the Friend spleen focus-forming virus long terminal repeat. This vector provides for high levels of protein expression in the Jurkat T cell line.

The cytoplasmic tail of ICAM-R has homology with a motif that has been identified in the cytoplasmic domains of some T cell antigen receptor subunits, B cell membrane immunoglobulin antigen receptor subunits and mast cell Fc receptor subunits Reth, M., Nature, 338:383-384 (1989)!. The motif (see below) is known by various acronyms, including antigen receptor activation motif (ARAM), T cell activation motif (TAM) and antigen receptor homology 1 (ARH 1). Since its identification, this motif has been demonstrated to be necessary and sufficient for the transduction of signals from the membrane. ##STR1##

Shown is the consensus motif (top line) aligned with the ICAM-R cytoplasmic tail residues 483-501. Residues that are homologous with the consensus are underlined. The asterisk indicates the position of the stop codon in the truncated protein haCT.DELTA..sub.286-493.

To determine what region of the ICAM-R tail consensus motif that might be required for signaling in T cells, two cytoplasmic domain truncations have been created from haWT.sub.286-518. These are truncated at residues 505 (haCT.DELTA..sub.286-505) and 493 (haCT.DELTA..sub.286-493) thus dividing the cytoplasmic domain roughly into thirds. The haCT.DELTA..sub.286-505 protein leaves the consensus motif intact but removes the carboxyl terminal 13 residues of the native protein, while haCT.DELTA..sub.286-493 divides the motif in half, leaving only one of the tyrosine residues in the motif present. These truncations have been subcloned into the expression vector pMHneo for analysis in the Jurkat T leukemic cell line.

B. Jurkat Cell Expression

Jurkat cells were grown in RPMI 1640 medium, 10% FBS supplemented with penicillin/streptomycin and L-glutamine (RPMI complete) under standard cell culture conditions. For each electroporation condition, 5.times.10.sup.6 Jurkat cells at midphase of logarithmic growth were pelleted and rinsed in PBS-D, pelleted again and resuspended in PBS-D to a density of 1.times.10.sup.7 /ml. One half of one ml of suspended cells was transferred to a sterile cuvette (0.4 cm electrode gap) and 20 ug of linear plasmid DNA were added, mixed gently and incubated on ice for 10 minutes. A sufficient quantity of DNA was linearized by BglII digestion and prepared as follows. The linear DNA was extracted once with phenol/chloroform and precipitated with ethanol. After a 70% EtOH rinse and lyophilization, the DNA was resuspended in PBS-D to 1 mg/ml. The DNA/cell mix was then subjected to a 0.625 V/cm electric field with a pulse capacitance of 960 uFd. After a 10 minute incubation on ice, the cells were placed on 4 ml medium and allowed to replicate at 37.degree. C. to allow for integration of the plasmid DNA. After 48 hours, the entire cell population from each electroporation condition was plated out into 96 well plates such that 1-3.times.10.sup.4 cells were plated into each well in medium supplemented with 1.25 mg/ml G418 (Life Technologies, Bethesda, Md.). After approximately four weeks of drug selection, sufficient numbers of wells had cells grown from partial to complete confluency. These cells, and subsequent ones that grew, were routinely screened by FACS for positive staining with the HA epitope tag antibody, 12CA5. Positive wells were expanded and restained for surface expression of the following proteins and controls: CD3 (OKT3 antibody), ICAM-R (ICR-1.1 antibody), HA (12CA5 antibody), IgG 2a and 2b matched controls and FITC-conjugated secondary alone. Those wells that were positive for CD3, ICAM-R and HA and were negative for the isotype controls and the secondary antibody detecting reagent alone were expanded and FACS sorted to retain cells exhibiting the highest expression levels of the HA antigens.

C. Co-Stimulation and Cytokine Release

Wells of a 96 well plate (Immulon 4, Dynatech) were coated with 50 ul of a 1 ug/ml OKT3 monoclonal antibody in PBS-D for 16 hours at 4.degree. C. This dose of antibody alone provides minimal signalling for IL-2 release. The OKT3 treatment was removed and replaced with buffer alone or anti-ICAM-R (ICR-1.1) (10 ug/ml in PBS-D) and incubated at 37.degree. C. for at least 2 hours. Monoclonal antibody coating was done in replicates of two or more wells and pooled to provide sufficient quantities of conditioned media for ELISA and/or bioassay. Two hundred fifty thousand Jurkat cells or transfectants thereof, were placed into antibody coated wells or buffer coated negative control wells in 0.1 ml of RPMI complete medium. Following incubation of about 16-24 hrs at 37.degree. C. in a humidified atmosphere containing 5% CO.sub.2, the medium was transferred to a fresh 96 well round bottom plate, spun to pellet cells carried over and transferred to a fresh plate. Samples were frozen and stored at -70.degree. C. IL-2 ELISAs were performed using commercially available kits (Biosource Intl. Co., Camarille, Calif.) by making serial dilutions of the samples.

Expression of the haWT.sub.286-518 and haCT.DELTA..sub.286-484 had differing effects on the ability of the cells to be costimulated by ICR-1.1. Expression of haWT.sub.286-518 inhibited by about 60% the co-stimulatory response delivered via ICAM-R as compared to the response from cells expressing haCT.DELTA..sub.286-484 or vector transfectants alone. A conclusion of these experiments is that intracellular signalling and modulation of the IL-2 response through ICAM-R requires an intact cytoplasmic domain. Furthermore, this result implies that this system can be used to define the critical amino acid residues in ICAM-R by introducing the mutations outlined above and in Table 12.

D. Associate Protein p23

When Jurkat transfectants expressing haWT.sub.286-518 and haCT.DELTA..sub.286-484 were metabolically labeled with .sup.35 S! methionine, lysed and immunoprecipitated using anti-HA antisera, proteins of approximately 45 and 40 kD were visualized. Under reduced conditions, both of the transfectants yielded immune complexes which also contained an associated protein which migrated in SDS-PAGE at approximately 23 kD (p23). Under non-reduced conditions, the 23 kD protein band apparently shifted mobility and formed a complex with haWT.sub.286-518 or haCT.DELTA..sub.281-484 of 68 and 60 kD respectively. In addition, higher molecular weight complexes were visualized in haWT.sub.286-518 lysates of approximately 100 kD which may be homodimers of the protein or heteromeric complexes.

When unlabeled Jurkat transfectants were immunoprecipitated with anti-HA serum, analyzed by SDS-PAGE and transferred to nitrocellulose, only the full length haWT.sub.286-518 and haCT.DELTA..sub.286-484 proteins were visualized when the membrane was probed with anti HA monoclonal antibody. In addition, nothing was visualized when an antisera that recognizes the ICAM-R cytoplasmic tail was used to probe the membrane. Therefore, it is unlikely that p23 is a proteolytic degradation product of the haWT.sub.286-518 and haCT.DELTA..sub.286-484 proteins.

Example 25

The ICAM-R binding site for LFA-1 was localized to the amino terminal domain (domain 1) of ICAM-R. In addition, specific residues in domain 1 involved in the interaction of ICAM-R with LFA-1 were identified.

A. Production of ICAM-R Immunoglobulin Chimeras

The entire extracellular coding region of ICAM-R (nucleotides 1 to 1470 comprising the leader peptide sequence and all five Ig-like domains) isolated from pVZ147 (Example 4) by PCR was ligated in frame with a DNA fragment encoding the hinge and CH2 and CH3 coding regions of human IgG.sub.1 antibody. The resulting chimeric construct, ICAM-R/IgG, was expressed from the CMV promoter in the plasmid pcDNA1/Amp (Invitrogen). Variant ICAM-R/IgG fusion proteins were also produced from expression constructs in which either ICAM-R domain 1 or domain 3 encoding sequences had been deleted or in which mutations encoding amino acid substitutions had been introduced by site directed mutagenesis.

COS cells were transfected with the expression constructs by the DEAE-dextran method and the culture supernatant was collected. Fusion protein was purified from the culture supernatants using a Protein-A affinity (Prosep-A, Bioprocessing Ltd., England) column. A 1 ml bed volume Prosep-A column was equilibrated with PBS and the culture supernatant was loaded by gravity flow. The column was serially washed with 10 ml of 0.5M diethanolamine and 10 ml of 0.05M citric acid pH 5.0 to remove weakly binding proteins. Subsequently the ICAM-3/IgG fusion protein (or variant fusion protein) was eluted with 0.05M citric acid pH 3.0. The eluate was neutralized with one-sixth volume of 1.5M Tris pH 9.0. SDS-PAGE analysis of the protein revealed a band of about 225 kD in size which migrated to about 110 kD size under reducing condition indicating that the secreted fusion protein is a dimer.

The purified wild type ICAM-R/IgG chimera was quantitated by Micro BCA reagent (Pierce, Rockford, Ill.) according to the manufacturer's instructions. An estimate of the concentration of variant ICAM-R/IgG proteins was determined by coating plastic microtiter wells with serial dilutions of mutant ICAM-R chimeras or highly purified (>95%) ICAM-R/IgG followed by detection with horseradish peroxidase (HRP)-conjugated goat anti-human IgG (Fc-specific) antibody.

B. Binding of JY Cells to Wild Type ICAM-R/IgG and ICAM-R Domain Deletion/IgG Chimeras

The binding of the lymphoblastoid cell line JY, which expresses LFA-1, to wild type ICAM-R/IgG chimera was examined in the presence and absence of LFA-1 specific antibodies or antibodies specific for either domain 1 or 2 of ICAM-R. ICAM-R chimeras were diluted to 10 .mu.g/ml in 0.1M Na-carbonate/bicarbonate buffer pH 9.6 and used to coat triplicate wells (50 .mu.l/well) of an Immulon 496-well plate (Dynatech) overnight at 4.degree. C. Wells were washed three times in PBS and blocked with 1% BSA (in PBS) for 1 hour at 37.degree. C. JY cells were labeled with the fluorescent dye Calcein (Molecular Probes, Eugene, Oreg.) at 8 .mu.g/ml at 37.degree. C. in serum free RPMI for 20 minutes. Cells were washed with RPMI and resuspended in binding buffer (0.2% HSA in RPMI). Approximately 1.times.10.sup.5 cells were added to each well containing the binding buffer with or without antibody (20 .mu.g/ml of ICR-2.1, ICR-1.1, ICR-3.1, ICR-5.1, ICR-7.1, ICR-8.1, ICR4.2, ICR-6.2, ICR-9.2, TS1/22 or 60.3) so that the final volume was 0.35 ml. Plates were incubated at 37.degree. C. in the dark for 45 minutes and input fluorescence was quantitated with a fluorescence reader (Millipore) in a 96-well format. Unbound cells were removed by inverting the plate in 0.1% BSA in PBS for 20 minutes. Bound cells were quantitated by measuring the remaining fluorescence and presented as percent input minus background binding to wells coated with BSA alone.

Adhesion of the JY cells to the plate-bound ICAM-R/IgG chimera was predominantly LFA-1 dependent as shown by the ability of the LFA-1 specific monoclonal antibody TS1/22 to completely block binding. Six ICAM-R monoclonal antibodies also blocked the adhesion of JY cells to ICAM-R to varying extents. ICR-2.1 inhibited about 85% of JY cell binding. Five others ICR-1.1, 3.1, 5.1, 7.1 and 8.1 inhibited JY cell adhesion approximately 60 to 80%. Three monoclonal antibodies ICR4.2, 6.2 and 9.2 had no appreciable effect on the adhesion of JY cells to ICAM-R. These results suggest that LFA-1 interacts with ICAM-R domain 1 because monoclonal antibodies specific for ICAM-R domain 1 but not domain 2 block binding.

Binding of JY cells to domain deletion chimeras was also tested in the foregoing assay. Deletion of domain 1 resulted in complete loss of the ability of ICAM-R to promote LFA-1 mediated cell adhesion whereas deletion of domain 3 had essentially no effect. Collectively, from these results and the monoclonal antibody blocking results, it is apparent that ICAM-R interacts with LFA-1 primarily through its amino terminal domain 1.

C. Binding of JY Cells to Amino Acid Substituted ICAM-R/IgG Chimeras

ICAM-R/IgG chimeras with single or double amino acid substitutions in domain 1 were also tested in the binding assay described in Section B above. The mutation, E32K/AS (wherein mutation nomenclature is the same as in Example 14) resulted in a significant decrease in LFA-1 binding. In addition, the mutations E37T/AS and Q75I/AS nearly abolished adhesion of JY cells. To determine the contribution of individual residues we generated two additional mutations, E37/A and T38/A. The E37/A mutation completely abrogated adhesion of JY cells to ICAM-R. The T38/A mutation also resulted in a significant (70-80%) reduction in cell adhesion. These residues are displayed in a model of ICAM-R domain 1 along with their effect on LFA-1 binding in FIG. 16, wherein .beta. strands (wide arrows) were based on secondary structure predictions of ICAM-R and ICAM-1 as well as on alignment with the tenth type 3 repeat of fibronectin. Similar to the epitopes mapped for blocking ICAM-R specific monoclonal antibodies, residues implicated in LFA-1 binding locate to one face of domain 1 in this model.

Because carbohydrates are known to influence the ligand binding properties of several cell adhesion molecules, the effects of N-linked glycosylation of ICAM-R on LFA-1 binding were determined. In domain 1 of ICAM-R there are five potential N-linked glycosylation sites. Of these sites, N71 and N82 are closest to the residue Q75I which is shown above to be involved in LFA-1 binding. Replacement of N71 or N82 with glutamine did not significantly alter the adhesive properties of ICAM-R.

Moreover, disruption of ICAM-R domain 2 structure did not decrease LFA-1 binding. A domain 2 mutation L121/P resulted in significant loss of the epitopes for all three domain 2 antibodies. When the L121/P mutant chimera was tested for capacity to support cell adhesion, wild type levels of LFA-1 mediated binding were observed.

Example 26

Experiments were performed that show that another leukocyte integrin .alpha..sub.d /CD18 is also a ligand for ICAM-R. The leukocyte integrin .alpha..sub.d is described in co-pending, co-owned U.S. patent application Ser. No. 08/173,497 and in co-owned, concurrently filed U.S. patent application Ser. No. 08/173,497.

A. Human .alpha..sub.d binds to ICAM-R in a CD18-dependent fashion

In replicate assays, soluble ICAM-1, ICAM-R, or VCAM-1 IgG1 chimeric fusion proteins were immobilized on plastic and the ability of .alpha..sub.d /CD18 or LFA-1 transfected CHO cells (see co-owned, concurrently filed U.S. patent application Ser. No. 08/286,889, now U.S. Pat. No. 5,470,953 to bind the immobilized ligand was determined. Transfected cells were labeled internally with calcein, washed in binding buffer (RPMI with 1% BSA), and incubated in either buffer only (with or without 10 ng/ml PMA) or buffer with anti-CD18 at 10 .mu.g/ml. Transfected cells were added to 96-well Immulon 4 microtiter plates previously coated with soluble ICAM-1/Ig, ICAM-R/Ig or VCAM-1/Ig chimera, or bovine serum albumin (BSA) as a negative control. Wells were blocked with 1% BSA in PBS prior to addition of labeled cells. After washing the plates by immersion in PBS with 0.1% BSA for 20 minutes, total fluorescence remaining in each well was measured using a Cytofluor 2300 (Millipore, Milford, Mass.).

In experiments with immobilized ICAMs, .alpha..sub.d /CD18 co-transfectants consistently showed a 3-5 fold increase in binding to ICAM-R/IgG wells over BSA coated wells. The specificity and CD18-dependence of this binding was demonstrated by the inhibitory effects of anti-CD18 antibody TS1/18. The binding of cells transfected with LFA-1 to ICAM-1/IgG wells was comparable to the binding observed with BSA coated wells. LFA-1 transfected cells showed a 2-3 fold increase in binding to ICAM-1/IgG wells only following pretreatment with PMA. PMA treatment of .alpha..sub.d /CD18 transfectants did not affect binding to ICAM-1/IgG or ICAM-R/IgG wells. No detectable binding of .alpha..sub.d /CD18 transfectants to VCAM-1/IgG wells was observed.

Binding of .alpha..sub.d /CD18-transfected cells to soluble ICAM-1/IgG, ICAM-R/IgG, or VCAM-1/IgG chimeras was determined by flow cytometry. Approximately one million .alpha..sub.d /CD18-transfected CHO cells (grown in spinner flasks for higher expression) per measurement were suspended in 100 .mu.l binding buffer (RPMI and 1% BSA) with or without 10 .mu.g/ml anti-CD18 antibody. After a 20 minute incubation at room temperature, the cells were washed in binding buffer and soluble ICAM-1/IgG or ICAM-R/IgG chimera was added to a final concentration of 5 .mu.g/ml. Binding was allowed to proceed for 30 minute at 37.degree. C., after which the cells were washed three times and resuspended in 100 .mu.l binding buffer containing FITC-conjugated sheep anti-human IgG1 at a 1:100 dilution. After a 30 minute incubation, samples were washed three times and suspended in 200 .mu.l binding buffer for analysis with a Becton Dickinson FACScan.

Approximately 40-50% of the .alpha..sub.d /CD18 transfectants indicated binding to ICAM-R/IgG, but no binding to ICAM-1/IgG or VCAM-1/IgG chimeric proteins. Pretreatment of transfected cells with PMA has no effect on .alpha..sub.d /CD18 binding to either ICAM-1/IgG or ICAM-R/IgG. Binding by ICAM-R was reduced to background levels after treatment of .alpha..sub.d /CD18 transfectants with anti-CD18 antibody TS1/18. Consistent with the immobilized adhesion assay, PMA treatment of transfected cells had no affect on .alpha..sub.d /CD18 interaction with ICAM-R/Ig.

The collective data from these two binding assays illustrate that .alpha..sub.d /CD18 binds to ICAM-R and does so preferentially as compared to ICAM-1 and VCAM-1. The .alpha..sub.d /CD18 binding preference for ICAM-R over ICAM-1 is opposite that observed with LFA-1 and Mac-1. Thus modulation of .alpha..sub.d /CD18 binding may be expected to selectively affect normal and pathologic immune function where ICAM-R plays a prominent role. Moreover, results of similar assays, in which antibodies immunospecific for various extracellular domains of ICAM-R were tested for their ability to inhibit binding of ICAM-R to .alpha..sub.d /CD18 transfectants, indicated that .alpha..sub.d /CD18 and LFA-1 interact with different domains of ICAM-R.

The failure of LFA-1 to bind ICAM-1/IgG or ICAM-R/IgG in solution suggests that the affinity of binding between LFA-1 and ICAM-1 or ICAM-R is too low to permit binding in solution. Detection of .alpha..sub.d /CD18 binding to soluble ICAM-R, however, suggests an unusually high binding affinity.

Example 27

ICAM-R was also determined to interact with the .beta..sub.1 integrin VLA-4 (.alpha..sub.4 /.beta..sub.1). ICAM-R domain specific monoclonal antibodies and an ICAM-R domain deletion IgG chimera were used to map the VLA-4 binding to a site in ICAM-R domains 3-5. VLA-4 is expressed on all leukocytes with the exception of neutrophils. VLA-4 is also expressed on non-hematopoietic cells, including fibroblasts and neural crest cells. VLA-4 monoclonal antibodies inhibit the binding of leukocytes to cytokine activated endothelium Elices et al., Cell, 60:577-584 (1990)!, lymphocyte homotypic aggregation Pulido et al., J. Biol. Chem., 266:10241-10245 (1991)! and cytotoxic T cell mediated killing Clayberger et al., J. Immunol., 138:1510-1514 (1987)!. Adhesion of VLA-4 positive leukocytes to endothelium has been implicated in the process of atherogenesis Cybulsky et al., Science, 251:788-791, (1991)!, encephalomyelitis Baron et al., J. Exp. Med., 177:57-68, (1993)!, allogeneic graft rejection Pelletier et al., J. Immunol, 149:2473-2481 (1992)! and rheumatoid arthritis Morales-Ducret et al., J. Immunol., 149:1424-1431 (1992)!.

A. Binding of Jurkat Cells to Plate-Bound ICAM-R

shICAM-R (Example 9) was diluted to 10 .mu.g/ml in 50 mM carbonate buffer pH 9.6. Fifty .mu.l/well of this dilution was used to coat wells of a 96-well plate (Dynatech) by incubation for 16 hours at 4.degree. C. Prior to adhesion assay, the wells were emptied and blocked for 1 hour at 37.degree. C. with 1% BSA (Cohn Fraction V, Sigma) in PBS. Jurkat cells were labeled with 8 .mu.g/ml Calcein AM (Molecular Probes) in serum free culture medium for 20 minutes at 37.degree. C. Rinsed cells were treated with .alpha..sub.4 or .beta..sub.1 specific monoclonal antibody, and 1.3.times.10.sup.5 cells were distributed per well in RPMI, 0.2% HSA (Calbiochem). After a 50 minute incubation at 37.degree. C., total fluorescence per well was determined using a Cytofluor 2300 (Millipore), then the plate was inverted in 37.degree. C. PBS, 1% BSA for 30 minutes. The plate was removed and about 100 .mu.l/well aspirated using a 12 place manifold to remove cells in suspension near the mouth of the wells. The remaining fluorescence was determined and percent bound calculated from triplicate wells for each condition tested.

The .alpha..sub.4 specific monoclonal antibodies IC/A4.1 (ICOS Corp., Bothell, Wash.), 163H (anti-CD49d obtained from Dr. Michael Longenecker, University of Alberta, Edmonton, Canada) and HP2/1 (anti-CD49d, AMAC, Westbrook, Me.) and the .beta..sub.1 specific monoclonal antibodies K20 (anti-CD29, AMAC) and 3S3 (anti-CD29 obtained from Dr. John Wilkins, University of Manitoba, Winnepeg, Canada) blocked adhesion of Jurkat cells. Since Fab fragments of the .beta..sub.1 monoclonal 3S3 also blocked adhesion, inhibition does not appear to be a consequence of signals transduced following an antibody-induced antigen crosslinking. The ability of multiple .alpha..sub.4 and .beta..sub.1 specific monoclonal antibodies to block VLA-4 binding to ICAM-R suggests a direct interaction between VLA-4 and ICAM-R. In contrast, these .alpha..sub.4 and .beta..sub.1 antibodies did not block the CD18-dependent JY cell binding to ICAM-R.

B. Localization of the VLA-4 Binding Site

The VLA-4 binding site on ICAM-R was localized to domains 3-5 by comparing Jurkat and JY cell binding to wild type and domain deletion ICAM-R/IgG chimeras (Example 25) in the presence or absence of domain specific monoclonal antibodies to ICAM-R. The domain deletion ICAM-R/IgG chimera included ICAM-R domains 3, 4 and 5 and thus lacked the LFA-1 binding site within ICAM-R domain 1. Adhesion assays were performed according to the procedure described in Example 26.

The binding of Jurkat and JY cells to the ICAM-R/IgG chimeras was compared to the binding results for soluble ICAM-R. Jurkat cells bound at equivalent levels to all three forms of ICAM-R and was VLA-4 dependent. ICAM-R antibody ICR-19.3 specific for domain 3 completely blocked Jurkat cell binding to all forms of ICAM-R. In contrast, JY cells did not bind to domain deletion ICAM-R/IgG chimera lacking the LFA-1 binding site in domain 1. Monoclonal antibody specific for ICAM-R domain 1 did not block binding of Jurkat cells to wild-type ICAM-R.

The presence of three integrin binding sites (for LFA-1, .alpha..sub.d /CD18 and VLA-4) on ICAM-R may have several functional implications. Integrin binding to different sites on ICAM-R may transduce distinct intracellular signals. In addition, if LFA-1, .alpha..sub.d /CD18 and VLA-4 could bind simultaneously to ICAM-R, distinct or synergistic signals may be transduced through VLA-4 or LFA-1 in an opposing cell.

C. Inhibition of T Cell Activation by ICAM-R Binding to VLA-4

In previous examples, the effects on responding cells of engagement of ICAM-R by specific monoclonal antibodies was determined. In this section, the effect of engagement of the integrin receptors LFA-1 and VLA-4 by rICAM-R protein was measured.

T cell costimulation by recombinant ICAM-R and CD3 antibody in the presence of monoclonal antibodies which block ICAM-R adhesion to VLA-4 was measured (Example 20F). The monoclonal antibodies tested were specific for domain 3 of ICAM-R or the .alpha..sub.4 subunit of VLA-4. All of the blocking monoclonal antibodies dramatically enhanced the proliferative response to immobilized CD3 antibody. These results suggest that ICAM-R binding to VLA-4 inhibits T cell activation.

ICAM-R/IgG chimeras that preferentially bind LFA-1 or VLA-4 were also used to demonstrate the differential effects of ICAM-3 interaction with either receptor on T-cell activation. An ICAM-R/IgG variant E37T-Ig which binds VLA-4 but not LFA-1, and an ICAM-R/IgG variant D231H-Ig which demonstrated reduced VLA-4 binding capacity while retaining full LFA-1 binding capacity, were prepared by methods similar to those described in Example 25. Both variants were tested for effects on T-cell proliferation in response to CD3 mAb. In comparison to the HSA control, wild-type ICAM-3/IgG and variant D231H-Ig enhanced CD3 mAb driven T-cell proliferation. Similar levels of costimulation were induced by the LFA-1 binding D231H-Ig or wild-type ICAM-R/Ig chimera which binds both LFA-1 and VLA-4. This result supports the concept that LFA-1 represents the dominant functional T-cell integrin receptor for ICAM-R in this model. A contrasting effect on T-cell proliferation was observed in cultures containing the selective VLA-4 binding variant E37T-Ig. In comparison to the HSA control, E37T-Ig markedly inhibited T-cell proliferation in response to CD3 mAb.

Similar results were obtained from assays of IL-2 production in response to costimulation by anti-CD3 monoclonal antibody and shICAM-R or mutant ICAM-R/IgG chimeras. shICAM-3, wild type ICAM-R/IgG chimera and to a lesser extent D231H-Ig enhanced IL-2 production in comparison to the HSA control. In contrast the VLA-4 binding mutant E37T-Ig inhibited IL-2 production in response to CD3 mAb. These results demonstrate that ICAM-R elicits contrasting effects on T-cell function by interacting with distinct T-cell integrins.

Since rICAM-R in soluble form is of sufficient affinity to bind integrins such as .alpha..sub.d /CD18 (see Example 26), these results are significant since it is collectively implied that variant soluble ICAM-Rs may be utilized therapeutically.

Example 28

Circulating forms of ICAM-R (cICAM-R) were identified in human serum. Using a sandwich-ELISA with two monoclonal anti-ICAM-R antibodies (ICR-4.2 and ICR-8.1), cICAM-R was found in concentrations between 40 to 360 ng/ml in all of 112 healthy controls. An analysis of patient sera from ten different immune-mediated diseases revealed a distinct pattern of expression. Significantly elevated cICAM-R levels were found in rheumatoid arthritis, systemic lupus erythematosus, Guillain-Barre syndrome and multiple sclerosis, but not in type I diabetes, Grave's disease, chronic autoimmune thyroiditis, ulcerative colitis or Crohn's disease. cICAM-R levels were significantly higher in lupus patients with active compared to non-active disease. There was no uniform increase of cICAM-R levels in chromic inflammatory/autoimmune conditions. Serum levels of cICAM-R did not correlate with cICAM-1 concentrations in either control samples or in patients. The majority of patients had either elevated cICAM-R or cICAM-1 levels, but not both.

A circulating form of ICAM-R is thus present in human sera. cICAM-R expression is elevated in certain immune-mediated disease states but occurs independently of cICAM-1.

Example 29

Clearly, polynucleotides (e.g., DNA and RNA) encoding ICAM-R are useful not only in securing expression of ICAM-R and variant polypeptides; they may readily be employed to identify cells (especially cells involved in immunological processes) which express ICAM-R in a normal or activated state. Typical detection assays involving ICAM-R DNA include Northern blot hybridization, RNAse protection, and in situ hybridization cytological assays wherein the DNA or RNA (in suitably labelled, detectable form) hybridizes to RNA in the sample. ICAM-R encoding DNA (especially DNA encoding the first, fourth and fifth domains which have less homology to DNAs encoding ICAM-1 and ICAM-2 than the DNAs encoding domains 2 and 3) is expected to be useful in isolating genomic DNA encoding ICAM-R including genomic DNA specifying endogenous expression control DNA sequences for ICAM-R DNA. As previously noted, knowledge of polynucleotide sequences encoding ICAM-R and/or controlling expression of ICAM-R makes available a variety of antisense polynucleotides useful in regulating expression of ICAM-R.

The present invention makes available the production of ICAM-R polypeptides and variants thereof, especially including soluble fragments thereof, such as fragments comprising one or more of the five immunoglobulin-like domains of ICAM-R in glycosylated, non-glycosylated, or de-glycosylated forms. Pharmaceutical compositions including the protein products of the invention have therapeutic potential in the modulation of immune cell activation/proliferation, e.g., as competitive inhibitors or stimulatory agents of intercellular and intracellular ligand/receptor binding reactions involving ICAM-R. Such therapeutic potential is especially projected for "immunoadhesin" type recombinant hybrid fusion proteins containing, at their amino terminal, one or more domains of ICAM-R and, at their carboxy terminal, at least one constant domain of an immunoglobulin. Such hybrid fusion proteins are likely to be available in the form of homodimers wherein the Ig portion provides for longer serum half life and the ICAM-R portion has greater affinity for the ICAM-R binding partner than ICAM-R itself. Other multimeric forms of ICAM-R which may have enhanced avidity are also projected to have therapeutic potential.

Antibody substances and binding proteins, especially monospecific antibodies including monoclonal and polyclonal antibodies, are made readily available by the present invention through the use of immunogens comprising cells naturally expressing ICAM-R, recombinant host cells producing polypeptide products of the invention, the ICAM-R polypeptide products themselves, and polypeptide products of the invention bound to an ICAM-R specific antibody that stimulates cell-cell aggregation (i.e., polypeptide products that may be in a "high affinity" binding conformation). Such antibodies and other ICAM-R specific binding proteins can be employed for immunopurification of ICAM-R and variants and in pharmaceutical compositions for therapies premised on blocking and/or stimulating the ligand/receptor binding of ICAM-R and soluble fragments thereof. For use in pharmaceutical compositions, ICAM-R specific antibody and anti-idiotypic antibody substances may be humanized (e.g., CDR-grafted) by recombinant techniques well-known in the art. As illustrated in the foregoing examples, antibodies to distinct regions of ICAM-R may be employed to block adhesive interactions mediated by distinct integrins (e.g., LFA-1, VLA-4 and .alpha..sub.d /CD18). Also, antibodies specific for distinct regions of ICAM-R may be employed in ELISA systems involving immunological "sandwiches" for monitoring inflammatory processes characterized by increases in amounts of soluble ICAM-R polypeptides in body fluids such as serum. As outlined in Example 28, it is anticipated that such monitoring of ICAM-R levels as a surrogate marker of disease progression will be particularly useful in syndromes such as systemic lupus erythematosus, rheumatoid arthritis, multiple sclerosis and Guillan-Barre syndrome and may be useful as an early predictor of the onset of clinical episodes so that therapeutic drugs can be applied in a more timely fashion. As well, the onset of syndromes such as preterm labor, which may be mediated in part through an inflammatory process, may also be monitored by assessing levels of circulating ICAM-R in body fluids.

Inflammatory conditions which may be treated or monitored with ICAM-R related products include conditions resulting from a response of the non-specific immune system in a mammal (e.g., adult respiratory distress syndrome, multiple organ injury syndrome secondary to septicemia, multiple organ injury syndrome secondary to trauma, reperfusion injury of tissue, acute glomerulonephritis, reactive arthritis, dermatosis with acute inflammatory components, stroke, thermal injury, hemodialysis, leukapheresis, ulcerative colitis, Crohn's disease, necrotizing enterocolitis, granulocyte transfusion associated syndrome, atherosclerosis and cytokine-induced toxicity) and conditions resulting from a response of the specific immune system in a mammal (e.g., psoriasis, organ/tissue transplant rejection and autoimmune diseases including Raynaud's syndrome, autoimmune thyroiditis, EAE, multiple sclerosis, rheumatoid arthritis, diabetes, and lupus erythematosus). ICAM-R products of the invention may also be useful in monitoring and treating asthma, tumor growth and/or metastasis, and viral infection (e.g., HIV infection).

In particular, it is anticipated that disease processes in which T cell activation plays a central and essential triggering role will be impacted beneficially by products of the invention described herein. This inference is drawn in part from the findings outlined in Example 20 wherein monoclonal antibodies specific to ICAM-R and recombinant forms of ICAM-R protein were shown to modulate the response of human T lymphocytes to activating stimuli. Moreover, the therapeutic use of ICAM-R analogs incorporating specific amino acid substitutions (e.g. E37T or D231H) chosen to enhance or diminish their specific immunomodulatory properties (see Example 27) are expected to be useful in this regard. Since analogs of ICAM-R expressed as chimeric fusions with human immunoglobulin constant regions were shown to bind at least one integrin, it is anticipated that administration of these molecules in soluble form will be therapeutically useful. Specific examples of T cell dependent diseases for which ICAM-R related products are anticipated to have utility include but are not limited to asthma, psoriasis, diabetes, graft vs. host disease, tissue transplant rejection, and multiple sclerosis.

As illustrated in the foregoing examples, products of the invention can also be used to modulate the biological responses of monocytic cells and adhesion mediated by at least one integrin expressed selectively by macrophages, .alpha..sub.d. Thus, diseases wherein macrophages play a central generative role are also expected to benefit from products of the invention. For example, the formation of atherosclerotic plaques both as occurs progressively over time in humans and also as a consequence of solid organ or vessel engraftment (e.g., coronary bypass surgery) involves the activities of macrophages at both early and late stages of lesion formation. Foam cells, a specialized form of lipid laden macrophage found in such lesions, are thought to be a particularly important element of this process. As outlined in Example 20, engagement of ICAM-R on monocytes in the presence of oxidated phospholipid elicits secretion of the chemokine, MCP-1, which is potentially pro-atherosclerotic. Therefore, it is anticipated that products of the invention which modulate ICAM-R function could be utilized to block this process.

As outlined in Example 18, ICAM-R expression on vascular endothelial cells occurs selectively on neovascularizing sites found in solid tumors and benign angiomas. Therefore, it is anticipated that products of the invention such as monoclonal antibodies specific to ICAM-R may be used therapeutically either on their own or when conjugated to other moieties (e.g., toxins, radionuclides) to therapeutically target and/or detect the presence of such neovascularizing sites.

The foregoing illustrative examples relate to presently preferred embodiments of the invention and numerous modifications and variations thereof will be expected to occur to those skilled in the art. Thus only such limitations as appear in the appended claims should be placed upon the scope of the present invention.

Claims

We claim:

1. An antibody substance selected from the group consisting of monoclonal antibodies, monoclonal antibody fragments, single chain antibodies, chimeric antibodies, and CDR-grafted antibodies which specifically bind with ICAM-R and inhibit the interaction between ICAM-R and .alpha..sub.d /CD18.

Patent Citations (24)

Patent	Date	Inventor
US5081034	1992-01-01	Bevilacqua et al.
US5437958	1995-08-01	Gallatin et al.
EPX289949A2	1988-11-01
EPX314317A2	1989-05-01
EPX314863A2	1989-05-01
EPX362531A1	1990-04-01
EPX386906A1	1990-09-01
EPX387668A1	1990-09-01
EPX408859A2	1991-01-01
EPX468257A1	1992-01-01
WOXWO88/06592	1988-09-01
WOXWO89/02922	1989-04-01
WOXWO90/05539	1990-05-01
WOXWO90/05786	1990-05-01
WOXWO90/06953	1990-06-01
WOXWO90/13300	1990-11-01
WOXWO91/10683	1991-07-01
WOXWO91/16928	1991-11-01
WOXWO91/18010	1991-11-01
WOXWO91/18011	1991-11-01
WOXWO92/00751	1992-01-01
WOXWO92/04034	1992-03-01
WOXWO92/06119	1992-04-01
WOXWO92/22323	1992-12-01

Non-Patent Literature (36)

Ashkenazi et al., "Protection Against Endotoxic Shock by a Tumor Necrosis Factor Receptor Immunoadhesin", Proc. Natl. Acad. Sci. USA, 88:10535-10539 (1991).
Beatty et al., "Definition of a Common Leukocyte Cell-Surface Antigen (Lp95-150) Associated with Diverse Cell-Mediated Immune Functions.sup.1 ", J. Immunol., 131:2913-2918 (1983).
Cannizzaro et al., "The Human Gene Encoding Phosphatidylinositol-3 Kinase Associated p85.alpha. Is at Chromosome Region 5q12-13.sup.1 ", Cancer Res., 51:3818-3820 (1991).
Kapecchi, "Altering the Genome by Homologous Recombination", Science, 244:1288-1292 (1989).
Capon et al., "Designing CD4 Immunoadhesins for AIDS Therapy", Nature, 337:525-531 (1989).
Chen et al., "High Efficiency Transformation of Mammalian Cells by Plasmid DNA", Molecular and Cellular Biology, 7:2745-2752 (1987).
Corpet et al., "Multiple Sequence Alignment with Hierarchical Clustering", Nucleic Acids Res., 16(22):10881-10890 (1988).
de Fougerolles et al., "Characterization of ICAM-2 and Evidence for a Third Counter-Receptor for LFA-1", J. Exp. Med., 174:253-267 (1991).
de Fougerolles et al., "Intercellular Adhesion Molecule 3, a Third Adhesion Counter-Receptor for Lymphocyte Function Associated Molecule 1 on Resting Lymphocytes", J. Exp. Med., 175:185-190 (1992).
de Fougerolles et al., "Cloning and Expression of Intercellular Adhesion Molecule 3 Reveals Strong Homology to Other Immunoglobulin Family Counter-receptors for Lymphocyte Function-associated Antigen 1", J. Exp. Med., 177:1187-1192 (Apr. 1993).
Dustin et al., "Structure and Regulation of the Leukocyte Adhesion Receptor LFA-1 and Its Counterreceptors, ICAM-1 and ICAM-2", CSH Symp. Qual., 54:753-765 (1989).
Dustin et al., "T-Cell Receptor Cross-Linking Transiently Stimulated Adhesiveness Through LFA-1", Nature, 341:619-624 (1989).
Edwards, "Cell Adhesion Molecules as a Target for Therapy", Current Opinion in Therapeutic Patents, 1(11):1617-1630 (1991).
Fawcett et al., "Molecular Cloning of ICAM-3, a third Ligand for LFA-1, Constitutively Expressed on Resting Leukocytes", Nature, 360:481-484 (1992).
Geppert et al., "Association of Various T Cell-Surface Molecules With The Cytoskeleton", J. Immunol., 146:3298 (1990).
Goding, "Conjugation of Antibodies with Fluorochromes: Modifications to the Standard Methods", J. Immunol. Meth., 13:215 (1976).
Hadam, "N11 Cluster Report: CDw50", pp. 667-670 in Knapp et al., eds., Leukocyte Typing IV, Oxford, Oxford University Press (1989).
Hochuli et al., "Genetic Approach to Facilitate Purification of Recombinant Proteins With A Novel Metal Chelate Adsorbent", Bio/Technology, 6:1321-1325 (1988).
Hunkapiller et al., "The Growing Immunoglobulin Gene Superfamily", Nature, 323:15-16 (1986).
Janknecht et al., "Rapid and Efficient Purification of Native Histidine-Tagged Protein Expressed by Recombinant Vaccinia Virus", Proc. Natl. Acad. Sci., USA, 88:8972-8976 (1991).
Johnstone et al., p. 52 in Blackwell, Immunochemistry in Practice, Oxford Press (1982).
Juan et al., "CDw50 and ICAM-3: Two Names for the Same Molecule", Eur. J. Immunol., 23:1508-1512 (1993).
Knapp et al., "CD Antigens 1989", Blood, 74(4):1488-1450 (Sep. 1989).
Kunkel et al., "Rapid and Efficient Site-Specific Mutagenesis Without Phenotypic Selection", Proc. Natl. Acad. Sci. USA, 82:488-492 (1985).
Kyte et al., "A Simple Method for Displaying the Hydropathic Character of a Protein", J. Mol. Biol., 157:105-132 (1982).
Ledbetter et al., "Antibodies to Tp67 and Tp44 Augment and Sustain Proliferative Responses to Activated T Cells", J. Immunol., 135(4):2331-2336 (1985).
Lozano et al., "Effect of Protein Kinase C Activators on the Phosphorylation and the Surface Expression of the CDw50 Leukocyte Antigen", Eur. J. Biochem., 203:321-326 (Mar. 1992).
Lozano et al., "Isolation and Characterisation of a CDw50 Negative Jurkat T-Cell Line Variant", Leukemia Research, 17(1):9-16 (1993).
Newman et al., "PECAM-1 (CD31) Cloning and Relation to Adhesion Molecules of the Immunoglobulin Gene Superfamily", Science, 247:1219-1222 (1990).
Pierschbacher and Ruoslahti, "Influence of Stereochemistry of the Sequence Arg-Gly-Asp-Xaa on Binding Specificity in Cell Adhesion", J. Biol. Chem., 262(36):17294-17298 (1987).
Simmons et al., "ICAM, An Adhesion Ligand of LFA-1, Is Homologous to the Neural Cell Adhesion Molecule NCAM", Nature, 331:624-627 (1988).
Springer, "Adhesion Receptors of the Immune System", Nature, 346:425-434 (1990).
Staunton et al., "The Arrangement of the Immunoglobulin-like Domains of ICAM-1 and the Binding Sites for LFA-1 and Rhinovirus", Cell, 61:243-254 (1990).
Vazeux et al., "Cloning and Characterization of a New Intercellular Adhesion Molecule ICAM-R", Nature, 360:485-488 (1992).
Vilella et al., "Involvement of the CDw50 Molecule in Allorecognition", Tissue Antigens, 36:203-210 (1990).
Williams et al., "The Immunoglobulin Superfamily--Domains For Cell Surface Recognition .sup.1,2 ", Ann. Rev. Immunol., 6:381-405 (1988).

Abstract

Description