WO1994014065A1

WO1994014065A1 - Methods for identifying and using immunosuppressant compounds

Info

Publication number: WO1994014065A1
Application number: PCT/US1993/011329
Authority: WO
Inventors: Hans Reiser
Original assignee: Dana-Farber Cancer Institute, Inc.
Priority date: 1992-12-14
Filing date: 1993-11-22
Publication date: 1994-06-23

Abstract

A method of identifying a potential immunosuppressive agent by assaying for the ability of a given candidate compound to inhibit T cell activation via the activation pathway involving stimulation by (i) a substance which stimulates the T cell CD28 antigen, and (ii) a phorbol ester; and use of such immunosuppressive agents to treat patients in need of immunosuppression.

Description

METHODS FOR IDENTIFYING AND USING IMMUNOSUPPRESSANT COMPOUNDS Statement as to Federally Sponsored Research This invention was made in the course of work supported by Grant Νo.AI-30169 of the National Institutes of Health; the United States government therefore has certain rights in the invention.

Background of the Invention The field of the invention is immunosuppressants.

Generation of a T cell immune response is a complex process involving cell-cell interactions and production of soluble immune mediators (lymphokines) such as interleukin-2 (IL-2) . Each T cell bears on its surface many copies of an antigen-recognizing structure termed the T cell receptor/CD3 complex; when presented with the appropriate foreign airtigenic peptide by an antigen-presenting cell (APC) , this structure is stimulated, resulting in the generation of critical intraceilular signals that can lead to activation of the T cell if supplemented with signals from a second type of T cell surf- ^e structure. This second type of T cell surface structure is termed the CD28 antigen. It is productively stimulated upon contact with a molecule, known as B7, found on the surface of activated B cells (including APCs) as well as certain other cell types (Gimmi et al., Proc. Natl. Acad. Sci. USA 88:6575-6579, 1991; Reiser et al., Proc. Natl. Acad. Sci. USA 89:271- 275, 1992) . Stimulation of the CD28 antigen can also be accomplished in the absence of B7 by the use of an anti- CD28 antibody (Thompson et al., PCT publication number WO 90/05541) . Similarly, stimulation of the T cell receptor/CD3 complex in the absence of properly presented antigenic peptides can be accomplished experimentally with lectins or with an anti-CD3 antibody. Co- stimulation of these two structures leads to clonal proliferation of the T cell, enhanced by the newly- induced synthesis of IL-2 and IL-2 receptors by the cell.

A third type of T cell surface structure is termed the CTLA-4 antigen. This antigen can also function as a receptor for the B7 molecule, but the functional significance of the CTLA-4/B7 interaction is not understood.

It has been shown that activation of T cells can also take place via an alternative pathway that bypasses the T cell receptor/CD3 complex. Activation via this alternative pathway is triggered by the action of a phorbol ester, such as phorbol myristate acetate (PMA) , directly on the intraceilular enzyme protein kinase C. As with the T cell receptor/CD3 pathway, activation triggered by a phorbol ester requires co-stimulation of CD28 by the CD28 ligand, B7, or by an anti-CD28 antibody. The biological significance of this alternative pathway, herein termed the phorbol ester/CD28 pathway, is not understood. Cyclosporine, a fungal metabolite that is widely used to suppress immune rejection of organ transplants, acts by inhibiting T cell activation. It has been shown that cyclosporine inhibits activation triggered by co- stimulation of CD28 and the T cell receptor/CD3 complex, but does not inhibit activation triggered by the phorbol ester/CD28 pathway (June et al., Mol. Cell. Biol. 7:4472- 4481, 1987) .

Summary of the Invention The screening and therapeutic methods of the invention are based upon the discovery that it is possible to inhibit the phorbol ester/CD28 activation pathway in T cells with a compound that is not an antibody or other polypeptide. Although a monoclonal antibody such as anti-B7 would presumably function to prevent stimulation of CD28 by binding to endogenous B7 on APCs, thereby blocking the binding of B7 to CD28 on T cells, therapeutics based upon polypeptides such as antibodies possess the disadvantage of being costly to manufacture; in addition, they are of limited use as therapeutics because (1) they tend to be highly immunogenic themselves, (2) they generally cannot be administered orally, and (3) they are typically destroyed by proteases present in the blood before they can reach therapeutic concentrations at the target site. It is therefore of great interest to develop non-polypeptide- based pharmaceuticals capable of suppressing the immune response, for use in preventing transplant rejection. The assay method described below revealed that Mycalamide A, a non-polypeptide compound with known anticancer properties which is isolated from a Mycale sp . sponge, is capable of inhibiting both pathways of T cell activation; furthermore, it inhibits T cell activation at far lower doses than either of the known immunosuppressant drugs to which it is herein compared. This discovery defines a new class of immunosuppressants hitherto unknown to exist (i.e., those which inhibit the PMA/CD28 activation pathway) , and validates the use of the screen of the invention as a means for discovering additional immunosuppressive agents of this class.

The invention thus features a method of identifying a potential immunosuppressive agent from a plurality of candidate compounds which are not polypeptides, which method includes the steps of contacting a first T cell with (i) a substance which stimulates the T cell CD28 antigen, and (ii) a phorbol ester, in the presence of a first candidate compound; contacting a second T cell with (i) the CD28- stimulating substance, and (ii) a phorbol ester, in the presence of a second candidate compound; contacting a third T cell with (i) the CD28- stimulating substance, and (ii) a phorbol ester, in the absence of the first and second candidate compounds; determining the level of activation of each of the test T cells; and selecting a candidate compound which inhibits activation of the T cells, such inhibition being an indication that the selected candidate compound is a potential immunosuppressive agent.

The CD28-stimulating substance is preferably either an anti-CD28 antibody or CD28-binding fragment thereof, or is B7, an antigen identified on the surface of activated B cells which acts as the natural ligand for CD28 and CTLA-4. Such B7 may be soluble B7 or may be anchored to a cell membrane or a liposome. The cell membrane may be a cell-free membrane preparation, an intact cell which naturally bears the B7 antigen, or a cell (e.g., a non-lymphocytic cell such as a Chinese Hamster Ovary [CHO] cell) transfected with a cDNA encoding B7 (e.g., human or urine B7) , and expressing this B7 on its surface. The phorbol ester can be phorbol myristate acetate (PMA) , or any phorbol ester capable of activating T cell via the same mechanism as PMA. Examples of such phorbol esters include l-oleoyl-2- acetylglycerol (OAG) , dioctanoylglycerol (diC₈) , and phorbol-12,13-dibutyrate (PDBu) . Activation can be detected by any standard means, including by measuring proliferation or lymphokine gene expression.

The screening method preferably includes an additional assay for activation via the T-cell receptor/CD3 complex pathway, which assay would include the steps of contacting a fourth, a fifth and a sixth T cell with the CD28-stimulating substance and a molecule which stimulates the T-cell receptor/CD3 complex: the fourth in the presence of the first candidate compound, the fifth in the presence of the second candidate compound, and the sixth in the presence of neither candidate compound; and selecting a candidate compound which is effective at inhibiting T cell activation by both activation pathways.

The screening method can alternatively be carried out by the following steps: contacting a first T cell with a candidate compound in the presence of (i) a substance which stimulates the T cell CD28 antigen, and (ii) a phorbol ester; contacting a second T cell with the candidate compound in the presence of (i) the CD28-stimulating substance, and (ii) a molecule which stimulates the T- cell receptor/CD3 complex of the second T cell; and determining whether the candidate compound inhibits activation of both the first and the second T cells, wherein inhibition of activation of both the first and the second T cells is an indication that the candidate compound is an immunosuppressant.

Also within the invention is a therapeutic method including the steps of identifying a patient (e.g., a human or other mammal such as a dog, cat, cow, horse, pig, goat, mouse, rat, rabbit, guinea pig or hamster) in need of an immunosuppressant; providing a compound which inhibits activation of T cells induced by a combination of (i) an anti-CD28 antibody and (ii) PMA; and administering an effective amount of the compound to the patient, provided that the compound is not an antibody or other polypeptide. By "effective amount" is meant an amount of the compound sufficient to decrease or prevent an increase in the level of activation of the patient's T cells. Such an amount would depend upon the potency of the particular compound used, the weight of the patient, the route of administration, and the intensity of T cell activation stimuli the therapy is intended to counteract. In preferred embodiments, the compound is additionally characterized as inhibiting the T cell activation pathway induced by a combination of (i) an anti-CD28 antibody and (ii) an anti-CD3 antibody. Examples of such a compound include the mycalamides, a class of compounds disclosed by Blunt et al., U.S. Patent No. 4,868,204; this class includes Mycalamide A and derivatives thereof. The treatment method of the invention can be used to decrease or prevent transplant rejection, graft-vs-host disease, or autoimmune disease. The therapeutic method of the invention may additionally include the steps of administering to the patient, in conjunction with the first compound, a second therapeutic compound (e.g., cyclosporine A, FK506, or rapamycin) , which second compound acts by inhibiting the T cell activation pathway triggered by a combination of (i) an anti-CD28 antibody and (ii) an anti-CD3 antibody. Such combination therapy may be accomplished by simultaneous administration of the two compounds (e.g., in a mixture) , or by administering one prior to administering the other. Such combination therapy has the advantage of suppressing both of the mechanisms of T cell activation, thereby increasing the effectiveness of the treatment.

Other features and advantages of the invention will be apparent from the detailed description set forth below, and from the claims.

Brief Description of the Drawings Figs. 1A and IB are graphs illustrating the effect of CsA on TCR- or mB7-induced T cell activation. (A) Costimulation of T cells with anti-CD3 mAb and mB7. 2xl0⁵ purified BALB/c CD4+ T lymphocytes were stimulated with the indicated concentrations of anti-CD3 mAb in the presence of medium (O) , 2xl0⁴ paraformaldehyde-fixed CHO- mB7 cells (D) , or 2xl0⁴ paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 g/ml (0.08 μ,M) CsA (A) or 1 μg/ml (0.8 μM) CsA (■) .

(B) Costimulation of T cells with mB7 and PMA. 2xl0⁵ purified BALB/c CD4+ T lymphocytes were stimulated with the indicated concentrations of PMA and either medium (O) , 2xl0⁴ paraformaldehyde-fixed CHO-mB7 cells (D) , or paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 μg/ml CsA (■) . All cultures ware incubated for 48 hours and pulsed with lμCi [³H]~thymidine/well for the last 6 hours of the incubation period to assay for T cell proliferation.

Figs. 2A and 2 are graphs showing that the effects of FK506 on mB7-mediated T cell activation parallel those of CsA. (A) Costimulation of T cells with anti-CD3 mAb and mB7. 2xl0⁵ purified BALB/c CD4⁺ T lymphocytes were stimulated with the indicated concentrations of anti-CD3 mAb in the presence of medium (0) , 2xl0⁴ paraformaldehyde-fixed CH0-mB7 cells (D) , or 2xl0⁴ paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 nM FK506 (T) , l nM FK506 (A) or 10 nM FK506 (I) .

(B) Costimulation of T cells with mB7 and PMA. 2xl0⁵ purified BALB/c CD4⁺ T lymphocytes were stimulated with the indicated concentrations of PMA and either medium (o) , 2xl0⁴ paraformaldehyde-fixed CH0-mB7 cells (D) , or paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 nM FK506 (T), 1 nM FK506 (A) or 10 nM FK506 (■) . All cultures were incubated for 48 hours and pulsed with lμCi [³H]-thymidine/well for the last 6 hours of the incubation period to assay for T cell proliferation. Figs. 3A and 3B are graphs showing that Mycalamide A blocks activation of murine CD4+ T cells by anti-CD3 mAb and mB7, as well as activation by mB7 and PMA.

(A) Costimulation of T cells with anti-CD3 mAb and mB7. 2xl0⁵ purified BALB/c CD4⁺ T lymphocytes were stimulated with the indicated concentrations of anti-CD3 mAb in the presence of medium (O) , 2xl0⁴ paraformaldehyde-fixed CHO- mB7 cells (D) , or 2xl0⁴ paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 pM Mycalamide A (▼) , 1 pM Mycalamide A (A) or 10 pM Mycalamide A (■) .

(B) Costimulation of T cells with mB7 and PMA. 2xl0⁵ purified BALB/c CD4⁺ T lymphocytes were stimulated with the indicated concentrations of PMA and either medium (O) , 2xl0⁴ paraformaldehyde-fixed CHO-mB7 cells (D) or paraformaldehyde-fixed CHO-mB7 cells in the presence of 0.1 pM Mycalamide A (T) , 1 pM Mycalamide A (A) or 10 pM Mycalamide A (!) . All cultures were incubated for 48 hours and pulsed with lμCi [³H]-thymidine/ well for the last 6 hours of the incubation period to assay for T cell proliferation.

Detailed Description The screening method of the invention is carried out by exposing a T cell to a candidate immunosuppressant compound in the presence of substances known to activate resting T cells via the cyclosporine- resistant activation pathway involving co-stimulation by (a) B7 antigen, anti-CD28 antibody, or another substance which stimulates CD28 on the T cell; and (b) a phorbol ester (in particular, PMA) or another substance which acts on protein kinase C in a manner similar to that of phorbol esters. In contrast to a second type of activation pathway involving co-stimulation of (a) the CD28 antigen and (b) the T-cell receptor/CD3 complex, this first pathway has never previously been shown to be affected by an immunosuppressant compound, and so there was no reason to believe it could serve as a basis for screening potential immunosuppressant compounds. The results disclosed below establish the feasibility of using this assay as such a screen. As described below, a compound capable of inhibiting activation via the PMA- induced pathway is potentially a much more powerful immunosuppressant than a compound which acts only via the CD28-and-T-cell receptor/CD3 complex pathway. Most preferred is a compound which is capable of blocking both palαways of activation, and so a preferred embodiment of the invention is a screen incorporating two separate assays: one assaying inhibition of the first pathway, and a second assaying inhibition of the second assay. Example I below describes the assay of the invention as applied to the known immunosuppressants cyclosporine A and FK506. In Example II, the compound Mycalamide A is used in the assay. Example III sets forth cytotoxicity data relating to Mycalamide A. Example IV describes how the assay of the invention would be used to screen other potential compounds, while Example V provides a description of the therapeutic method of the invention.

Materials and Methods

Immunosuppressive Reagents. Cyclosporine A (CsA) (Sandimmune^R) was purchased in liquid form from Sandoz Corporation, East Hanover, NJ. FK506 was kindly provided by Lymphomed/Fujisawa Corporation, Melrose Park, II the compound was provided in powder form and was dissolved in DMSO. It is a neutral macrolide produced by Streptomyces tsukubaensis (Kino et al. , J. Antibiotics 40:1249-1255, 1987). Mycalamide A was kindly provided by Harbor Branch Oceanographic Institution, Fort Pierce, FL, in lyophilized form, and was reconstituted with ethanol. It can be extracted from marine sponges of the genus Mycale , family Mycalidae, and order Poecilosclerida, as described by Blunt et al., U.S. Patent No. 4,868,204 (herein incorporated by reference) . All compounds were diluted in medium prior to addition to the cultures. Monoclonal Antibodies. The following monoclonal antibodies (mAbs) were used in this study: mAb M5/114 = anti-I-A^b'^d'3, anti-I-E^d' (available from the American Type Culture Collection, Bethesda, MD) (Bhattacharya et al., J. Immunol. 127:2488, 1981); mAb ADH4 = anti-CD8.2 (Gottlieb et al., Immunogenetics 10:545, 1980); mAb 145-2C11 = anti-CD3 (Pharmingen Corp, San Diego, CA) (Leo et al., Proc. Natl. Acad. Sci. USA 84:1374, 1987).

Cell Lines. The CH0-mB7 and CHO cells have been previously described (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992). CHO-mB7 is a clone of CHO cells stably transfected with cDNA encoding murine B7, which cells express murine B7 (mB7) on their surfaces (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992). These cell lines were grown in a 1:1 mixture of F12 (Gibco, Grand Island, NY) and Dulbecco's modified Eagle media (4.5 g/1 glucose formula, Irvine Scientific, Santa Ana, CA) supplemented with the following ingredients: 10% fetal calf serum (FCS) , 10 M HEPES (Irvine Scientific) , 4mM L-glutamine (Irvine Scientific) , and 400 μg/ml [effective drug concentration] G418 (Gibco) .

Cell Cultures. Mouse T cells were purified as previously described (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992; Reiser and Benacerraf, Proc. Natl. Acad. Sci. USA 86:10069, 1989; Reiser, J. Immunol. 145:2077, 1990). Briefly, splenocytes were depleted of red blood cells by treatment with Tris-NH₄C1. T cells were enriched by nylon wool fractionation (Julius et al., Eur. J. Immunol. 3:645, 1973). CD4+ T cells from BALB/c mice were purified by two-fold treatment with a mixture of anti-MHC Class II and anti-CD8 mAbs and rabbit complement (Cedarlane, Hornby, Ontario) (Reiser and Benacerraf, Proc. Natl. Acad. Sci. USA 86:10069, 1989). CHO and CHO-mB7 cells were fixed with 1% paraformaldehyde prior to addition to the cultures (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992). Microcultures were set up in duplicate in 96-well plates. Briefly, 2xl0⁵ T cells were cultured with or without 2xl0⁴ paraformaldehyde-fixed CHO transfectants in 0.2 ml of RPMI 1640 (Irvine Scientific), supplemented with 10% FCS, 4mM L-glutamine, 10 mM HEPES, 5xl0^~5 M 2-ME, antibiotics, and non-essential amino acids. The precise culture constituents are described in the respective experimental protocols. All cultures were incubated for 48 hours and pulsed with lμCi [³H] thy idine/well for the last 6 hours of the incubation period to assay for T cell proliferation (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992; Reiser and Benacerraf, Proc. Natl. Acad. Sci. USA 86:10069, 1989).

Example I. Activation of murine CD4+ T cells by anti-CD3 mAb and mB7 is sensitive to CsA and FK 506 while activation by mB7 and PMA is CsA- and FK 506-resistant.

CsA is an inhibitor of T cell activation which blocks Ca²⁺-dependent activation pathways that lead to transcription of lymphokine genes. It binds a cytosolic receptor of the immunophilin class, termed cyclophilin, but its precise mechanism of action is unknown (Schreiber, Science 251:283, 1991). Recent experiments suggest that complexes of CsA and cyclophilin can bind to a calmodulin-dependent serine/threonine phosphatase in Ca²⁺-dependent fashion (Friedman and Weissman, Cell

66:799, 1991; Liu et al., Cell 66:807, 1991) and inhibit its activity [Liu et al., Cell 66:807, 1991; Fruman et al., Proc. Natl. Acad. Sci. USA 89:3686, 1992). Studies in the human system have shown that stimulation of T cells by anti-CD3 mAb in the presence of anti-CD28 mAb is sensitive to CsA. In contrast, stimulation with anti-CD28 mAb and PMA is CsA-resistant (June et al., Mol. Cell. Biol. 7:4472, 1987). The molecular basis for this differential sensitivity is unknown.

The effects of CsA on stimulation of murine CD4+ T cells by the CD28 ligand mB7 in the presence of either anti-CD3 mAb or PMA were first assayed. A representative experiment is shown in Fig. 1. CD4+ T cells were purified as detailed in Material and Methods. The resulting population is >95% pure as judged by flow fluorocytometry (data not shown; Reiser and Benacerraf, Proc. Natl. Acad. Sci. USA 86:10069, 1989) and does not respond to mitogenic concentrations of the anti-CD3 mAb 145-2C11 mAb in the absence of a costimulatory signal (Fig. 1A) . In contrast, when purified CD4+ T cells are incubated with anti-CD3 mAb in the presence of fixed CHO- mB7 cells, a dramatic increase in T cell proliferation is observed. The effect of mB7 on T cell activation is specific, as only a marginal increase in T cell proliferation is induced by vector-transfected CHO cells (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992, data not shown) . As shown in Fig. 1A, CsA blocks T cell activation by anti-CD3 mAb and CH0-mB7 cells in dose- dependent fashion.

The addition of either CHO-mB7 cells or PMA to purified murine T cells does not lead to T cell proliferation (Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992, data not shown). In the presence of CHO- mB7 cells, PMA can induce T cell proliferation in dose- dependent fashion (Fig. IB) . Activation of murine CD4+ T cells by this pathway is CsA-resistant (Fig. IB) . A second, recently discovered ir nosuppressive agent, FK506, also inhibits T cell activa n. FK506 binds to another soluble, cytosolic T eel, niolecule of the immunophilin class, FKBP (Schreiber, Science 251:283, 1991) . Available evidence suggests that the FK506:FKBP and CsA:cyclophilin complexes affect the same signalling component (Liu et al., Cell 66:807, 1991; Fruman et al., Proc. Natl. Acad. Sci. USA 89:3686, 1992). As shown in Fig. 2A, FK506, like CsA, blocks the activation of purified CD4+ T cells by anti-CD3 mAb and CHO-mB7 cells. On a molar basis, FK506 is approximately 100 times more potent than CsA (Compare Figs. 1A and 2A) . In agreement with the findings disclosed herein, Kino et al. have shown that FK506 can block mixed lymphocyte reactions or the generation of cytotoxic T lymphocytes at a concentration 100-fold lower than CsA (Kino et al., J. Antibiotics 40:1249, 1987; Kino et al., J. Antibiotics 40:1256, 1987). Activation of T cells with CHO-mB7 cells and PMA is FK506-resistant (Fig. 2B) . Thus, the effects of FK506 on mB7-mediated T cell activation qualitatively parallel those of CsA.

Example II. Mycalamide A blocks activation of murine CD4+ T cells by anti-CD3 mAb and mB7, as well as activation by mB7 and PMA. Mycalamide A is a compound isolated from a

Mycale sp. sponge. This reagent is a potent inhibitor of tumor replication and displays activity against various lymphomas and carcinomas (Burres and Clement, Cancer Research 49:2935, 1989). The molecular basis for the effects of Mycalamide A has not been elucidated. In the preliminary biochemical studies carried out to date, Mycalamide A specifically inhibited the biosynthesis of the p21^ras protein and reverted the morphology of ras- transformed NRK cells to the normal phenotype [Ogawara et al., Chem. Phar . Bull. 39:2152, 1991).

When tested in the T cell activation assays described herein, Mycalamide A was found to block the activation of purified CD4+ T cells by either activation pathway. In the assay using anti-CD3 mAb and CHO-mB7 cells to stimulate activation (Fig. 3A) , Mycalamide functioned with very high efficiency compared to FK506 and Cyclosporine. Its 50% inhibitory concentration is approximately 1 pM, and T cell proliferation is abrogated at a concentration of 10 pM (Fig. 3A) . On a molar basis, Mycalamide A appears to be at least 10-fold more potent than FK506 and approximately 1000-fold more potent than Cyclosporin A in inhibiting the activation of murine CD4+ T cells. Most interestingly, the effects of Mycalamide A on the activation of murine CD4+ T cells by mB7 and PMA parallel its effects on the mB7 and anti-CD3 MAb pathway (Fig. 3B) . Thus, the mechanism of immunosuppressive action of this compound is distinct from that of CsA and FK506, which block T cell activation through the TCR complex in combination with B7, but do not affect T cell stimulation with mB7 and PMA.

Example III. Cytotoxicity of Mycalamide A.

The viability of T cells following treatment with Mycalamide A was assaying using the standard technique of trypan blue exclusion as an indicator of viability. Approximately 1.5 x 10⁶ T cells were seeded out in 1 ml of culture medium. Compounds were added as indicated in Table 1. Cells were harvested 20 hours after initiation of the culture, and treated with trypan blue to distinguish viable from nonviable cells. Indicated in the table is the number of cells excluding trypan blue and, in parentheses, the percentage of trypan blue-positive cells. The table shows two independent experiments, with each data point representing the mean value of two independent cultures.Importantly, incubation of T cells with Mycalamide A does not significantly affect cell viability as assessed by trypan blue exclusion. Thus, Mycalamide A does not appear to act merely by intoxicating the T cell.

Table 1. Effect of CsA, FK 506, and Mycalamide A on T cell viability

Treatment Experiment l Experiment 2 none 1.59 xlO⁶ (<2%) 1.39 xlO⁶

(<2%)

CsA 0.1 μg/ml 1.48 XlO⁶ (<2%) 1.67 lO⁶

(<2%)

CsA 1 / g/ml 1.36 XlO⁶ (<2%) 1.59 XlO⁶ (<2%)

FK506 InM 1.40 XlO⁵ (<2%) 1.64 XlO⁶

(<2%)

Mycalamide A lpM 1.39 XlO⁶ (<2%) 1.45 XlO⁶

(<2%) Mycalamide A 5pM 1.41 XlO⁶ (<2%) 1.68 XlO⁶

(<2%)

Mycalamide A lOpM N.D. 1.49 XlO⁶

(<2%)

Example IV. The screening assay.

A preparation of primary T cells (CD4+ and/or CD8+) is obtained from mice or a healthy human donor, and depleted of red blood cells and B lymphocytes as described above. Alternatively, cells from an appropriate T cell line such as Jurkat (ATCC) can be used; where the cells, like Jurkat, proliferate independently of activation stimuli, lymphokine gene expression would be the criterion tested. Samples of the cells are exposed to varying concentrations of the candidate compound in the presence of (1) a phorbol ester such as PMA, l-oleoyl-2- acetylglycerol (OAG) , dioctanoylglycerol (diC₈) , or phorbol-12,13-dibutyrate (PDBu) (such compounds are available from commercial sources and/or may be prepared by standard methods) ; and (2) (a) an anti-CD28 antibody such as that produced by Clone 37.51 (Pharmingen Corp., San Diego, CA) , or an antigen-binding fragment thereof, which antibody is either in soluble form or attached to a surface by standard means (e.g., incubation on a polystyrene surface, e.g. in a microtiter plate) ; or (b) a source of B7 antigen such as soluble recombinant B7, or paraformaldehyde-fixed CHO cells transformed with a cDNA encoding B7 (as described above and by Reiser et al., Proc. Natl. Acad. Sci. USA 89:271, 1992), or liposomes containing on their surface recombinant B7 antigen (prepared as described by, for example, Mimms et al., Biochemistry 20:833-840, 1981), or cell-free membrane preparations including B7 antigen (prepared by standard methods, e.g. N₂-cavitation of B7-containing cells, followed by centrifugation at 100,000xG to produce pelleted membranes, which can then be used as is or solubilized in detergent-containing buffer) , or cells which naturally express B7 antigen but not the class of MHC antigen to which the T cells utilized respond. For example, mouse L cells express B7 but not MHC class II antigen, and so could be used in studies of CD4+ T cells (Razi-Wolf et al., Proc. Natl. Acad. Sci. USA 89:4210- 4214, 1992) . The amount of B7 present in each assay sample should be at least 10⁷ molecules per 10⁵ T cells.

The concentration of PMA or other phorbol ester used would ideally be optimized for the particular T cell preparation, but would be expected to be approximately 10 ng/ml (ranging from approximately 0.1 ng/ml to 10 μg/ml) . The candidate compound should be present at a variey of concentrations, from zero up to at least 10 μM. Activation of the T cells can be measured as a function of proliferation of the cells, commonly determined by exposing the cells to radioactively labelled thy idine for a period of time, and then measuring the amount of radioactivity incorporated into acid-precipitable DNA. A second means of determining activation of T cells is by measuring expression of a lymphokine such as IL-2, IL-4, tumor necrosis factor-alpha, lymphotoxin, granulocyte- macrophage colony stimulating factor, or γ-interferon. The amount of a given lymphokine secreted by a sample of T cells can be readily determined by assaying the supernatant of the cells with an antibody specific for the lymphokine, e.g. in an E ISA or other type of immunoassay, or by determining the ability of the supernatant to support growth of a type of cell dependent upon the lymphokine being assayed. Alternatively, the level of lymphokine gene transcription in the T cells can be determined by standard methods, e.g. by Northern blot. A compound which inhibits proliferation and/or lymphokine production by T cells stimulated by treatment with phorbol ester and either B7 or anti-CD28 is a potential immunosuppressive agent. Such a compound can also be tested to determine whether it, like Mycalamide A, inhibits activation by the second activation pathway involving CD28 and the T-cell receptor/CD3 complex. Mycalamide A can serve as a useful positive control for inhibition by both pathways, while cyclosporine and FK506 can serve as positive controls for the latter pathway and negative controls for the former. Example V. Method of treatment.

A compound, such as Mycalamide A, which inhibits activation of T cells by the PMA/CD28 pathway can be used to suppress immune rejection of a xenogeneic or allogeneic organ or cellular transplant, either by itself or in combination with a second immunosuppressant which inhibits the second (T-cell receptor/CD3-based) pathway of T cell activation: e.g., cyclosporine, FK506, or rapamycin. The animal so treated can be a human or any other mammal, and the transplant can be, for example, a kidney, heart, lung, liver or other organ, or any appropriate type of cellular transplant (e.g., bone marrow, neural cell, myocyte, hepatocyte, or islet cell) . In order to prevent graft-versus host disease, the transplant can be pretreated with the compound ex vivo, prior to insertion into the patient. Treatment may be by any standard route: e.g., oral, intravenous, subcutaneous, intramuscular, or intraperitoneal, and may be by injection or implantation of a slow-release implant containing the compound. The optimal dosage and treatment regimen can be readily determined by one of ordinary skill in the art of pharmacology. It is expected that an IV dosage of approximately 1 μg to 10 mg per kg per 12 hour period, or an oral dosage of 1 μg to 200 mg per kg per 12 hour period, will be useful for preventing transplant rejection. Furthermore, treatment with such an immunosuppressive compound in similar dosages and regimens may be useful not only for prevention of organ rejection or graft-vs-host disease, but also for reversal of such a condition.

The compounds of the invention are also useful for preventing or decreasing the effects of autoimmune disease, including type 1 diabetes, rheumatoid arthritis, systemic lupus erythmatosis, and alopecia areata. The compounds would be administered as described above.

Claims

What is claimed is:

1. A method of identifying a potential immunosuppressive agent from a plurality of candidate compounds which are not antibodies, which method comprises the steps of contacting a first T cell with (i) a substance which stimulates the T cell CD28 antigen, and (ii) a phorbol ester, in the presence of a first candidate compound; contacting a second T cell with (i) said substance, and (ii) a phorbol ester, in the presence of a second candidate compound; contacting a third T cell with (i) said substance, and (ii) a phorbol ester,^" in the absence of said first and second candidate compounds; determining the level of activation of each of said T cells; and selecting a candidate compound which inhibits said activation, said inhibition being an indication that said selected candidate compound is a potential immunosuppressive agent.

2. A method of screening for potential immunosuppressive compounds, which method comprises the steps of contacting a first T cell with a candidate compound in the presence of (i) a substance which stimulates the T cell CD28 antigen, and (ii) a phorbol ester; contacting a second T cell with said candidate compound in the presence of (i) said substance, and (ii) a molecule which stimulates the T-cell receptor/CD3 complex of said second T cell; and determining whether said candidate compound inhibits activation of both said first T cell and said second T cell, wherein inhibition of activation of both said first T cell and said second T cell is an indication that said candidate compound is a particularly potent immunosuppressant.

3. The method of claim 1 or claim 2, wherein said substance which stimulates the T cell CD28 antigen is B7.

4. The method of claim 3, wherein said B7 is on the surface of a cell transfected with a cDNA encoding said B7.

5. The method of claim 4, wherein said cell is a non-lymphocytic cell.

6. The method of claim 5, wherein said cell is a CHO cell transfected with a cDNA encoding said B7.

7. The method of claim 3, wherein said B7 is human B7.

8. The method of claim 3, wherein said B7 is murine B7.

9. The method cf claim 1 or claim 2, wherein said substance is an anti-CD28 antibody.

10. The method of claim 1 or claim 2, wherein said phorbol ester is phorbol myristate acetate (PMA) .

11. The method of claim 1 or claim 2, wherein said activation is determined by detecting proliferation of said T cells.

12. The method of claim 1 or claim 2, wherein said activation is determined by detecting secretion of a lymphokine by said T cells.

13. The method of claim 12, wherein said lymphokine is interleukin-2, interleukin-4, tumor necrosis factor-alpha, lymphotoxin, granulocyte- macrophage colony stimulating factor, or γ-interferon.

14. The method of claim 2, wherein said candidate compound is a mycalamide.

15. A non-antibody pharmaceutical compound used to suppress the immune system of a patient, which compound is characterized in that it inhibits the T cell activation pathway triggered by a combination of (i) an anti-CD28 antibody and (ii) PMA.

16. The compound of claim 15, further characterized as inhibiting the T cell activation pathway induced by a combination of (i) an anti-CD28 antibody and (ii) an anti-CD3 antibody.