WO1994011400A1 - Peptides from human icam-2 and from human icam-1 and their analogs for use in therapy and diagnosis - Google Patents

Abstract

ICAM-1 and ICAM-2 peptides and analogs are disclosed which are useful in preventing aggregation or adhesion of leukocytes or lymphocytes to endothelial cells. Such peptides and analogs may also be used to enhance the activity of leukocytes to target cells. Prevention of aggregation or adhesion of leukocytes or lymphocytes aids in the prevention of undesired immune responses, such as transplant rejection.

Description

PEPTIDES FROM HUMAN ICAM-2 AND FROM HUMAN ICAM-1 AND THEIR ANALOGS FOR USE IN THERAPY AND DIAGNOSIS

This application is a continuation-in-part of United States Patent Application Serial Number 07/977,699, filed November 18, 1992.

FIELD OF THE INVENTION The present invention relates in general to ICAM molecules, and in particular to peptides from human ICAM-2, to peptides from human ICAM-1, and to analogs of both.

BACKGROUND Many leukocyte functions depend on the ability of the cells to specifically adhere to other leukocytes, to various target cells, or to homotypic leukocytes. Adherence involves the binding of heterodimeric leukocyte integrins, which consist of one of three specific α-chains (CDlla, CDllb, or

CDllc) and a common β-chain (CD18) . The biological significance of leukocytes is exemplified by Leukocyte Adhesion Deficiency Syndrome, in which alterations in CD18 give rise to diminished leukocyte binding which, in turn, results in various immunological disorders and the possibility of repeated infection.

The CD11/CD18 leukocyte integrins, which bind to intercellular adhesion molecules (hereinafter referred to as "ICAM") present on leukocytes and various other cells, are important in mediating diverse cell-cell interactions and leukocyte adhesion. The ICAM polypeptides belong to the immunoglobulin superfamily of adhesion ligands. ICAM-1 (CD54) contains five immunoglobulin domains. ICAM-2 contains two immunoglobulin domains which show a high degree of homology to the two NH₂- terminal domains of ICAM-1.

ICAM-1 is widely distributed, and its expression may be up-regulated by various cytokines. It acts as a receptor for the malaria parasite P . falciparum and for rhinoviruses. Soluble ICAM-1 may be useful in the treatment of asthma and rhinovirus infections. ICAM-1 binds both to CDlla/CD18 and CDllb/CD18.

In contrast to ICAM-1, ICAM-2 has been reported to bind only to CDlla/CD18, and not to CDllb/CD18 or to rhinovirus. ICAM-2 is reported to be cloned by a functional adhesion assay using COS- cells transfected with an endothelial cell expression library. ICAM-2 cDNA may be synthesized by polymerase chain reaction, and may be expressed in E . coli as a fusion protein. ICAM-2 may also be used to generate polyclonal antiserum. A monoclonal antibody raised against expressed ICAM-2, such as the monoclonal antibody designated 6D5, inhibits ICAM-2-dependent cell adhesion. The ICAM-2 protein has an apparent molecular weight of approximately

55,000, and may be heavily N-glycosylated. The 6D5 antibody reacts with the first immunoglobulin domain of ICAM-2.

In further contrast to ICAM-1, ICAM-2 appears to be constitutively expressed in most cells and tissues. However, in vivo, the reactivity of monoclonal antibody 6D5 with endothelium in malignant lymph nodes is higher than in nonmalignant lymph nodes. ICAM-2 has a costimulatory effect during the activation of human T cells.

It is important to identify binding sites on ICAMs in order to develop reagents that interfere with leukocyte adhesion. Several integrins bind to the arginine-glycine-aspartic acid (hereinafter referred to as "RGD") sequence in various proteins such as fibronectin, fibrinogen, and snake venom proteins, but there are also other recognition sequences. Those sequences are not found in human ICAM-1 or I CAN-2 , and are therefore excluded as binding sites.

Mutations may be made in ICAM-1, which then may be assayed for binding to CDlla/CD18 in transfected COS-cells. The most obvious effects on CDlla/CD18 binding are reported to result from mutations in the first domain.

The regulation of CDlla/CD18-ICAM interactions has been extensively studied using phorbol ester-stimulated aggregation as a model.

Phorbol esters induce a rapid and sustained increase in lymphocyte aggregation which is mediated by CDlla/CD18 interaction.^" Patarroyo, et al . , Scand . J . Immunol . , 22 : 171 (1985) . However, little is known regarding ICAM-2 interactions with the leukocyte integrins.

SUMMARY OF THE INVENTION The present invention relates generally to purified and isolated ICAM-1 and ICAM-2 peptides and fragments thereof (hereinafter collectively referred to as "ICAM peptides or fragments") . Specifically disclosed are fragments which stimulate aggregation of cells expressing at least one heterodimeric leukocyte integrin. In a preferred embodiment, ICAM peptides or fragments stimulate homotypic leukocyte aggregation. Also in a preferred embodiment of the invention, ICAM peptides or fragments which stimulate natural killer cell activity are disclosed. ICAM peptides or fragments according to the invention are shown to bind to purified CDlla/CD18 leukocyte integrin.

The present invention further provides a peptide having an amino acid sequence selected from the group consisting of: (SEQ ID NO: 4) ; and an amino acid sequence differing from SEQ ID NO: 4 by the addition or substitution of an amino acid which does not destroy integrin binding activity such as preferably, the amino acid sequence of SEQ ID NO: 12.

Another peptide according to the present invention has an amino acid sequence selected from the group consisting of: (SEQ ID NO: 5) ; and an amino acid sequence differing from SEQ ID NO: 5 by the addition or substitution of an amino acid which does not destroy integrin binding activity; preferably the amino acid sequence of SEQ ID NO: 13. Yet another peptide according to the present invention has an amino acid sequence selected from the group consisting of: (SEQ ID NO: 6) ; and an amino acid sequence differing from SEQ ID NO: 6 by the addition or substitution of an amino acid which does not destroy integrin binding activity; such as the amino acid sequence of SEQ ID NO: 14.

Still another peptide according to the present invention has an amino acid sequence selected from the group consisting of: (SEQ ID NO: 11) ; and an amino acid sequence differing from SEQ ID NO: 11 by the addition or substitution of an amino acid which does not destroy integrin binding activity; such as the amino acid sequence of SEQ ID NO: 17. Methods of treating a patient suspected of being infected by a virus comprising administration of an amount of an ICAM peptide or fragment sufficient to stimulate natural killer cell activity are disclosed. The present invention also provides methods for inhibiting transplant rejection including the steps of identifying a patient as being susceptible to transplant rejection, and administering to the patient a peptide according to the present invention in combination with a pharmaceutically acceptable diluent.

Analogs of SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO: 9, including their respective tyrosine addition analogs SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, and SEQ ID NO: 17 are also presented in the application.

In addition, the present invention provides substitution analogs of SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO: 9 in which a substituted amino acid is similar to the amino for which it was substituted in at least one of the properties of hydropathicity, charge, size of side groups, hydrogen bonding capacity, salt bridge formation capacity, optical rotation, disulfide bridge formation, and effect on peptide backbone conformation. Substitution analogs of SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6 according to the present invention minimally retain integrin binding properties as exhibited by SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6.

Addition analogs according to the present invention may be shorter in length than naturally- occurring ICAM-1 and ICAM-2 peptides and may not include integrin binding sites other than those in SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 and SEQ ID NO: 9. Preferably, addition analogs according to the present invention are shorter than human ICAM-2 (SEQ ID NO: 1) or human ICAM-1 (SEQ ID NO: 2), or have a different amino acid sequence therefrom. According to the present invention, a method for activating natural killer cells in a patient includes the steps of identifying a patient as benefitting from activation of natural killer cells, and administering to the patient a peptide according to the present invention in a pharmaceutically acceptable diluent.

A method for constructing a peptide having integrin binding activity according to the present invention includes the step of chemically synthesizing according to methods known to those skilled in the art a peptide according to the present invention. Alternately, a method for constructing a peptide having integrin binding activity includes the step of expressing from DNA, according to methods known to those skilled in the art, a peptide having the amino acid sequence of a peptide or fragment according to the present invention.

The present invention also provides kits for performing an assay according to the present invention. Such kits includes a peptide according to the present invention in combination with a signal for identifying integrin binding of the peptide.

The present invention also provides alternative forms of peptides according to the present invention, including anti-idiotypic antibodies produced according to methods known to those skilled in the art.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic depiction of the location of certain peptides in the amino acid sequence of human ICAM-2 (SEQ ID NO: 1) , human ICAM- 1 (SEQ ID NO: 2) and mouse ICAM-1 (SEQ ID NO: 3) ;

FIG. 2 is a reproduction of a gel exhibiting the results of polyacrylamide gel electrophoresis of purified CDlla/CD18;

FIG. 3A is a bar graph depicting the binding of ¹²⁵I-labeled PI (SEQ ID NO: 12) peptide to purified CDlla/CD18;

FIG. 3B is a bar graph depicting the relative binding of ¹²⁵I-labeled PI (SEQ ID NO: 12) peptide to purified CDlla/CD18 in the presence of competing peptides;

FIG. 3C is a graph depicting the relative binding of ¹²⁵I-P1 (SEQ ID NO: 12) peptide to purified CDlla/CD18 by comparison with the concentration of peptide;

FIG. 4A is a bar graph depicting inhibition endothelial cell binding to purified CDlla/CD18 by various peptides and antibodies; FIG. 4B is a graph depicting inhibition of endothelial cell binding to purified CDlla/CD18 versus concentration of peptides used; FIG. 5 is a bar graph depicting binding of NAD-20 cells to peptide-coated plates;

FIG. 6 is a three dimensional view of the first domain of ICAM-2; FIG. 7 is a graph depicting an aggregation score versus time for various peptides showing the effects on homotypic aggregation of Ramos cells;

FIG. 8 is a bar graph of an aggregation score for various monoclonal antibodies in a blocking assay;

FIG. 9 is a bar graph depicting the PI (SEQ ID NO: 4) and P7 (SEQ ID NO: 10) inhibition of the relative binding of ¹²⁵I-P1 (SEQ ID NO: 12);

Fig. 10A is a graph of the time-course of phorbol ester (squares)- and PI (circles) -induced aggregation of T-cells;

Fig. 10(B) is a graph of percent T-cell aggregation with differing amounts of PI;

Fig. 10(C) shows the percent T-cell aggregation when cells were pretreated with an inhibitor or control and then treated with PI;

Fig. 10(D) shows various microscopic fields of cells treated with P(Bu)₂ or PI;

Fig. 11 shows the extent of inhibition of aggregation induced by peptides according to the present invention with several compounds;

Fig. 12A shows the effect of peptides according to the invention and control peptides on natural killer cell binding; Fig. 12B shows the effect of peptides according to the invention and control peptides on natural killer cell cytotoxicity;

Figs. 13A-13D show the kinetics and dose response of Na alwa B lymphoma cells and U-937 myelomonocytic leukemia cells upon treatment with peptides according to the invention;

Fig. 14 shows the expression of CD11/CD18 integrins and ICAMs on Namalwa and U-937 cells; Fig. 15 shows the inhibition of Pl-induced adhesion in Namalwa and U-937 cells with different monoclonal antibodies;

Figs. 16A and 16B show inhibition of Pl- induced cell adhesion in Namalwa and U-937 cells with various monoclonal antibodies which were the same as those used in Figure 15;

Figs. 17A and 17B show the inhibition of Pl-induced aggregation in Namalwa and U-937 cells at 4 °C and with various chemical pre-treatments; Figs. 18A-18F show blockade of Pl-induced aggregation of Namalwa cells with protein kinase, phosphatase, and protein G inhibitors;

Fig. 19A and 19B show the inhibitory effects of staurosporine and okadaic acid, respectively, on PI (circles)- and p(Bu)₂-induced aggregation of T-cells;

Fig. 20 is a graph of the kinetics of PI (ICAM-2 peptide) -induced migration of natural killer cells; and Fig. 21 is a graph of the dose-dependent

PI (ICAJM-2 peptide) promotion of natural killer cell migration in a Boyden chamber. DETAILED DESCRIPTION The abbreviations used in the present application include CD, cluster of differentiation antigens; CDlla/CD18, lymphocyte function-associated antigen-1 (also abbreviated "LFA-1") CDllb/CD18, complement type 3 receptor (also abbreviated as Mac- 1) ; ICAM-1, intercellular adhesion molecule-1; ICAM-2, intercellular adhesion molecule-2 ; HAT, hypoxanthine-aminopterin-thymidine; BSA, bovine serum albumin; TNF- , tumor necrosis factor-α; PI, peptide 1, etc.; and VLA-4 , very late antigen-4.

The human ICAM-2 and ICAM-1 NH₂-terminal immunoglobulin domains are 35% identical and are related to murine ICAM-1, which also binds to human CDlla/CD18. Similar motifs in the CDlla/CD18 binding regions of all three ICAMs are therefore expected. FIG. 1 shows an alignment of the amino acid sequences from the first domains of human ICAM-2 (SEQ ID NO: 1), ICAM-1 (SEQ ID NO: 2), and mouse ICAM-1 (SEQ ID NO: 3) which show the highest homology. According to the present invention, a peptide has been defined from this part of the first domain of ICAM-2. This peptide specifically binds to CDlla/CD18. Numerous leukocyte functions depend on adhesive intercellular interactions. The leukocyte- specific integrins CDlla/CD18 (LFA-1) [Springer, Nature, 346, 425-434 (1990)] and CDllb/CD18 (Mac-1) , which bind to the intercellular adhesion molecules ICAM-1 and ICAM-2, play a key role in adhesion.

Little is known about the details of molecular binding events which occur in adhesion.

The present invention provides a defined peptide region from the first immunoglobulin domain of human ICAM-2 (SEQ ID NO: 4) and a peptide region from human ICAM-1 (P6, SEQ ID NO: 9) , which peptide regions are specifically involved in binding to CDlla/CD18. Synthetic peptides from these portions of ICAM-2 and ICAM-1 bind to purified CDlla/CD18 and inhibit the adhesion of endothelial cells to the CDlla/CD18 integrin. Leukocytes also bind to plastic coated with the aforementioned peptides. Furthermore, such peptides induce leukocyte adherence to target cells.

Controlling leukocyte functions by inhibiting leukocyte target cell binding with peptides, such as those disclosed in SEQ ID NO: 4 and SEQ ID NO: 9 or their analogs, or by inducing leukocyte activation with peptides according to SEQ ID NO: 4 and SEQ ID NO: 9 or their analogs, is useful for therapeutic application in conditions of undesired leukocyte target cell binding (e.g., to inhibit transplant rejection) , and is also useful in the treatment of infection and cancer or in cases where activation of leukocytes is desired (e . g . , activation of natural killer cells) .

Peptides according to the present invention may be administered to patients in order to interfere with leukocyte binding to endothelial cells. In this way, inflammation and leukocyte migration into tissues are prevented.

Due to the high specificity of peptides according to the invention, such as those disclosed in SEQ ID NO: 4 and SEQ ID NO: 9 or their analogs, these peptides may also be used to detect, inter alia , the CDlla/CD18 receptor for diagnostic purposes. EXAMPLE 1 Computer Modelling of Domain 1 of ICAM-2

In order to determine structural information regarding the PI fragment of ICAM-2, a domain model was constructed and energy minimized using the Insightll package (BIOSYM Technologies Inc. , San Diego, CA) . Known three-dimensional structures of several im unoglobulins, CD4, and CD8, and a model of ICAM-1 [Giranda et al . , Proteins , 7, 227-233 (1990)] were used as templates.

A. Model of the First Immunoglobulin Domain of ICAM-2

The external portion of ICAM-2 is relatively simple, containing only two immunoglobulin domains. Therefore, ICAM-2 offers some advantages over ICAM-1 as a model protein. In both ICAM-1 and ICAM-2, domain 1 is most important in binding to CDlla/CD18. Because not only human ICAM-1 and ICAM-2, but also murine ICAM-1, binds to human integrin, the binding regions of the first immunoglobulin domains in each of those proteins may be closely related.

Computer modeling of the first immunoglobulin-like domain of ICAM-2 , as illustrated in FIG. 6, reveals a close resemblance to the first immunoglobulin domain of ICAM-1. [Berendt et al., Cell , 68, 71-81 (1992); Giranda et al., Proteins, 7, 227-233 (1990)] The main differences are that the ICAM-2 domain contains a larger loop between beta strands A and B has a cis-peptide bond in Pro-12 which is not present in ICAM-1. In FIG. 6 the location of PI (SEQ ID NO: 4) is indicated in dark color. The NH₂-terminal of the peptide beginning at Gly-21 is well exposed, whereas most of the B and C strands are deeply embedded in the structure. The COOH-terminal is exposed.

EXAMPLE 2 Construction of Synthetic Peptides and Fragments

Synthetic peptides were prepared by solid- phase synthesis on an Applied Biosystems model 430A peptide synthesizer (Applied Biosystems, Foster City, California) , using t-BOC-chemistry. Peptides were reduced with dithiothreitol for two hours at room temperature before purification by HPLC to greater than 98% purity using a C18 μ Bondapak column with a linear gradient of 0-70% acetonitrile in 0.1% trifluoroacetic acid. The structures of the synthetic peptides so made were confirmed by amino acid analysis and plasma desorption mass spectrometry.

The synthesized peptides comprised: a 22- amino acid peptide (residues 21 to 42) from the first domain of ICAM-2 , as shown in FIG. 1 and designated PI (SEQ ID NO: 4) ; sub-peptides of PI (SEQ ID NO: 4) which represent sequential deletions of the NH₂-terminal amino acid(s) of PI to form P2 through P4 [P2 (SEQ ID NO: 5), P3 (SEQ ID NO: 6), and P4 (SEQ ID NO: 7)]; P5 (SEQ ID NO: 8)]; P5a (SEQ ID NO: 11); P5b (SEQ ID NO:19) ; the peptide corresponding to PI from human ICAM-1 (P6, SEQ ID NO: 9) ; and a control peptide, P7 (SEQ ID NO: 10) , which contains the same amino acids as PI (SEQ ID NO: 4) in random order with the exception of two conserved cysteine residues. A COOH-terminal tyrosine was added to all peptides to facilitate ¹²⁵I-labeling as follows: Pl-Tyr (SEQ ID NO: 12) ; P2-Tyr (SEQ ID NO: 13); P3-Tyr (SEQ ID NO: 14); P4- Tyr (SEQ ID NO: 15); P5-Tyr (SEQ ID NO: 16); P6-Tyr (SEQ ID NO: 17); and P7-Tyr (SEQ ID NO: 18) .

FIG. 1 shows an alignment of homologous sequences from the first immunoglobulin domains in human and mouse ICAMs. The sequences corresponding to ICAM-2 peptides P1-P5 (SEQ ID NO; 4 - SEQ ID NO: 8) and the ICAM-1 peptide P6 are indicated by lines beneath the sequences. The boundary between the two immunoglobulin domains (DI and D2) is also shown.

EXAMPLE 3

Binding Characteristics of PI

A. Binding of ¹²⁵I-labeled PI (SEQ ID NO: 12) Peptide to Purified CDlla/CD18

CDlla/CD18 and CDllb/CD18 were purified from human buffy coat cell lysates by immunoaffinity chromatography on TS2/4-monoclonal antibody 60.1-

Sepharose®, respectively, and eluted at pH 11.5 in the presence of 2 mM MgCl₂ and 1% octylglucoside

[Dustin et al., Nature, 341, 619-624 (1989)]. The purity of each was checked by polyacrylamide gel electrophoresis in the presence of SDS [Laemmli, Nature, 227 , 680-685 (1970)]. The detergent- solubilized protein was diluted with buffer (25 mM Tris, pH 8.0, 150 mM NaCl, 2 mM MgCl₂) and attached to ELISA plates (Nunc, Roskilde, Denmark) , 5 μg in 50 μl/well, by overnight incubation at 4°C. The wells were blocked with 1% BSA for one hour at room temperature. ¹²⁵I-labeled PI (SEQ ID NO: 12), labeled by the chloramine T method [Greenwood et al., Biochem . J . , 89, 114-123 (1963)] was added to each well in 50 μl of medium (0.15 M NaCl, 0.01 M sodium phosphate, pH 7.4, PBS, 2 mM MgCl_2; 0.5% BSA, 0.02% NaN₃) with or without non-radioactive peptides, and incubated at 37°C for one hour. After washing, the attached ¹²⁵I-P1 was solubilized with 1% SDS and counted.

B. Binding of Peptides to CDlla/CD18

The purified CD11/CD18 preparation was analyzed by polyacrylamide gel electrophoresis in the presence of SDS. The preparation contained the expected CDlla and CD18 polypeptides, and no impurities were observed.

FIG. 2 illustrates the results obtained from polyacrylamide gel electrophoresis of purified CDlla/CD18. Purified CDlla/DC18 was run on an 8 percent polyacrylamide gel in the presence of SDS and stained with Coomassie blue.

Peptide PI (SEQ ID NO: 4; derived from ICAM-2) bound to plastic which had been coated with purified CDlla/CD18 almost completely inhibited the binding of ¹²⁵I-P1 (SEQ ID NO: 12) , as illustrated in (FIG. 3A) .

FIG. 3 illustrates binding of ¹²⁵I-labeled PI (SEQ ID NO: 12) peptide to purified CDlla/CD18 in the absence or presence of competing peptides. ¹²⁵ι- labeled PI (SEQ ID NO: 12) was incubated in

CDlla/CD18-coated wells together with 100 μM of different non-radioactive peptides (with results as shown in FIG. 3A and FIG. 3B) and various concentrations of non-radioactive PI (SEQ ID NO: 4; shown by X's), P6 (SEQ ID NO: 9; shown by closed circles) or P7 (SEQ ID NO: 10; shown by closed squares) peptides. The differences in binding in the absence of competing peptides and in their presence were statistically significant as shown in FIG. 3B, wherein **=P<0.01, ***P<0.001. The standard deviations in FIG. 3C were less than 12%. The ability of various Pi subpeptides disclosed above to bind to CDlla/CD18 was determined using ¹²⁵I-labeled PI. The peptide P2 (SEQ ID NO: 5) , lacking the NH₂-terminal glycine of PI (SEQ ID NO: 4) , was about 25 times less efficient than PI at blocking ¹²⁵I-labeled PI binding to CDlla/CD18; and P3 (SEQ ID NO: 6) , lacking the next serine was even less active. The ICAM-1 peptide, P6 (SEQ ID NO: 9) , partially blocked binding. Peptides P4 (SEQ ID NO: 7), P5 (SEQ ID NO: 8), and the control peptide P7 (SEQ ID NO: 10) were not inhibitory (FIG. 3B) . The binding of ¹²⁵I-P1 (SEQ ID NO: 12) was inhibited in a concentration-dependent manner by nonradioactive PI (SEQ ID NO 4) , less so by P6 (SEQ ID NO: 9) , and not at all by the control peptide P7 (SEQ ID NO: 10) (FIG. 3C) . A 50% inhibition was obtained with about 5 μM of PI (SEQ ID NO: 4) peptide. The results suggest that both the NH₂- terminal and the COOH-terminal of PI are important for activity. ICAM-2 peptide P5 (SEQ ID NO: 8) , which ended at Pro-34, had no CDlla/CD18 binding activity. Peptide P2 (SEQ ID NO: 5) , lacking the NH₂-terminal glycine of PI (SEQ ID NO: 4) , was about 25 times less active than PI (SEQ ID NO: 4) in binding to CDlla/CD18, and further deletion of the NH₂-terminal serine from P2 further lowered activity. Therefore, it is presumed that those NH₂-terminal amino acids are important for activity. Surprisingly, when the PI peptide (SEQ ID NO: 4) was elongated with a lysine at its NH₂-terminal, which is found in the natural sequence of ICAM-2, the peptide was not active. The lysine may change the conformation of the peptide by forming a salt bridge to one of the three glutamic acids in the peptide.

C. Binding of Peptides to CDllb/CD18

CDllb/DClδ was purified from human buffy coat cells by immunoaffinity chromatography using anti-CDllb mAb. An adhesion assay was performed in a way similar to that described in Example 3.

Labeled PI (SEQ ID NO: 12) also binds to purified CDllb/CDlδ, unlabeled PI (SEQ ID NO: 4) efficiently blocks this binding, but the control peptide, P7 (SEQ ID NO: 10) , does not (as shown in FIG. 9) . FIG. 9 illustrates a comparison of the relative binding percent of ¹²⁵I-P1 (SEQ ID NO: 12) with the background % binding of BSA and with the degree of inhibition of binding by unlabeled PI (SEQ ID NO: 4) or unlabeled P7 (SEQ ID NO: 10) . Degree of inhibition is shown as lower ¹²⁵I-P1 (SEQ ID NO: 12) binding. The antibody 60.1, which reacts with CDllb/CD18, also blocks binding of labeled PI (SEQ ID NO : 12) . This finding indicates that ICAM-2 may also bind to CDllb/CDlδ and not only to CDlla/CDlδ. Accordingly PI (SEQ ID NO: 4) may interfere with CDllb/CDlδ-dependent functions.

The ICAM-1 peptide P6 showed less activity in all test systems, and the binding capacity of ¹²⁵I-labeled P6 (SEQ ID NO: 17) to CDlla/CD18 was less than 10% as compared to that of labeled PI (SEQ ID NO: 12; data not shown). P6 (SEQ ID NO: 9) may bind to the same region in CDlla/CDlδ as does PI (SEQ ID NO: 4) , but the affinity is lower. Small adjustments in the peptide sequence may greatly influence its activity as is the case for PI (SEQ ID NO: 4) .

Seth et al. (Seth et al. , FEBS Lett . , 282 , 193-196 (1991)) report an ICAM-2 peptide from the second Ig-domain which, when coated on plastic, was able to bind B lymphoblastoid cells. Some inhibition of binding is reported to be obtained both with CDlδ and very late antigen-4 (VLA-4) antibodies. Fecondo et al. [Fecondo et al., Proc . Natl . Acad . Sci . USA, 88, 2δ79-2δ82 (1991)] report a peptide from the fourth domain of ICAM-1, which is reported to inhibit T-cell cytotoxicity toward K562 cells, and the spontaneous aggregation of Raji Burkitt's lymphoma B cells. These reports are contrary to other reports [ e . g . , Staunton et al., Cell , 61, 243-254 (1990)], that domains D3-D5 in ICAM-1 can be deleted without loss of CDlla/CDlδ binding activity. On the other hand, it is reported that peptides from ICAM-1, spanning residues 1-20, 26-50, and 132-146 partially inhibited the binding of Molt-4 lymphoblastic cells to endothelial cells. [Ross et al., J. Biol . Chem . , 267, δ537-δ543 (1992)] The first two peptides (1-20 and 26-50) partially overlap with the peptide P6 (SEQ ID NO: 9) . The most efficient peptide inhibited 29% of the binding, but the contribution of ICAM-1 in the adhesion assay was estimated to be only 35%. Several regions in ICAM-1 may show some binding activity.

Leukocyte functions are modified by peptides according to the present invention. These function may be used to determine the binding sites on CDlla/CDlδ and CDllb/CDlδ, as is reported for the platelet IIB/IIIa integrin [D'Souza et al., J . Biol . Chem . , 263, 3440-3446 (1990); D'Souza et al. , Nature, 350, 66-6δ (1991)].

EXAMPLE 4

Effects of PI on Endothelial Cell Adhesion

A. Cell Lines and Cell Culture

The cell line Eahy926 is a hybrid between vascular endothelial and carcinoma cells and expresses endothelial cell markers [Eahy 926 cells were obtained from Dr. C. Edgell and are described in Edgell et al., Proc . Natl . Acad . Sci . USA, 80, 3734-3737 (1983)]. Cells of this cell line were cultured in Dulbecco's HAT medium containing 10% fetal calf serum. Eahy926 was induced by incubating the cells with 10 ng/ml of TNF-α (Boehringer Mannheim, Mannheim, Germany) overnight [Nortamo et al., Eur . J . Immunol . , 21, 2629-2632 (1991)]. The Epstein-Barr virus-transformed B cell line, NAD-20, and Burkitt's lymphoma cell line, Ramos, were maintained in RPMI 1640 medium containing 20% fetal calf serum.

B. Inhibition of Endothelial Cell Adhesion Purified CDlla/CDlδ (50 μg in 1 ml/plate) was coated on 35 mm bacteriological dishes (Greiner GmbH, Kre smϋnster, Austria) and the plates were saturated with 1% BSA as described in Example 3 above. Control plates were treated with BSA only. TNF-α induced Eahy926 cells were removed from tissue culture flasks with 5 mM EDTA in PBS, washed and resuspended in medium (HAT medium 40 mM HEPES, pH 7.2, 2 mM MgCl₂, 5% fetal calf serum). Cells (7 x 10⁵) in 1 ml were added to each plate in the presence of peptides as indicated in FIG. 4A and FIG. 4B, and were incubated for one hour at room temperature. Unbound cells were removed by washing three times with the same medium. The binding was quantitated by counting bound cells/microscopic field, with 15 fields for each plate.

C. Antibodies

Samples of monoclonal antibodies 7E4 (anti-CDlδ) [Nortamo et al., Scand. J . Immunol . , 28, 537-546 (1988)], LB-2 (anti-ICAM-1) [obtained from Dr. E.A. Clark and described in Clark et al., Hum . Immunol . , 16, 100-113 (1986)], TS2/4 (anti-CDlla) [obtained from Dr. T.A. Springer and described in Sanchez-Madrid et al., Proc . Natl . Acad . Sci . USA, 79, 7489-7493 (1982)], H12 (obtained from Dr. M. Patarroyo) , 60.1 (obtained from Dr. P. Beatty) , IC2/2 [obtained from Dr. T.A. Springer and described in deFougerolles et al., J . Exp . Med . , 174 , 253-267 (1991)], and the irrelevant antibody 84-3C1 (anti- CD43) [obtained from D. J. Vives and described in Borche et al., Eur . J . Immunol . , 17, 1523-1526 (19δ7) ] were prepared using precipitation with lδl mg/ml of Na₂S0_,.

D. Inhibition of Endothelial Cell Binding to

CDlla/CDlδ by Peptides

FIG. 4A and FIG. 4B illustrate inhibition of endothelial cell binding to purified CDlla/CDlδ by peptides according to the present invention. One hundred μg each of ICAM-2 and ICAM-1 peptides PI

(SEQ ID NO: 4) and P6 (SEQ ID NO: 9) , control peptide P7 (SEQ ID NO: 10) , and monoclonal antibodies 7E4 (CDlδ) , and 64-3C1 (DC43) in (FIG. 4A) , and indicated concentrations of PI (SEQ ID NO: 4) (X's), P6 (SEQ ID NO: 9) (closed circles), and P7 (SEQ ID NO: 10) (squares) (in FIG. 4B) were used to block the adhesion of Eahy926 cells to purified CDlla/CDlδ. The cells bound as a percentage of maximum is provided in the Figures. The inhibitory effects of peptides according to the present invention were significant, as shown in FIG 4A wherein *** = p < 0.001.

Peptide PI (SEQ ID NO: 4) efficiently inhibited the binding of TNF-α-induced Eahy926 cells to purified CDlla/CDlδ (FIG. 4A) . The ICAM-1 derived peptide P6 (SEQ ID NO: 9) showed a smaller inhibition. The binding was efficiently blocked by the CDlδ antibody 7E4. The inhibition was concentration-dependent as shown in FIG. 4B. A 50% inhibition was obtained with about 18 μM (43 μg/ml) of peptide PI (SEQ ID NO 4) . The control peptide P7 (SEQ ID NO: 10) had no apparent effect on binding.

The binding of endothelial cells to CDlla/CDlδ was efficiently inhibited by the PI (SEQ ID NO: 4) peptide. Induced Eahy926 endothelial cells contain more ICAM-1 than ICAM-2. [Nortamo et al., Eur . J . Immunol . , 21, 2629-2632 (1991)], and therefore, the effect of PI (SEQ ID NO: 4) may be partially due to interference with the binding of ICAM-1 to CDlla/CDlδ. ICAM-2 and ICAM-1 may bind to the same or closely spaced sites in CDlla/CDlδ. E. Binding of NAD-20 Cells to Peptide-Coated Plates

Peptides PI (SEQ ID NO: 4), P6 (SEQ ID NO:

9) , and P7 (SEQ ID NO: 10) were covalently attached to amino-surface plates (COSTAR Europe Ltd.

Badhoevedorp, the Netherlands) by incubating 10 μg of peptide with 0.4 mg l-ethyl-3-(3- dimethylaminopropyl) carbodiimide hydrochloride

(Pierce Europe BV, Oud-Beijerland, The Netherlands) in 100 μl of dilute HCl, pH 4.0 in each well for two hours at 22°C. The wells were then saturated with 1% nonfat dry milk solution by incubating for one hour at room temperature. The amount of peptide attached was about 5 ng/well for each peptide as determined by competition with ¹²⁵I-labeled peptides. NAD-20 cells (δ x 10^A) in 100 μl (RPMI1640 40 mM HEPES, pH 7.2 2 mM MgCl₂, 5% fetal calf serum) were incubated for one hour at 22°C in the presence of 5 μg LB-2 antibody to avoid spontaneous intercellular aggregation. After washing three times, the bound cells in each microscopic field were counted. Four fields were evaluated for each well.

F. Binding of Leukocytes to Peptides Coated on Plastic NAD-20 B lymphocytes bound efficiently to peptide PI (SEQ ID NO: 4) but not to the control peptide (P7, SEQ ID NO: 10) coated on plastic (FIG. 5) . The binding was not blocked by the monoclonal antibody 7E4. PI (SEQ ID NO: 4) does not block the spontaneous aggregation of NAD-20 cells (not shown) . This may be because it does not block binding between CDllb/CDlδ and the third immunoglobulin domain of ICAM-1 [Diamond et al., Cell , 65, 961-971 (1991)]. ICAM-3 [deFougerolles et al., J . Exp. Med . , 175, 165-190 (1992)], or other unknown ICAMs may have a different binding specificity with which PI may not interfere.

FIG. 5 illustrates binding of NAD-20 cells to peptide-coated plates. The binding of NAD-20 cells to PI (SEQ ID NO: 4) was significantly (p < 0.01) higher than binding to the control peptide P7 (SEQ ID NO: 10) . The monoclonal antibody 73R had no effect on the adhesion of cells to PI (SEQ ID NO: 4) . The number of cells bound/microscopic field is given.

EXAMPLE 5 Induction of Leukocyte Aggregation

A. Inducing Homotypic Leukocyte Aggregation with Peptides of the Invention

Peptides PI (SEQ ID NO: 4) , P6 (SEQ ID

NO: 9) , and P7 (SEQ ID NO: 10) were covalently attached to amino-surface plates (COSTAR Europe Ltd. Badhoevedorp, the Netherlands) by incubating 10 μg of peptide with 100 μl Qf 100 mM l-ethyl-3-(3- dimethylaminopropyl) carbodiimide hydrochloride, 5 mM N-hydroxysulphosuccimnimide (Pierce Europe BV, Oud-Beijerland, the Netherlands) in H₂0 overnight at 4°C. The wells were then saturated with 1% nonfat dry milk solution by incubating for one hour at room temperature. The amount of peptide attached was about 5 ng/well for each peptide as determined with ¹²⁵I-labeled peptides. Ramos cells (1 x 10⁶) in medium 100 μl RPMI 1640, 40 mM HEPES, pH 7.2-2 mM MgCl₂, 5% fetal calf serum were incubated at 37°C in the absence or presence of specific antibodies. The stage of aggregation was evaluated by counting the percentage of aggregated cells and showing the data by using aggregation scores as follows: 0, no aggregated cells; 1, less than 5% of cells aggregated; 2, 5-10% of cells aggregated; 3, less than 30% of cells aggregated; 4, 30-50% of cells aggregated; 50-70% of cells aggregated.

B. Induction of Homotypic Leukocyte Aggregation Peptides PI (SEQ ID NO: 4) and P6 (SEQ ID

NO: 9) , but not the control peptide P7 (SEQ ID NO: 10) , efficiently induced the aggregation of Ramos B lymphocytes, as shown in the kinetic curves in FIG. 7 wherein: PI (SEQ ID NO: 4) is identified by an unshaded square, P6 (SEQ ID NO: 9) is identified by a shaded circle, and P7 (SEQ ID NO: 10) is identified by a shaded square. The aggregation induced by the peptide PI (SEQ ID NO: 4) is totally blocked with monoclonal antibody 7E4 and partially with monoclonal antibodies H12, 60.1, LB-2 and IC2/2 as illustrated in FIG. δ.

PI (SEQ ID NO: 4) coated on plastic efficiently bound leukocytes, but the CDlδ antibody, 7E4, which efficiently blocks CDll/CDlδ-dependent adhesion [Nortamo et al., Scand . J . Immunol . , 28, 537-546 (1988)], had no effect. The 7E4 antibody also did not prevent the binding of ¹²⁵I-P1 (SEQ ID NO: 12) to purified CDlla/CDlδ (not shown) . The results show that peptides according to the present invention are able to bind to CDlla/CDlδ in the presence of antibody in situations wherein ICAMs cannot do so. This may be due to less steric hindrance in the case of the smaller peptide fragments as compared to the natural ligands.

Peptides PI (SEQ ID NO: 4) and P6 (SEQ ID NO: 9) induced homotypic aggregation of Ramos cells. The control peptide P7 (SEQ ID NO: 10) had no effect, demonstrating that the inducing effect by peptides PI (SEQ ID NO: 4) and P6 (SEQ ID NO: 9) is specific. Since the peptides were attached to solid phase and washed extensively, there was no free peptide present. Thus, the aggregation of cells is due to cellular activation rather than the cross- linking capabilities of the peptides. The monoclonal antibody 7E4 completely blocked the peptide Pl-induced aggregation, showing that the aggregation is CDlδ dependent. The effects of monoclonal antibodies H12 (anti-CDlla) , 60.1 (anti- CDllb) , LB-2 (anti-ICAM-1) , and IC2/2 (anti-ICAM-2) indicates that the aggregation is mediated by CDlla/CDlδ, CDllb/CDlδ, ICAM-1, and ICAM-2.

C. Activation of Leukocyte Adhesion By PI

The ability of peptides according to the invention to activate leukocyte adhesion was determined in an assay involving human T cells.

Human T cells were isolated from buffy- coat samples using the Ficoll-Paque (Pharmacia,

Uppsala, Sweden) technique with a nylon wool column. The purity of isolated cells was determined to be 90%. The cells were then cultured overnight in RPMI 1640 culture medium (Gibco, NY) containing 10% fetal calf serum (Flow Laboratories, Scotland) . After culturing, the cells were washed and resuspended in binding medium (RPMI 1640 with 40mM HEPES, 2 mM MgCl₂ and 5% fetal calf serum) . T lymphocytes were depleted with paramagnetic beads (Dynal, Oslo Norway) from the low density fractions using OKT3 monclonal antibodies (Ortho Pharmaceuticals, Raritan, NJ) . Peptides according to the invention were prepared as above by solid-phase synthesis on an Applied Biosystems model 430A peptide synthesizer, using t-BOC-chemistry. Peptide structures were confirmed by amino acid analysis and plasma desorption mass spectrometry. Monoclonal antibodies were constructed for use as competitive inhibitors in the assays. The monoclonal antibodies used were 7E4, an anti-CD18 antibody reported in Nortamo, et al . , Scand . J . Immunol . , 28 : 537-546 (198δ) ; TS1/22, an anti-CDlla antibody reported in Sanchez-Madrid, et al . Proc . Nat . Acad . Sci . , 79 : 7469-7493 (1982); 60.1, an anti-CDllb antibody reported in Wallis, et al , Blood, 67 : 1007-1013 (1986); 6D5, an anti-ICAM-2 antibody reported in Nortamo, et al . , J . Immunol . , 146 : 2530-2535 (1991); B-H19, an anti-ICAM-1 antibody; IOP49d, an anti-VLA-4α antibody (Immunotech S.A., France). Mouse IgG (Dakopatts A/S, Glostru, Denmark) was used as a negative control. The T-cell aggregation assay was performed as follows. Aliquots of 2 x 10⁵ T-cells in 100 μL binding medium per well were placed in microtiter plates (Dynatech Laboratories, Virginia) and incubated with P(Bu)₂ (4/3-phorbol-l2 , 13 dibutyrate, Sigma, St. Louis, Mo.), PI, or P7 at 37° C. Free cells from four randomly-chosen areas per well were quantitated by counting free cells/2.5 mm² microscopic field (200x magnification) . Percent aggregation was calculated as 100 x [1-(number of free cells) / (total number of cells)]. In order to determine that aggregation was due to Pi, another group of T-cells was pre-incubated with one or more of the monoclonal antibodies listed above for 10 minutes at room temperature before addition of PI. The results are shown in Figures 10A-10D. Aggregation of T lymphocytes occurred within 30 minutes of the addition to PI to a suspension of blood-derived T-cells and thereafter cells remained in clusters. The control peptide, P7 , which had the same amino acids as PI, but in random order with the exception of two conserved cysteines, failed to stimulate aggregation. Pl-induced aggregation of T cells was largely CDlla/CD18-ICAM-l dependents, as shown by separate binding studies using anti-CD18, anti-CDlla, and anti-ICAM-1 antibodies, all of which blocked the full extent of aggregation (See, Figure 10C) . Also, the remaining aggregation after treatment with any of the aforementioned antibodies small and loose as shown in Figure 10D. After activation with PI and washing, T-cells exhibit increased binding to purified recombinant ICAM-1 coated on plastic, which binding is inhibited by anti-CDlδ antibody. Finally, shortening of PI by removal of the amino-terminal glycine resulted in an inactive peptide and the removal of amino acids from the carboxyl terminus reduced activity. Those results are shown in Table 1 below. TABLE 1

Peptide Fragment Amino Acid Seq. Aggregation percent(S.D. )

Pl(ICAM-2) GSLEVNCSTTCNQPEVGGLETS 71(3)

P2(ICAM-2) SLEVNCSTTCNQPEVGGLETS 2(2)

P5a(ICAM-2) GSLEVNCSTTCNQPEVGGLE 45(4)

P5b(ICAM-2) GSLEVNCSTTCNQPEVGGL 20(6)

P6(ICAM-1) GSVLVTCSTSCDQPKLLGIETP 0

P7 (control) EVGTGSCNLECVSTNPLSGTEQ 0

In order to determine whether the adhesion of T cells shown above was due to integrin binding or was merely a bridging effect, the same experiment as above was performed under conditions in which adhesion, but not bridging, is inhibited. It is well-known that adhesion generally does not take place at temperatures of 4°C and that adhesion requires energy and divalent cations. Accordingly, as shown in Figure 11, when the above-noted experiments were run at low temperature or in the presence of sodium azide and deoxyglucose or in the presence of ethylenediamine tetraacetic acid (EDTA) , no aggregation occurred. Also, the protein kinase C inhibitor, staurosporine, blocked phorbol ester induced T cell aggregation but not aggregation induced by PI; whereas the phosphatase inhibitor, okadaic acid blocked both PI and phorbol ester induction of aggregation. Both results are shown in Figure 12A and 12B, respectively. EXAMPLE 6

Characteristics of PI Induction of Cellular Aggregation

Several experiments were performed in order to characterize the PI induction of cellular aggregation. The cell lines utilized included

Burkitt's lymphoma cell lines Namalwa, Ramos, and

BL-41; the T-cell leukemia cell line Jurkat; myelo- monocytic cell line U-937; and the K562 erythroleukemia cell line. Each group of cells was cultured in RPMI 1640 medium (Gibco, Grand Island, NY) supplemented with 10% fetal calf serum (Flow Laboratories, Irvine Scotland) , penicillin (50 IU/ml) , streptomycin (50 μg/ml), and glutamine (0.29 mg/ml) . The EBV-transformed B-cell line, NAD-20, and the promyelocytic cell line, HL-60, were grown in RPMI 1640 medium containing 20% fetal calf serum. Culture media were changed one day prior to any experiment. T lymphocytes and granulocytes were isolated from blood buffy coat samples obtained from the Finnish Red Cross Blood Transfusion Center, Helsinki using the Ficoll-Paque technique (Pharmacia, Uppsala, Sweden) . T-cells were further purified using a nylon wool column. After isolation, the cells were incubated at 37° C in RPMI 1640 medium with 10% fetal calf serum overnight before the assays.

Peptides for use in the assays were prepared as above. The structure of the peptides was confirmed by amino acid analysis and plasma desorption mass spectrometry according to the procedure of Li, et al . , J .Biol . Chem . , 268:17513 (1993) . Peptide PI was confirmed to be the structure indicated in SEQ ID NO: 4, corresponding to amino acids 21-42 of ICAM-2. Monoclonal antibodies identical to those above were also prepared.

A. Aggregation Assay In the aggregation assay, cells (as described immediately above) were washed with RPMI 1640 medium containing 40 mM HEPES, 2 mM MgCl₂, and 5% fetal calf serum. The cells were then resuspended to a concentration of 2 x 10⁶ cells/ml for T-cells and granulocytes and 10⁶ cells/ml for all other cells. Aliquots of 100 μl of each cell suspension were added to wells of flat-bottom 96- well microtiter plates either in the presence or absence of P(Bu)₂ or peptide PI or P7 (control) and incubated at 37°C for appropriate times (1,2,4,7, or 24 hours) . For inhibition of Pl-induced aggregation, cells were pre-incubated with inhibitor for 15 minutes at room temperature. Quantitative measurement of cell aggregation was accomplished by counting free cells from four randomly-selected areas (2.5 mm²) per well.

Peptide PI induced aggregation of T-cells, granulocytes, NAD-20 cells, Namalwa cells, Ramos cells, and U-937 cells, but had no effect on cell lines which do not express both the CDlla/CDlδ integrin and ICAM-1, such as BL-41, HL-60, and K562. The control peptide, P7, did not induce any aggregation and P(Bu₂) produced aggregation similar to PI. Results are shown in Table 2 below, wherein the amount of aggregated cells was express as percent aggregation (100 x [1 - (number of free cells) / (total number of cells)] and "None" indicates the negative control (i.e, nothing was added to the cells) .

The ability of Pi to stimulate aggregation of U-937 cells, on which CDllb but not CDlla is expressed, and the inhibition of aggregation by anti-CDllb antibodies indicates that PI activates CDllb/CD18. These findings evidence a novel role for the ICAM-2 peptide in ligand recognition by integrin, namely that ICAM-2 may serve as a "trigger" for high avidity ligand binding.

B. Kinetics and Dose Response of Pl-induced Cell Aggregation

The time course of U-937 and Namalwa cell aggregation induced by Pi was similar and each resembles induction by P(Bu₂) , as shown Figures 13A and 13B, wherein unshaded squares represent treatment with P(Bu)₂, shaded circles represent treatment with PI or P7 , P7 being the lower curve in each case, and shaded triangles represent cells left untreated. After one-hour incubation with PI or P(Bu₂) , cells clearly aggregated as compared to cells treated with the control peptide (P7) or cells left untreated. The maximum level of aggregation was observed after a 4-hour incubation, after which the cells remained in clusters throughout the 24-hour experiment. PI induction of aggregation was stronger than P(Bu)₂ aggregation in U-937 cells as shown in Figure 13B. Further, PI induced aggregation in a concentration-dependent manner (Figures 13C and 13D) . For example, a 50% aggregation was achieved with 5 μg/ml of PI for

Namalwa cells; whereas 100 μg/ml was need to achieve 50% aggregation with U-937 cells.

C. Expression of ICAMs and Integrins on Namalwa and U-937 Cells and Inhibition With Antibodies

Both Namalwa and U-937 cells expressed

CDlδ, ICAM-1, ICAM-2, and ICAM-3 , as shown in Figure

14. Both cell types also expressed CDlla and CDllb, however, U937 cells expressed more CDllb and Namalwa cells expressed more CDlla. Micrographs showing the effects of various monoclonal antibodies on Pl- induced aggregation of Namalwa cells is shown in Figure 15, wherein the specific antibodies are indicated next to each panel and wherein CDlδ represents an anti-CDlδ monoclonal antibody, etc. Monoclonal antibodies against CDlδ and those directed against ICAM-1 effectively inhibited the homotypic adhesion of both Namalwa and U-937 (Figures 16A-16C) . The blocking effect was larger against CDllb as compared to CDlla in U-937 cells, as shown in Figure 16B. Antibodies against ICAM-2 , ICAM-3,as well as control antibodies, failed to inhibit aggregation.

D. Requirements for Pl-induced aggregation No Pl-induced homotypic adhesion of Namalwa or U-937 cells was detected when experiments were conducted at 4° C. Aggregation was partially blocked by NaN₃ and completely blocked by NaN₃ with 2-deoxy-D-glucose, as shown in Figures 17A and 17B, thus indicating that proper function of the cell's energy-producing machinery is required. Cytochalasin B, an inhibitor of microfilament formation blocked aggregation. Ethylenediamine tetraacetic acid also blocked aggregation, indicating the need for divalent cations. Finally, treatment with dibutyryl cyclic AMP slightly blocked aggregation of Namalwa cells but not U-937 cells. Consistent with data reported above, staurosporine, the protein kinase inhibitor, effectively blocked P(Bu)₂-induced aggregation but had no effect on Pl-induced aggregation (See, Figure lδA) . Also okadaic acid, the serine/threonine phosphatase inhibitor, inhibited both P(Bu)₂- and P-

1-induced aggregation. The cAMP- and cGMP-dependent protein kinase inhibitors, KT5720 and KT5823 and the inhibitory G protein inhibitor, pertussis toxin, had little effect on aggregation, as shown in Figures lδC-lδD, respectively. Finally, as shown in Figure

18E, the stimulatory G protein inhibitor, cholera toxin, effectively blocked aggregation induced by P(Bu₂) but not that induced by PI. In Figures 18A- lδF, Shaded circles represent aggregation with Pi alone, unshaded circles represent the effects of the inhibitor on Pl-induced aggregation, open squares represent the effects of the inhibitors on P(Bu)₂- induced aggregation.

E. Activation of Natural Killer Cells

The binding of natural killer cells to K562 leukemic cells was greatly increased by pretreatment with PI. Natural killer cells were preincubated with 100 μg/ml of either PI or P7 in RPMI 1640 medium complemented with 0.5% bovine serum albumin in standard cell culture conditions (albumin at 37 °C in a humidified 5% C)₂ atmosphere) for various periods (See, Figure 19B) before being tested for cytotoxicity against K562 target cells in a 4-hour ⁵¹Cr assay as described in Timonen, et al . , J. Exp . Med . , 153 : 569-5δ2 (19δl), at a 10:1 effector/target cell ratio. The binding capacity of natural killer cells to target was tested at a 1:2 ratio after centrifugation for δ minutes at 120g. Binding is express as a percent of binding from a total pool of lymphocytes. At least 300 cells were counted in each combination. Results showing Pl- induction of natural killer cell activity (i . e . , binding and cytotoxicity) are presented in Figures 19A and 19B. The foregoing results indicate that Pi directly activates natural killer cells since mere clustering of natural killer cells with their targets has been shown not to lead to toxicity.

Timonen, et al . , J . Exp . Med . , 153 : 569-5δ2 (1981) . EXAMPLE 7

Effect of PI on Migration of Natural Killer Cells

The ability of PI to induce migration of natural killer cells was determined. Migration of natural killer cells through a Boyden chamber or similar apparatus indicates that the cells have been activated.

Cells were prepared from buffy coat samples from healthy blood donors (blood samples were obtained from the Finnish Red Cross Transfusion Service) . Mononuclear cells were isolated by Ficoll-Isopaque (Pharmacia Fine Chemicals AB, Uppsala, Sweden) gradient centrif gation and subsequent filtration through nylon wool columns in RPMI 1640 Medium supplemented with 0.29 mg/ml glutamine (Gibco) , 100 IU/ml penicillin, 10 μg/ml streptomycin, and 5% heat-inactivated fetal calf serum (Gibco) . Natural killer cells were further enriched by discontinuous four-step density gradient centrifugation on Percoll (Pharmacia Fine Chemicals AB, Uppsala, Sweden) as described in Timonen, et al . , J . Immunol . Meth . , .36 : 285-291 (I960), incorporated by reference herein, using the four uppermost gradients instead of seven. The cell composition was phenotyped by flow cytometry.

The migration assay was conducted in a Boyden chamber as described by Axelsson, et al . , J . Immunol . Meth . , 46:251-258 (1981) . Polycarbonate filters with 3 μm diameter pores (Nuclepore Corp. ,

Pleasonton, NY) were placed between the upper and lower compartments of the Boyden chamber. The lower compartment was filled with 400 μL RPMI 1640 buffer supplemented with 0.5% human AB serum and 0.5% human serum albumin (both obtained from the Finnish Red Cross Transfusion Service) . Natural killer cells were added to the upper compartment of the chamber. The cell number was adjusted to 10 xlO⁶ cells/ml and 200 μl of cell suspension was added to the upper compartment. ICAM-2 peptides or control peptides were added in amounts varying from 0 to 100 μg/10 xlO⁶ cells/ml to natural killer cell fractions 30 minutes prior to the assay and were left in the supernatants during the assay. As a non-migratory control, other natural killer cell fractions were treated with lOmM ethylenediamine tetraacetic acid (EDTA) 30 minutes prior to assay. In previous experiments, the Applicants determined that EDTA- treated cells fall passively in the lower compartment of the chamber through the incidental double or triple pores of the filters. After incubation in humidified air with 5% C0₂ at 37 °C for 1-4 hours, migratory cells in the lower chamber were counted and those cells were then subjected to flow cytometry analysis. The number of EDTA-treated cells was subtracted from experimental cells and the results are shown in Figure 20, wherein the control peptide was P7 (SEQ ID NO: 10) , as indicated by shaded circles. Open circles represent results obtained from PI

As shown in Figure 20, migration of natural killer cells was dependent upon the ICAM-2 peptide concentration. ICAM-2 began to activate natural killer cells at approximately 3 hours of incubation as shown in Figure 21, wherein shaded circles represent results obtained with the control peptide, P7, and unshaded circle represent results obtained with PI. The number of natural killer cells which migrated through the filters during four hours of incubation was significantly (p<0.001) higher than the number of control-peptide-treated natural killer cells which migrated during the same period.

Although the present invention has been described in terms of its preferred embodiments, it is expected that modifications and variations will occur to those skilled in the art upon consideration of the present disclosure. Accordingly, the present invention is intended to include all such modifications and variations, and, in particular, those within the scope of the following claims.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: Helsinki University Licensing Ltd Oy

(ii) TITLE OF INVENTION: Peptides From Human ICAM-2 and From Human ICAM-1 and Their Analogs for Uβe in Therapy and Diagnosis

(iii) NUMBER OF SEQUENCES: 19

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Helsinki University Licensing Ltd Oy

(B) STREET: Teollisuuskatu 23

(C) CITY: Helsinki

(D) STATE:

(E) COUNTRY: Finland

(F) ZIP: FIN-00510

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

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

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

(2) INFORMATION FOR SEQ ID Nθ:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 31 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: 1inear

(ii) MOLECULE TYPE: peptide

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

Ala Val Glu Pro Lys Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys 1 5 10 15

Asn Gin Pro Glu Val Gly Gly Leu Glu Thr Ser Leu Asn Lys lie 20 25 30

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 31 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: lie Leu Pro Arg Gly Gly Ser Val Leu Val Thr Cys Ser Thr Ser Cys

1 5 10 15

Asp Gin Pro Lys Leu Leu Gly lie Glu Thr Pro Leu Pro Lys Lys 20 25 30

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 31 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

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

Phe Leu Pro Gin Gly Gly Ser Val Gin Val Asn Cys Ser Ser Ser Cys 1 5 10 15

Lys Glu Asp Leu Ser Leu Gly Leu Glu Thr Gin Trp Leu Lys Asp 20 25 30

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

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

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val 1 5 _ 10 15

Gly Gly Leu Glu Thr Ser 20

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly 1 5 10 15

Gly Leu Glu Thr Ser 20

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

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

Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly Gly 1 5 10 15

Leu Glu Thr Ser 20

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 19 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: 1inear

(ii) MOLECULE TYPE: peptide

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

Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly Gly Leu 1 5 10 15

Glu Thr Ser

(2) INFORMATION FOR SEQ ID NO:8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 14 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID Nθ:8:

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro 1 5 10 (2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

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

Gly Ser Val Leu Val Thr Cys Ser Thr Ser Cys Asp Gin Pro Lys Leu 1 5 10 15

Leu Gly lie Glu Thr Pro 20

(2) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

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

Glu Val Gly Thr Gly Ser Cys Asn Leu Glu Cys Val Ser Thr Asn Pro 1 5 10 15

Leu Ser Gly Thr Glu Gin 20

(2) INFORMATION FOR SEQ ID NO:11:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 14 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: 1inear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:11:

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val 1 5 10 15

Gly Gly Leu Glu 20 (2) INFORMATION FOR SEQ ID NO:12:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 23 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val 1 5 10 15

Gly Gly Leu Glu Thr Ser Tyr 20

(2) INFORMATION FOR SEQ ID NO:13:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:

Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly 1 5 10 15

Gly Leu Glu Thr Ser Tyr 20

(2) INFORMATION FOR SEQ ID NO:14:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:

Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly Gly 1 5 10 15

Leu Glu Thr Ser Tyr 20 (2) INFORMATION FOR SEQ ID NO:15:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 20 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:15:

Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val Gly Gly Leu 1 5 10 15

Glu Thr Ser Tyr 20

(2) INFORMATION FOR SEQ ID NO:16:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 15 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:16:

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Tyr 1 5 10 15

(2) INFORMATION FOR SEQ ID NO:17:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 23 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:17:

Gly Ser Val Leu Val Thr Cys Ser Thr Ser Cys Asp Gin Pro Lys Leu 1 5 10 15

Leu Gly lie Glu Thr Pro Tyr 20

(2) INFORMATION FOR SEQ ID NO:18:

(i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 23 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: peptide (xi) SEQUENCE DESCRIPTION: SEQ ID NO:18:

Glu Val Gly Thr Gly Ser Cys Asn Leu Glu Cys Val Ser Thr Asn Pro 1 5 10 15

Leu Ser Gly Thr Glu Gin Tyr 20

(2) INFORMATION FOR SEQ ID NO:19:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 19 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: 1inear

(ii) MOLECULE TYPE: peptide

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:19:

Gly Ser Leu Glu Val Asn Cys Ser Thr Thr Cys Asn Gin Pro Glu Val 1 5 10 15

Gly Gly Leu

Claims

1. A peptide having an amino acid sequence selected from the group consisting of:

SEQ ID NO: 4; and an amino acid sequence differing from SEQ

ID NO: 4 by the addition or substitution of at least one amino acid which does not destroy integrin binding activity.

2. The peptide of claim 1 having the amino acid sequence according to SEQ ID NO: 12.

3. A peptide having an amino acid sequence selected from the group consisting of:

(SEQ ID NO: 5) ; and an amino acid sequence differing from SEQ ID NO: 5 by the addition or substitution of an amino acid which does not destroy integrin binding activity.

4. The peptide of claim 1 having the amino acid sequence of SEQ ID NO: 13.

5. A peptide having an amino acid sequence selected from the group consisting of: (SEQ ID NO: 6) ; and an amino acid sequence differing from SEQ ID NO: 6 by the addition or substitution of an amino acid which does not destroy integrin binding activity.

6. The peptide of claim 1 having the amino acid sequence of SEQ ID NO: 14. 7. A peptide having an amino acid sequence selected from the group consisting of:

(SEQ ID NO: 9) ; and an amino acid sequence differing from SEQ ID NO: 9 by the addition or substitution of an amino acid which does not destroy integrin binding activity.

δ . The peptide of claim 1 having the amino acid sequence of SEQ ID NO: 17.

9. A method for inhibiting transplant rejection comprising the steps of: identifying a patient as being susceptible to transplant rejection; and administering a peptide as recited in claim 1, 2, 3, 4, 5, 6, 7 or δ in combination with a diluent to the patient.

10. A method for activating natural¹ killer cells in a patient comprising the steps of: identifying a patient as benefiting from activation of natural killer cells; and administering a peptide as recited in claim 1, 2, 3, 4, 5, 6, 7 or δ in combination with a diluent to the patient.

11. A method for constructing a peptide having ICAM binding activity comprising the step of chemically synthesizing a peptide having the amino acid sequence of a peptide as recited in claim 1, 2, 3, 4, 5, 6, 7 or δ.

12. A method for constructing a peptide having ICAM binding activity comprising the step of expressing from DNA a peptide having the amino acid sequence of a peptide as recited in claim 1, 2, 3, 4, 5, 6, 7 or δ.

13. An assay kit including a peptide according to one of claim 1, 2 , 3, 4, 5, 6, 7 or δ in combination with a signal for identifying integrin binding of the peptide.

14. A formulation comprising a peptide as shown in SEQ ID NO: 4 in an acceptable diluent, adjuvant, or carrier.