US 5,935,803 AGrant

Methods to Identify Immunomodulators Using Cognate Interaction of PKC-theta

Issue Date:1999-08-10

•8 Claims

•17 Drawing Sheets

Abstract

Modulators of the immune system can be identified by determining the ability of candidate substances to affect the interaction of PKC-theta or a fragment thereof with its cognate. The interaction can be measured using binding of the cognate as an index, or can be measured using a physiological response.

Metadata

Assignee

Terrapin Technologies, Inc.

Inventors

Nicki J. Vasquez
Dorit Ron
Anna F. Voronova
Eugene W. Napolitano

Application Information

Application Number:US 6656471

Filing Date:1996-06-18

Priority Date:1994-02-01

Art Unit:162

Classifications

IPC:

C12Q 102C12Q 148C12Q 168

Field of Search:

4356;7.8;15;29

Patent Drawings (17 sheets)

Description

TECHNICAL FIELD

The field of this invention is methods for identifying pharmaceutical agents for use in modulating activities of the immune system and methods of using agents identified in the disclosed methods.

BACKGROUND ART

PCT Application WO 95/21252, discloses and claims peptide compositions which alter the activity of a signal-generating protein with respect to its cognate protein wherein the cognate protein contains at least one WD-40 region which putatively interacts with the signal-generating protein. The peptide compositions mimic the WD-40 regions, thus competing with the interaction of the cognate with the signal-generating protein. This competition results either in inhibiting the signal-generation or activating it.

One specifically exemplified signal-generating protein is protein kinase C (PKC); the illustrated cognate receptor for activated kinase C (RACK), in this case specific for .beta.PKC, was designated RACK1. The gene encoding RACK1 was cloned and sequenced, showing that RACK1 contains the requisite WD-40 regions.

The above PCT application and U.S. Ser. Nos. 08/473,089, 08/477,346, and 08/487,072 further describe methods to identify additional pairs of signal-generating proteins and their cognates and methods for recognizing WD-40 sequences in the cognates. These applications also note that such interactions can be used as a system to identify additional molecules that bind the signal-generating protein by measuring the effect of candidate binding molecules on the interaction between the signal-generating protein and either its cognate per se or the polypeptide compositions that mimic the WD-40 regions of the cognate.

In U.S. Ser. No. 08/541,964, several specific peptides were identified that bind either to the signal-generating protein or to the cognate protein in a signal-affecting manner. The use of the signal-generating protein/cognate system to assay for modulators of signal transduction in assays which are independent of the purity of these participants was described. The PKC enzyme system was illustrated as a specific embodiment. In addition, peptides which reside on the signal-generating protein, as well as those which reside on the cognate or mimics thereof, were described as being useful to modulate the signal-generating interactions and biological activities which are mediated by the signal-generating interactions.

In U.S. Ser. No. 08/594,447, experiments that demonstrated the identity of a cognate protein for PKC-theta as the fyn protein were described. Since it is well established that fyn is involved in mediation of T-cell responses, it is apparent that disruption of the interaction of PKC-theta with its fyn cognate mediates the immune response. It is also apparent that PKC-theta is a mediator of the immune response. Substances which can be shown to disrupt the interaction between PKC-theta and its cognate fyn, or, as described hereinbelow, to influence the interaction of PKC-theta with any cognate also have immunomodulating activity. The identification of fyn as a PKC-theta cognate, and the consequences of interfering with this association, demonstrate that PKC-theta is a significant signaling protein involved in the immune response.

DISCLOSURE OF THE INVENTION

The present invention is directed to an efficient assay system to identify modulators of the immune system. The assay system measures the effect of candidate substances on the interaction between a signaling protein--in this case PKC-theta, or a fragment thereof, with its cognate. The cognate comprises the relevant portion of a molecule which binds to PKC-theta. Because the method takes advantage of inherent biological specificity, it can be conducted on impure preparations of the participants in the signal pathway--i.e., the above-mentioned signal-generating protein PKC-theta and its cognate receptor. The assay is conducted by assessing the interaction between the signal-generating protein and its cognate either by measuring binding directly or by measuring a physiological or metabolic effect. The measurement is made in the presence and in the absence of a candidate substance. Successful candidates which agonize the signal effect an increase in a metabolic or physiological output; antagonists effect a decrease. Both antagonists and agonists compete for binding between cognate and signal-generating protein.

Among successful candidates will be peptides, which mimic regions on either the signal-generating protein or the cognate, as well as nonpeptide small molecules. Due to their ease of identification, successful peptide candidates are particularly useful as participants representing the signal protein or cognate in the screening assays, in particular where binding is detected. The assay methods enable sophisticated screening of candidates. The results and uses of these methods can be combined with other assay-independent techniques for efficiently selecting leads.

The methods described herein may involve peptides derived from the cognate or signalgenerating protein. By "derived from" is meant that such peptides are either found in the cognate or signal-generating protein, or are modified by a limited number of conservative changes. Preferably the conservative changes represent less than about 20%, more preferably less than about 10%, and most preferably less than about 5% of the residues in the identified native sequence. One way to identify a suitable peptide is to compare sequences among species as is described in Example 4.

The invention is also directed to methods to screen libraries of candidate substances using the above-described methods and to peptides representative of sites on the signal-generating protein and cognate which are themselves useful in these assays as well as in other applications involving the relevant interaction.

The present invention is also directed to specific embodiments in which the above disclosed methods are employed using PKC-theta protein, and its fragments along with the cognates of PKC-theta. The fyn protein and certain fragments thereof were found to bind to PKC-theta. Utilizing this interaction and that of PKC-theta and its fragments with other cognates, the present invention provides methods to identify immunomodulators by detecting modulators of biological and pathological processes which are mediated by PKC-theta/cognate interaction. Such substances are useful in modulating an activity of the immune system, particularly the activity and differentiation of T-cells.

The present invention is thus based on using the interaction of PKC-theta with its cognates as an index to substances that modulate activities of the immune system. In the Examples, evidence is presented that the fyn protein binds to PKC-theta. This observation is important because no binding partners of PKC-theta were known prior to applicants' invention. By contrast, fyn had previously been found to be associated with several T-cell receptor affiliated proteins, including CD3 zeta chain, ZAP-70 and Grb-2. The identified fyn/PKC-theta interaction of the present invention can be used as a basis for making and identifying agents which can modulate immune responses. Competitive assays using PKC-theta and the fyn peptide, or a PKC-theta or fyn equivalent, can be used to identify compounds which block fyn/PKC-theta interaction. A PKC-theta or fyn "equivalent" represents a peptide that can mimic the interactive binding activity of PKC-theta or fyn which is derived from the amino acid sequences of the appropriate binding regions. In addition, surrogate cognates can be used which may have no structural relationship to the fyn protein, but which bind PKC-theta in a manner similar to that displayed by fyn. Additionally, peptide and protein modeling techniques can be used to study the specific interactions of the cognate partners with PKC-theta to rationally design or rationally select agents for testing. Successful agents can be used as therapeutics to inhibit or otherwise modulate immune responses.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows, diagrammatically, the known general primary sequence and domains of various PKC isoenzyme families.

FIG. 2 shows the result of experiments demonstrating that PKC mediated effects on contraction of cardiac myocytes is inhibited by a fragment of the regulatory domain of .epsilon.PKC but not by a corresponding fragment of .delta.PKC.

FIG. 3 shows the specific effect of an 8 amino acid peptide derived from the RACK-binding site in the regulatory domain of .epsilon.PKC on the contraction rate of cardiac myocytes; an analogous peptide from the .beta.PKC has no effect.

FIG. 4 shows the specific effect of peptides derived from the RACK-binding site of .beta.PKC on maturation of Xenopus oocytes.

FIG. 5 shows the effects of peptides derived from RACK1 on PKC mediated maturation of Xenopus oocytes.

FIG. 6 shows the effect of various peptides on the binding of .theta.PKC V1 fragment to RACK1 in vitro.

FIG. 7 is a photocopy of a blot obtained from a gel electrophoresis performed on proteins precipitated with antibodies raised against the V1/C1 boundary region of PKC-theta and probed with antibodies to phosphotyrosine and to fyn.

FIG. 8 shows the regions of fyn (panel A) used in a yeast two hybrid system; the resulting interactions are diagrammed in panel B. In panel B, ++ is a strong interaction (color development in <2 hr), + is a weak interaction (color development within 12 hours), and - is a null interaction (no color at 24 hours). U refers to the Unique sequence portion of fyn; SH3 and SH2 are the src homology domains; kinase is the conserved tyrosine protein kinase catalytic domain.

FIG. 9 (SEQ ID NO:1) shows a detailed map of sequence motifs present in fyn, including the WD40 repeats, smaller RACK1 homologies, and the most conserved consensus PKC phosphorylation site.

FIGS. 10A and 10B show that the T2 peptide (aa 36-46) from the V1 region of PKC-theta reduces translocation of PKC-theta, but not of PKC-beta, from the soluble to the particulate fraction after activation of Jurkat T-cells with PMA/PHA (panel A); peptide is taken up spontaneously from the medium in Jurkat T-cells. In parallel experiments, T2 peptide reduces expression of IL-2 as measured by ELISA in supernatant from stimulated cells (panel B).

FIGS. 11A and 11B is aphotocopy of a blot which shows the effect of activation by various agents on translocation of PKC-theta and PKC-beta.

FIG. 12 is a photocopy of a blot which shows the effect of TER14687 on PKC-theta translocation.

FIG. 13 shows the results of binding of peptides encoded by the fyn kinase domains and two surrogates, clones 2-10 and 2-32, to PKC-theta V1 in the absence or presence of TER14687 using a two-hybrid system assay.

FIG. 14 shows the effect of TER14687 on anti-CD3-induced production of IL4, IL5 and .gamma.IFN.

FIG. 15 shows the nucleotide sequences and encoded amino acid 2-10 (SEQ ID NO:2 and SEQ ID NO:3); 2-32 (SEQ ID NO:4 and SEQ ID NO:5); 10 (SEQ ID NO:6 and SEQ ID NO:7); 1-22 (SEQ ID NO:8 and SEQ ID NO:9); 2-18 (SEQ ID NO:10 and SEQ ID NO: 11); 2-20 (SEQ ID NO:14 and SEQ ID NO:25); and 3-1 (SEQ ID NO:12 and SEQ ID NO:13).

MODES OF CARRYING OUT THE INVENTION

General Background

PKCs represent a family of signal-generating isoenzymes, at least several of which are present in most cell types. Upon activation by a suitable agent, typically phosphatidylserine (PS) and diacylglycerol (DAG), and in some cases calcium ion, a PKC is translocated subcellularly, generally from the soluble fraction to another location in the cell that is associated with the particulate fraction. Each isoenzyme in this family apparently has one or more cognates (or RACKs) which are anchoring proteins at the appropriate locations associated with the physiological or metabolic effect of the activation of each particular isoenzyme. Thus, for example, one or a subset of PKCs contained in cardiac myocytes, when activated, results in a slowing of the contraction rate. One or a subset of PKCs contained in Xenopus oocytes, when activated, effect maturation of the egg. One or a subset of PKCs, when inhibited at the catalytic site, blocks T-lymphocyte activation. See, Birdchall et al. J Pharm. Expt'l Ther (1994) 268:922. The interaction of a particular PKC isoenzyme with its cognate RACK is required for the metabolic or physiological effect; therefore interference with this interaction will modulate that effect. Alternatively, the effect of the modulation may be agonistic if the interaction of the modulator promotes a conformational change in the signal generating partner corresponding to that normally occurring only upon the concurrent binding of activators (e.g., PS or DAG) and cognate protein, or otherwise results in signal activation.

The known PKC isoenzymes can be divided into three major groups as shown in FIG. 1. All of the isoenzymes, regardless of group, contain a protein kinase domain represented by two constant (C) and two variable (V) regions. The regions which are responsible for the enzymatic activity are highly homologous or constant; the C4 region toward the carboxy terminus is thought to contain the catalytic site; the C3 regions upstream are responsible for binding ATP.

Upstream of the protein kinase domain in each case is a regulatory domain. All of the three families, the conventional (cPKC), the novel (nPKC) and atypical (aPKC) contain variable regions at the N-terminus designated VI, and constant regions immediately downstream marked C1. The C1 regions are thought to be involved in activation by phosphatidylserine, diacylglycerol, or pharmacological mimics such as phorbol esters. The C2 region is found only in the cPKC family and is thought to be the site for activation by calcium. However, the picture may not be quite so simple as C1 regions may also be involved in calcium binding, and the atypical class respond poorly to agents such as phorbol esters.

Nevertheless, it now appears clear that sequences within the regions shown as the regulatory domain are responsible for the interaction of the particular PKC with its cognate RACK. They may also contain a RACK-mimicking region, called a pseudo-RACK site, that prevents binding of PKC to its RACK when the PKC has not been activated. This situation is analogous to the pseudosubstrate sequence which is located elsewhere in the primary sequence and blocks the catalytic site prior to activation of the PKC. It is shown hereinbelow that the relevant regions are specific for the particular isoenzyme involved in a designated signal-generation event.

In the parent applications herein, published as PCT WO 95/21252, the cognate RACK1 protein which interacts with PKC-beta (a member of the cPKC family) was cloned and the WD-40 regions putatively responsible for binding to PKC-beta were identified through structural analogy. One of these WD-40 peptides was found to induce the kinase activity of PKC in the absence of PKC activators; both this peptide and another representing a WD-40 region rendered the PKC-beta susceptible to proteolysis, a characteristic of activated PKC forms. All of these peptides were also shown to inhibit the binding of PKC-beta to RACK1. In principle, the WD-40 regions of the appropriate RACK can serve as antagonists or agonists of the signal generation associated with the corresponding PKC. As described, an assay which shows the effects of members of a library of candidate modulators on interaction between the relevant PKC and its cognate or the relevant PKC and a WD-40 domain derived from said cognate can be used as a screening assay to identify modulators of this signal pathway.

In the illustrative work described below, modulation of signal generation can be achieved by supplying, to a reaction mixture containing an appropriate PKC, either WD-40 peptides derived from the relevant RACK or pseudo-RACK peptides from the PKC regulatory domain which themselves mimic the RACK's binding domains, or a surrogate cognate and examining the effect of these peptides on binding or signal generation. Substances that interact--i.e., behave as cognates--will bind PKC and/or alter its signal generation. PKC or cognate-binding peptides derived from PKC can also be used as assay reagents in combination with the appropriate cognate to screen for modulators of the signal-generating pathway by virtue of the ability of the successful candidate to affect the binding of the cognate protein to the signal-generating protein. In general, systems which employ a PKC/cognate combination whose interaction can be observed can be used to screen for substances that affect this interaction. Thus, the interaction is observed in the presence of a candidate and in its absence and the difference in the interaction under these two conditions reflects the ability of the candidate to modulate this interaction.

A particularly important signal-generating protein/cognate system is described hereinbelow. The signaling associated with PKC-theta is particularly significant in modulating the immune system, particularly T-cell responses. Therefore, substances that modulate the interaction between PKC-theta and its cognate are also modulators of the immune response, and thus can be used in contexts where modification of an immune response is desirable, such as in the treatment of allergies and asthma, preventing allograft rejection, and in autoimmune diseases. Described below are assay systems which employ PKC-theta or a fragment thereof wherein the cognate is that normally present in a host cell, is the fyn protein or a relevant fragment thereof, or a surrogate cognate defined according to the methods of the invention. As used herein, the terms "peptide" and "protein" will be used interchangeably without regard to size. It will be apparent in some cases which is intended according to conventional terminology; in others, both possibilities may be included. Applicants draw no arbitrary dividing line between peptides and proteins in terms of length of amino acid sequence.

Thus, in summary, various possible counterpart interactions can be tested; in no case are purified components required:

In the foregoing table, any assay method that detects the interaction appropriate to the components can be used. If the interaction is intracellular, certain physiological effects on the cell can be measured. If the interaction is in vitro, direct measurement of binding of the cognate components to the signal-generating protein is probably most appropriate.

In general, the present invention is directed to screening methods to identify modulators of particular signal pathways. Each assay will involve a cognate that binds sufficiently and specifically to a catalytically active signal-generating protein, via a noncatalytic site, to permit assay in impure preparations. The interaction of these two components is observed in the presence and absence of a candidate modulator. Depending on the assay system chosen, the interaction and its modification can be observed in a variety of ways, including intracellular binding assays affecting an observable parameter; either a physiological readout, such as change in subcellular distribution, or an artificial construct, such as transcription of a reporter gene, can be used. In no case, however, are purified reagents required, although it may be convenient in some cases, for example, to utilize the peptides identified as illustrated below which represent regions of the signal-generating protein (illustrated by PKC-theta) or its cognate binding protein (represented by the fyn) that are responsible for interaction.

It will be noted that the relevant endogenous PKC and the relevant endogenous cognate can be used in observing the effects of candidate substances--e.g., the ability of the candidate to alter intracellular translocation patterns upon activation of the signal-generating protein. Alternatively, surrogate cognate proteins can be used, such as those representing a WD40 region of a RACK or cognates identified using the approach described below by virtue of their ability to bind to the relevant PKC. If a surrogate is used in an in vitro system, generally binding is the measurable interaction. As used in the present application, the term "cognate" is applied to a substance that binds sufficiently and specifically to a catalytically active signal-generating protein via a noncatalytic site, as described above. "Cognate" therefore refers both to the endogenous cognate protein (or the relevant fragment) or equivalents derived from these and to a surrogate substance which shares the above-mentioned property. Although it is required that the cognate bind to a noncatalytic site, the cognate may, in addition, be a substrate for catalysis by the signal-generating protein.

As further described below, the peptides which can be substituted for one or the other component in the assay method are themselves identifiable through conduct of the assay. Thus, the ability of a compound to affect the interaction of the cognate protein and the signal-generating protein will identify it as a useful component of the assay, as well as a modulator of the signal pathway per se. Once the appropriate components are identified, the individual labeled components could be used to assess the level of binding. The labeled component may represent a region of the signal-generating peptide measured against a composition containing the cognate protein or, conversely, a component representing a portion of the cognate protein measured against the composition containing the signal-generating protein. These compositions may be whole cells or cell-free extracts or partially purified extracts.

It will be apparent that when a nonendogenous substance is chosen as one component of the assay, the screening tests are preferably performed by measuring only binding per se.

Alternatively, both the signal-generating protein and the cognate protein may be contained in a crude preparation and the method for assessing their interaction may include measuring localization of the signal-generating protein within the preparation per se or measuring a metabolic effect of the interaction, such as, for example, maturation of Xenopus oocytes or effect on the contraction rate of cardiac myocytes. The particular method of assessing the interaction will, of course, be appropriate to the partners in the interaction, and can readily be ascertained by taking advantage of the specificity of the signal pathways and their components as illustrated below.

In addition, the "two-hybrid" system may be used to effect interaction between a PKC or a fragment thereof and its cognate, and the effect of a candidate on this interaction can be observed. The "two-hybrid" system is described in U.S. Pat. No. 5,283,173, incorporated herein by reference. Briefly, as applied to the present invention, a recombinant host, typically yeast, is transformed with two expression systems each encoding a fusion protein. One fusion protein contains a portion of a transcription-activating factor fused to a PKC or cognate-binding fragment thereof, the other fusion protein contains the complementary portion of the transcription-activating portion fused to the cognate. Typically, the transcription-activating factor is an activator for RNA polymerase, and one portion represents the DNA-binding portion, the other the activator for the polymerase. When the cognate and PKC bind, the two portions of the transcription factor are brought into sufficient proximity that they are able to perform the function of activating transcription. The "two-hybrid" assay thus, also, will include a reporter expression system which is activated by the completed transcription factor to produce a reporter protein, such as .beta.-galactosidase or chloramphenicol acetyl transferase. As defined herein, "two-hybrid assay" refers to this general approach.

For convenience, the assays to identify modulating candidate compounds will sometimes be described as measuring the effect of the candidate on the "binding" of the counterpart components in the reaction mixture. It will be understood that in the instance where both the cognate protein and the signal-generating protein are the active components of the composition participating in the assay, binding may be measured not only directly, but also by the resulting metabolic or physiological effects.

The fyn/PKC-theta Interaction

It has been shown hereinbelow that the signal generation mediated by PKC-theta is significant in regulating the activation of T-cells. Thus, modification of PKC-theta mediated signaling will result in modification of the immune response. The substances that can be identified as modulating this signaling pathway are thus useful as immunomodulating agents in preventing allergic reactions or reducing their severity, or in treating asthma, in ameliorating the effects of autoimmune disease, and in reducing the risk of rejection of transplants. It has also been demonstrated below that PKC-theta interacts endogenously with the fyn protein which serves as at least one of its endogenous cognates. Thus, the PKC-theta/fyn interaction per se can be used as an assay system for identifying candidate substances that will behave as immunomodulators.

In one form of the assay, the PKC-theta signaling protein and the fyn cognate occur endogenously in a test cell, and the effect of the candidate on the PKC signaling function can be measured directly. As illustrated below, translocation of PKC-theta occurs on activation and thus binds to the endogenous cognate protein. The effect of a candidate substance on this translocation can be directly determined as a measure of its affect on interaction of PKC-theta with its cognate.

Alternatively, assays can be constructed in vitro or in vivo involving PKC-theta and fyn directly. Specifically, to identify an agent which modulates fyn/PKC-theta interaction, the fyn protein, a fragment of the fyn protein containing the domain which binds to PKC-theta or a fusion protein containing the domain of the fyn protein which binds to PKC-theta, is provided in the same environment with PKC-theta, a fragment of PKC-theta containing the fyn binding domain or a fusion protein containing the fyn binding domain, wherein the environment is such that ordinarily the signal-generating protein (PKC-theta) and the cognate (fyn) would interact. The environment is provided in the presence and in the absence of a substance to be tested. Differences in interaction between the signaling protein and the cognate in the presence and absence of the substance to be tested are compared; agents that block or otherwise affect the interaction can be identified by determining the differences in interaction between these two circumstances.

The fyn protein represents a family of splicing variants, and by "fyn" is meant any of these spliced variants and peptides which bind PKC-theta which are derived from them.

As used herein, an agent is said to block or decrease fyn/PKC-theta binding when the presence of the agent prevents or reduces the amount of association of the PKC-theta peptide with the fyn peptide. One class of agents will reduce or block the association by binding to the PKC-theta peptide while another class of agents will reduce or block the association by binding to the fyn peptide. Two examples of the first class of agent include antibodies which bind to the PKC-theta peptide and block the fyn binding site on PKC-theta and peptides which contain the PKC-theta binding site found on fyn. Two examples of the second class of agents include antibodies which bind to the fyn peptide and block the PKC-theta binding site on fyn and peptides which contain the fyn binding site found on PKC-theta. Other types of interaction may also be envisioned.

The fyn peptide used in the present method can either be the entire fyn protein whose amino acid sequence is known in the art, a fragment of the fyn peptide which binds the PKC-theta, or a small region thereof that retains the binding activity or a protein which contains the PKC-theta binding site of fyn, such as a fusion protein containing the appropriate fyn sequence. Alternatively, the fyn peptide can contain more than one copies of the fyn sequence, such as in a palindromic or tandem repeat. A cell or virus expressing the fyn peptide can also be used.

As an alternative to compounds containing the fyn sequence, agents identified in the present method can be substituted for the fyn peptide. For example, an agent which is found to block fyn/PKC-theta binding by binding to PKC-theta can be used in place of the fyn peptide.

The PKC-theta peptide used in the present method can be any member of the PKC-theta family of proteins so long as the member binds the fyn peptide. As used herein, a PKC-theta family member refers to proteins currently known in the art which are members of the PKC-theta family of proteins (for a review see Baier et al J Biol Chem (1993) 268(7):4997-5004 and Baier et al., Eur J Biochem (1995) 225(1): 195-203). These include PKC-theta isolated from organisms such as humans, mice, etc., as well as the various splice forms of PKC-theta found in each organism. The PKC-theta family member can be used in its entirety or a fragment of the PKC-theta protein which contains the fyn binding site can be used. The preferred fragment will be derived from the V1 region of PKC-theta; the strongly isozyme-specific regions V3 or V5 also can be used. Alternatively, a cell or virus expressing the PKC-theta, or PKC-theta fragment, can be used.

The fyn and PKC-theta peptides used in the present invention can be used in a variety of forms. The peptides can be used in a highly purified form, free of naturally occurring contaminants. Alternatively, a crude preparation containing a mixture of cellular components as well as the fyn and PKC-theta peptides can be used. Further, the fyn or PKC-theta peptides can be isolated from cells which naturally express these peptides, from cells which have been altered, using recombinant methods, to express these peptides, or can be synthesized using standard peptide synthesis methods. So long as the association of the PKC-theta peptide with the agent to be tested and/or the fyn peptide can be identified in the sample, the fyn and PKC-theta peptides are in a suitable form for use in the above described assay.

The fyn and/or PKC-theta peptides can additionally be modified to contain a detectable label or signal generation system to facilitate detection. Methods for attaching agents such as fluorescence tags or fluorescence polarization and secondary labeling agents such as biotin, are well known in the art.

A variety of art-known methods can be adapted and employed to detect whether an agent blocks or reduces the interaction of the fyn peptide with the PKC-theta peptide. Such methods include, but are not limited to, assays which employ a solid support, assays in solution phase, assays performed in a gel-type media, and assays which use a combination of these environments. An example of a solid phase assay would be one in which one or both of the fyn and PKC-theta peptides are immobilized on a solid support and is incubated in a solution phase with the agent to be tested and the other peptide of the fyn/PKC-theta pair. A secondary detection means, such as an antibody, is then used to determine the amount of the second peptide which binds to the immobilized peptide. Alternatively, the second peptide of the fyn/PKC-theta pair can be detectably labeled and its binding to the immobilized first peptide is directly assessed. One format which is preferably suitable for a solid phase based assay is immobilization of one of the peptides in a 96-well micro-titer plate. Such titer plates provide an efficient assay format for rapidly processing multiple samples.

Alternatively, both peptides of the fyn/PKC-theta binding pair can be in solution. After mixing, the binding of the fyn peptide to the PKC-theta peptide can be detected using a variety of methods, for example detecting mobility shifts using electrophoretic means. One skilled in the art can readily appreciate how numerous assay-type formats which are known in the art for use in competitive assays can be modified to use the fyn/PKC-theta peptide pair.

As described above, the binding of PKC-theta and fyn or a fyn substitute can be detected using the two-hybrid assay system. The "fyn substitute" can be any alternative peptide or other molecule which is found in the assays of the invention to bind to PKC-theta so as to interfere with the action of PKC-theta with fyn or other endogenous cognates. Thus, the discussion herein is not to be considered limited to PKC-theta/fyn interaction per se, but includes interactions of PKC-theta with any cognate.

Direct binding to the PKC-theta peptide or the fyn peptide can, but need not, be used as first step in identifying agents which block fyn/PKC-theta interaction. In such methods, agents are first screened for the ability to bind to the PKC-theta or fyn peptides. Agents which bind to either of the two peptides are then screened for the ability to block fyn/PKC-theta interaction, or for the ability to modulate a function of the immune system.

Agents which are assayed in the above methods can be randomly selected or rationally selected or designed. As used herein, an agent is said to be randomly selected when the agent is chosen randomly without considering the specific sequences involved in the association of the fyn peptide with the PKC-theta peptide. An example of randomly selected agents is the use of a chemical library or a peptide combinatorial library.

As used herein, an agent is said to be rationally selected or designed when the agent is chosen on a nonrandom basis which takes into account the sequence of the target site and/or its conformation in connection with the agent's action. As described above, two sites of actions for agents of the present invention are the fyn peptide and the PKC-theta peptide. Agents can be rationally selected or rationally designed by utilizing the peptide sequences which make up the contact sites of the fyn/PKC-theta pair. For example, a rationally selected peptide agent can be a peptide whose amino acid sequence is identical to the fyn contact site found on PKC-theta. Such an agent will reduce or block the association of fyn with PKC-theta by binding to fyn.

The agents of the present invention can be peptides, small molecules, vitamin derivatives, as well as carbohydrates. A skilled artisan can readily recognize that there is no limit as to the structural nature of the agents of the present invention or those used in the present assay methods.

As provided above, one class of agents of the present invention is that of peptide agents whose amino acid sequences are chosen based on the amino acid sequence of fyn and in particular the PKC-theta contact site found on fyn, while a second class of agents is that of peptide agents whose amino acid sequences are chosen based on the amino acid sequence of PKC-theta and in particular the fyn contact site found on PKC-theta. The fyn contact site on PKC-theta and the PKC-theta contact site on fyn can readily be determined using art-known methodologies. For example, tryptic digestion of the PKC-theta protein can be performed and the various fragments of PKC-theta can be tested for their ability to bind the fyn peptide. Alternatively, a modification of a bind and chew assay can be used in which the fyn and PKC-theta peptides are allowed to interact and the interactive pair is subject to protein digestion. Regions of the PKC-theta peptide which are contacted by the fyn peptide will be protected from digestion and can be later characterized to determine the amino acid sequence which is bound and protected. Alternative assays which detect binding of the various fragments can also be used.

All of the peptide agents of the invention, when an amino acid forms the C-terminus, may be in the form of the pharmaceutically acceptable salts or esters. Salts may be, for example, Na.sup.+, K.sup.+, Ca.sup.+2, Mg.sup.+2 and the like; the esters are generally those of alcohols of 1-6 C. In all of the peptides of the invention, one or more amide linkages (--CO--NH--) may optionally be replaced with another linkage which is an isostere such as --CH.sub.2 NH--, --CH.sub.2 S--, --CH.sub.2 CH.sub.2, --CH.dbd.CH-- (cis and trans), --COCH.sub.2 --, --CH(OH)CH.sub.2 -- and --CH.sub.2 SO--. This replacement can be made by methods known in the art. Alternative peptide linking moieties can also be used to decrease the rate of degradation of peptide based agents. The following references describe preparation of peptide analogs which include these alternative-linking moieties: Spatola, A. F., Vega Data (March 1983), Vol. 1, Issue 3, "Peptide Backbone Modifications" (general review); Spatola, A. F., in "Chemistry and Biochemistry of Amino Acids Peptides and Proteins," B. Weinstein, eds., Marcel Dekker, New York, p. 267 (1983) (general review); Spatola, A. F., et al., Life Sci (1986) 38:1243-1249 (--CH.sub.2 --S); Hann, M. M., J Chem Soc Perkin Trans I (1982) 307-314 (1982) (--CH--CH--, cis and trans); Almquist, R. G., et al., J Med Chem (1980) 23:1392-1398 (--COCH.sub.2 --); Jennings-White, C., et al., Tetrahedron Lett (1982) 23:2533 (--COCH.sub.2 --); Holladay, M. W., et al., Tetrahedron Lett (1983) 24:4401-4404 (--C(OH)CH.sub.2 --); and Hruby, V. J., Life Sci (1982) 31:189-199 (--CH.sub.2 --S--).

In addition to analogs which contain isosteres in place of peptide linkages, the peptides or proteins of the invention include peptide mimetics in general, such as those described by Olson, G. L. et al. J Med Chem (1993) 36:3039-3049 and retro-inverso type peptides as described by Chorev, M. et al. Science (1979) 204:1210-1212; and Pallai, P. V. et al., Int J Pept Protein Res (1983) 21:84-92.

Another class of agents of the present invention is that of antibodies immunoreactive with critical positions of the fyn protein or with the PKC-theta protein. Since the target for action of the agents of the present invention is within a cell (cell signaling involved in fyn/PKC-theta interaction), antibody agents are most useful in immunodiagnostic methods and find use as substitutes for either the fyn or PKC-theta peptides in the present methods. However, using currently available recombinant technologies, antibodies can be produced intracellularly, and thus can participate in intracellular forms of the assays as well.

Antibody agents are obtained by immunization of suitable mammalian subjects with peptides, containing as antigenic regions, those portions of the fyn or PKC-theta proteins which are intended to be targeted by the antibodies. Critical regions include, but are not limited to, the contact sites involved in the association of fyn with PKC-theta and sites which provide steric interference with the contact sites upon binding.

Antibody agents are prepared by immunizing suitable mammalian hosts in appropriate immunization protocols using the peptide haptens alone, if they are of sufficient length, or, if desired, or if required to enhance immunogenicity, conjugated to suitable carriers. Methods for preparing immunogenic conjugates with carriers such as BSA, KLH, or other carrier proteins are well known in the art. In some circumstances, direct conjugation using, for example, carbodiimide reagents may be effective; in other instances linking reagents such as those supplied by Pierce Chemical Co., Rockford, Ill., may be desirable to provide accessibility to the hapten. The hapten peptides can be extended at the amino or carboxy terminus with a cysteine residue or interspersed with cysteine residues, for example, to facilitate linking to carrier. Administration of the immunogens is conducted generally by injection over a suitable time period and with use of suitable adjuvants, as is generally understood in the art. During the immunization schedule, titers of antibodies are taken to determine adequacy of antibody formation.

While the polyclonal antisera produced in this way may be satisfactory for some applications, for pharmaceutical compositions, use of monoclonal preparations is preferred. Immortalized cell lines which secrete the desired monoclonal antibodies may be prepared using the standard method of Kohler and Milstein or modifications which effect immortalization of lymphocytes or spleen cells, as is generally known. (See Harlow: Antibodies Cold Spring Harbor Press NY 1989) The immortalized cell lines secreting the desired antibodies are screened by immunoassay in which the antigen is the peptide hapten or is the PKC-theta or fyn peptide. When the appropriate immortalized cell culture secreting the desired antibody is identified, the cells can be cultured either in vitro or by production in ascites fluid.

The desired monoclonal antibodies are then recovered from the culture supernatant or from the ascites supernatant. Fragments of the monoclonals or the polyclonal antisera which contain the immunologically significant portion can be used as antagonists, as well as the intact antibodies. Use of immunologically reactive fragments, such as the Fab, Fab', of F(ab').sub.2 fragments is often preferable, especially in a therapeutic context, as these fragments are generally less immunogenic than the whole immunoglobulin. The antibodies or fragments may also be produced, using current technology, by recombinant means. Regions that bind specifically to the desired regions of receptor can also be produced in the context of chimeras with multiple species origin.

It is, of course, unnecessary to conduct initial production of the antibodies through immunization and isolation of immortalized cells that secrete monoclonals. Techniques are available for cloning immunoglobulin genes so that production can be effected through de novo recombinant synthesis. Regardless of whether the genes originate in immortalized B cells or result from cloning efforts, manipulation of the genes permits redesign of the immunoglobulins, including production of single-chain immunoglobulins such as F.sub.v fragments.

Uses for Substances which Modulate Interaction of Cognate with PKC-theta

Interaction between PKC-theta and its cognate has been implicated in modulating a variety of biological responses. According to the present invention, this interaction results in modulation of the immune system, particularly those activities involving T-cell activity. Therefore, substances which affect PKC-theta interaction with its cognate can be used to modulate activities of the immune system.

Specifically, immune system activity, such as T-cell mediated responses, can be modulated by administering to a subject a substance which affects the interaction of an appropriate cognate with PKC-theta. The subject can be any vertebrate in need of modulation of immune activity. These substances are particularly useful in treating human subjects.

Immune system activity refers to the wide variety of cellular events in which cells of the immune system participate. Examples of situations where it is desirable to modulate such activity include, but are not limited to, transplant surgery, autoimmune disorders, and response to allergens. In each of these situations, it is desirable selectively to reduce T-cell responsiveness.

A substance modulates an immune system activity when it reduces the severity of a pathological condition mediated by the immune system, or when it affects either positively or negatively the normal immune activity of the subject. For example, an agent is said to modulate the immune system activity involved in graft rejection when it reduces the rate of onset of graft rejection or reduces the severity of graft rejection. Other effects on the immune system can be determined more precisely. For example, the resultant effect may be to diminish the production of cytokines such as IL-4 or IL-5 as illustrated below, preferably with no effect on .gamma.-IFN production.

Administration of Agents which Modulate Immune System Activity

The agents of the present invention can be provided alone, in combination with another agent that modulates a function of the immune system, or in combination with drugs having additional physiological effects. For example, a substance of the present invention that reduces T-cell activity can be administered in combination with other immunosuppressive agents, and/or with another substance so identified. As used herein, two agents are said to be administered in combination when the two agents are administered simultaneously or are administered independently in a fashion such that the agents will act at the same time.

The agents of the present invention can be administered via parenteral, subcutaneous, intravenous, intramuscular, intraperitoneal, transdermal, or buccal routes. Alternatively, or concurrently, administration may be by the oral route, or by inhalation. The dosage administered will be dependent upon the age, health, and weight of the recipient, kind of concurrent treatment, if any, frequency of treatment, and the nature of the effect desired.

The present invention further provides compositions containing one or more agents of the present invention. While individual needs vary, determination of optimal ranges of effective amounts of each component is within the skill of the art. Typical dosages are in the range of 0.01-50 mg/kg body weight. Preferred dosages are in the range of 0.1-10 mg/kg; most preferably 0.1-1 mg/kg.

In addition to the pharmacologically active agent, a composition comprising an agent of the present invention may contain suitable pharmaceutically acceptable carriers such as excipients and auxiliaries which facilitate processing of the active compounds into preparations which can be used pharmaceutically for delivery to the site of action. Suitable formulations for parenteral administration include aqueous solutions of the active compounds in water-soluble form, for example, water-soluble salts. In addition, suspensions of the active compounds as appropriate oily injection suspensions may be administered. Suitable lipophilic solvents or vehicles include fatty oils, for example, sesame oil, or synthetic fatty acid esters, for example, ethyl oleate or triglycerides. Aqueous injection suspensions may contain substances which increase the viscosity of the suspension include, for example, sodium carboxymethyl cellulose, sorbitol, and/or dextran. Optionally, the suspension may also contain stabilizers. Liposomes can also be used to encapsulate the agent for delivery into the cell.

The pharmaceutical formulation for systemic administration according to the invention may be formulated for enteral, parenteral or topical administration. Indeed, all three types of formulation may be used simultaneously to achieve systemic administration of the active ingredient.

Suitable formulations for oral administration include hard or soft gelatin capsules, pills, tablets, including coated tablets, elixirs, suspensions, syrups or inhalations and controlled release forms thereof.

Suitable formulations for administration by inhalation include metered dose inhalers and dry powder devices. For nasal absorption aqueous and nonaqueous suspensions or dry powders may be used.

Methods for Targeting

The effective affinity of the substances of the present invention can be increased by covalently linking the agent to a second agent which also binds either PKC-theta or its cognate. Such a second agent will bind to another site on either the cognate or PKC-theta molecule and bring the first agent into close proximity to the target site, augmenting the overall avidity. Such second agents can be, but are not limited to, antibody and peptide agents. The second agent can be covalently attached to the first agent using art-known methods. Methods which employ linkers are particularly well suited for this use.

Without further description, it is believed that one of ordinary skill in the art can, using the preceding description and the following illustrative examples, practice the claimed invention. The following working examples illustrate preferred embodiments of the present invention, and are not to be construed as limiting.

EXAMPLE 1

Specificity of Negative Chronotropy for .epsilon.PKC Translocation

Neonatal rat cardiac myocytes were used in this assay. These cells, when prepared in culture, exhibit contractions at approximately the rate of 40-50/15 sec., and it is known that the phorbol ester, 4-.beta. phorbol 12-myristate-13-acetate (PMA) reduces the contraction rate (Johnson, J. A. et al., Circ Res (1995) 76:654-653). Previous work has also shown that treating cardiac myocytes with PMA or with norepinephrine (NE) causes translocation of .alpha.PKC to the nuclear boundary, .beta.IPKC to the interior of the nuclei, .delta.PKC to the fibrillar and perinuclear structures, and .epsilon.PKC to cross-striated structures (Disatnik, M-H. et al., Exp Cell Res (1994) 210:287-297). It has also been shown that exogenously added activated PKCs bind similarly (Mochly-Rosen, D. et al., Molec Biol Cell (1990)1:693-706). Since the location to which the various isoenzymes are translocated are different, it has been suggested that the variable regions specific for each isoenzyme (Nishizuka, Y., Nature (1988) 334:661-665) should contain at least part of the specific RACK binding site (Disatnik, M-H. et al., Exp Cell Res (1994) 210:287-297). Furthermore, it has been suggested that the V1 region of .epsilon.PKC determines its substrate specificity (Pears, C. et al., Biochem J (1991) 276:257-260).

To show that only translocation of the corresponding isoenzyme is inhibited by one of its fragments--e.g., only translocation of .epsilon.PKC is inhibited by an .epsilon.PKC-V1 fragment, cells cultured on chamber slides were permeabilized with saponin (50 .mu.g/ml) in the absence or presence of 100 .mu.g/ml rat recombinant .epsilon.PKC-V1 or .delta.PKC-V1 fragments containing amino acids 2-144 in each case. Cellular functions, including cell viability, spontaneous and stimulated contraction rates, gene expression and hypertrophy are unaffected by the saponin treatment.

These fragments were prepared by amplifying the relevant portion of the gene from a cDNA library (Stratagene). A FLAG.TM. epitope (DYKDDDK) (SEQ ID NO:16) was engineered at the 5' end of the fragment and the 0.45 kb PCR fragment was subcloned into pMAL-C2 vector (New England Biolabs) for overexpression as a fusion protein with maltose binding protein in E. coli. Protein purification and Factor Xa proteolysis of the fusion proteins was as described by Ron, D. et al., Proc Natl Acad Sci USA (1994) 91:839-843.

The intracellular concentration of each fragment was approximately 300 nM or about 3% of the extracellular concentration as determined by quantitative Western blot of washed and extracted cells.

After the .epsilon.PKC-V1 or .delta.PKC-V1 fragments were administered by permeabilization, the cells were incubated with either 4-.alpha. or 4-.beta. PMA. (4.alpha. PMA is not active and is used as a control.) The cells were then fixed with methanol and acetone and PKC isoenzyme localization was determined by immunofluorescence; the antisera used to detect .delta.PKC and .epsilon.PKC do not recognize the administered fragments. Multiple fields of cells for each treatment group and for PKC isoenzymes .alpha., .beta.I, .delta., and .epsilon. were observed and the data were presented as a percentage of cells having the tested enzyme at the activated site. When the cells were treated with 100 nM PMA for five minutes, it was apparent that neither .delta.PKC-V1 nor .epsilon.PKC-V1 had any effect on translocation of the .alpha. or .beta. isoenzymes whereas each of the .delta. and .epsilon. fragments specifically inhibited the translocation of the corresponding isoenzyme, but not the other isoenzyme. An additional experiment measuring translocation of .epsilon.PKC at the much lower level of 3 nM PMA also showed complete inhibition by the .epsilon. fragment. It has previously been shown that 3 nM PMA is only marginally effective in translocation of PKC isoenzymes other than the .epsilon. form (Johnson, J. A. et al., Circ Res (1995) 76:654-663).

The localization of .epsilon.PKC to the cross-striated structures suggested that the .epsilon. isoenzyme might mediate the effect of PMA on the contraction rate. Cells were cultured and permeabilized with saponin in the presence or absence of 150 .mu.g/ml .delta. or .epsilon.PKC-V1 fragments described above. Basal contraction rates were monitored for 10 min, and the cells were then treated with 3 nM PMA. The rate of contraction was monitored over the next 20 minutes. The results are shown in FIG. 2.

As shown, in cells where no fragment was added, the contraction rate is reduced almost to zero within 15 min of the addition of PMA. Similarly, in cells where the .delta. PKC fragment is added, the contraction rate is thus reduced. However, in cells where the .epsilon.PKC fragment was added, the contraction rate is maintained. Thus, the .epsilon.PKC-V1 fragments specifically prevented PMA-induced inhibition of spontaneous contraction. These data, combined with the data described above with respect to translocation and the fact that the .epsilon.PKC-V1 fragment does not affect the catalytic activity of .epsilon.PKC in vitro, demonstrate that the translocation of .epsilon.PKC is an essential step in signaling the chronotropic effect of PMA and that this signaling is inhibited by a fragment containing the V1 region.

The effect of PMA in reducing the contraction rate can be mimicked by controlling the .alpha.1 and .beta.1 adrenergic receptors of the myocytes, providing a more physiologically relevant phenomenon. If both the .alpha.1 and .beta.1 receptors are activated with NE, an increase in contraction rate occurs; when both receptors are inhibited, NE no longer has this effect. If the .alpha.1 receptor is inhibited alone by prazosin, the initial increase in contraction rate is higher; if the .beta.1 receptor alone is inhibited, the contraction rate decreases.

When either the .delta. or .epsilon. fragments described above is substituted for the known inhibitors of the .alpha.1 and .beta.1 receptors, the behavior of the cells in response to NE is unaffected by the presence of the .delta. fragment; however, addition of the .epsilon. fragment gives a response similar to that obtained in the presence of prazosin. These data are consistent with the role of the .epsilon. fragment in controlling contraction rate since the .alpha.1 receptor (inhibited by prazosin) mediates PKC translocation.

FIG. 3 shows the results of a similar experiment using stimulation with 3 nM PMA, and providing peptides of less than 10 residues that block localization of .epsilon.PKC and .beta.PKC using a 0.1 .mu.M .epsilon.PKC-derived peptide .epsilon.V1-2 (sequence EAVSLKPT) (SEQ ID NO:17) or 20 .mu.M of a .beta.PKC-derived peptide .beta.C2-4 (sequence SLNPEWNET) (SEQ ID NO:18). As shown in FIG. 3, stimulation with 3 nM PMA without adding peptides to the permeabilized cells or in the presence of 20 .mu.M of the .beta.PKC localization inhibitor results in negative chronotropy as above.

EXAMPLE 2

Specific Inhibition of .beta.PKC Translocation by C2-Derived Peptides

The parent application herein described binding sites on a particular RACK, RACK1, which are responsible for binding .beta.PKC. It is recognized that if the binding site on PKC is identified, peptides that mimic this binding site could also serve as modulators of .beta.PKC translocation and function. Furthermore, it should be noted that PKC may itself contain pseudo-RACK peptide sequences that mimic the binding sites on RACK and regulate the exposure of the binding site for RACK on PKC. The following experiments do not distinguish between these possibilities; nevertheless, whichever function on the PKC sequence is represented, mimics of the sequence will be effective modulators of the relevant signal pathway.

The cPKC class of isozymes comprises the only members of the PKC general family that contains C2 regions. Other C2-containing proteins such as synaptotagmin and phospholipase C.gamma. also bind to a mixture of RACKs prepared from cell particulate fractions. It has also been demonstrated that recombinant fragments of synaptotagmin containing the C2 homologous region bind to mixtures of RACKs and inhibit PKC binding to RACKs (Mochly-Rosen, D. el al., Biochemistry (1992) 31:8120-8124).

The following experiments demonstrate that certain peptides residing in the C2 region of .beta.PKC are able to inhibit translocation of .beta.PKC and the maturation of Xenopus oocytes. Ron, D. et al., J Biol Chem (1995) 270:24180-24187.

The following .beta.PKC-derived peptides were prepared:

.beta.C2-1 (SEQ ID NO:19): KQKTKTIK (210-217);

.beta.C2-2 (SEQ ID NO:20): MIDPNGLSDPYVKL (186-198);

.beta.C2-3 (SEQ ID NO:21): IPDPKSE (201-207);

.beta.C2-4 (SEQ ID NO:18): SLNPEWNET (218-226);

Scrambled .beta.C2-1 (SEQ ID NO:22): TKQKKITK;

Control Peptide (SEQ ID NO:23): LQKAGVDG (266-271).

Recombinantly produced fragments of .beta.PKC were expressed as fusion proteins with GST: Fusion L9 includes the V1 region, the pseudosubstrate sequence, and the C1 and V2 regions (residues 3-182) of .beta.PKC. L10 includes the V1 region, the pseudosubstrate sequence and the first cysteine repeat from the C1 region, as well as the entire C2 and V3 regions (residues 3-76 and 143-339). The numbering is as described in Luo, J-H. et al., J Biol Chem (1993) 248:3715-3719.

Standard overlay assays were performed by blotting RACKI onto nitrocellulose as described by Mochly-Rosen, D. et al., Proc Natl Acad Sci USA (1991) 88:3997-4000. Strips of the nitrocellulose sheet containing 0.1-1 .mu.g RACK1 per strip were incubated in overlay buffer with or without the test fragment added at approximately 10 .mu.M. Addition was in the presence or absence of 50 .mu.g/ml phosphatidyl serine (PS) and 1 mM calcium. The mixture was further incubated for 30 min at room temperature. The strips were then washed and binding of fragment of L9 or L10 to RACK1 was detected with anti-GST polyclonal antibodies followed by labeling with anti-rabbit horseradish peroxidase-linked antibodies and development by addition of substrate.

Using this assay, L10, but not L9 was found to bind RACK1. The PKC activators phosphatidyl serine and calcium did not increase the binding of L10 to RACK1, although these activators are necessary for the binding of intact PKC to RACK1. Thus, these data are consistent with the suggestion that the PKC activators are required to expose the RACK binding site in the intact PKC; this site is already exposed in the C2-containing fragment L10.

To determine whether L10 would inhibit the binding of intact .beta.PKC to RACK1, RACK1 was immobilized on an amylose column and .beta.PKC binding in the presence of PS, DAG and calcium and in the presence of L10 or L9 was determined. In the presence of L10, .beta.PKC binding to RACK1 was completely inhibited; however, this was not true of L9. Similar results were obtained in an overlay assay.

Similar overlay assays were conducted using the above-listed peptides as candidate inhibitors for the binding of L10 to RACK1. The C2-derived peptides .beta.C2-1, .beta.C2-2 and .beta.C2-4 peptides were successful in inhibiting binding of L10 to RACK1; however .beta.C2-3 and scrambled .beta.C2-1 were not.

In addition to the foregoing cell-free assays, the association of .beta.PKC with RACK1 and the ability of peptides derived from the C2 region to interrupt this interaction was tested in rat neonatal cardiac myocytes in culture. The presence of RACK1 in these cells was confirmed by immunostaining. RACK1 was found at perinuclear structures and throughout the cytosol. Treating with NE or PMA did not alter these locations. It was also demonstrated that activated .beta.II PKC, but not C2-less isoenzymes .delta. or .epsilon.PKC, colocalized with RACK1.

The C2-derived peptides that had been shown to inhibit .beta.PKC binding to RACK1 in vitro were then tested for their ability to inhibit activation-induced translocation in myocytes.

The myocytes were exposed to 100 nM PMA for 15 min after transient permeabilization with saponin (50 .mu.g/ml) in the presence and absence of the test peptides. 80% of the cells that had not been treated with peptides showed localization of .beta.1 PKC to perinuclear structures. However, when .beta.C2-1, .beta.C2-2 or .beta.C2-4 at 10 .mu.M extracellular concentration had been supplied to the permeabilized cells, translocation of both .beta.I PKC and .beta.II PKC isoenzymes was inhibited by 65-95%. .beta.C2-4 was the most effective. Control peptides described above did not affect translocation.

Consistent with the results in Example 1, treating nonpermeabilized cardiac myocytes with 100 nM PMA resulted in translocation of .epsilon.PKC from the nucleus to the perinuclear and cross-striated structures and of .delta.PKC from the perinuclear and fibrilar cytosolic structures in 80% and 90% of the cells respectively. Permeabilization and treatment of the cells with the C2 peptides derived from .beta.PKC had no effect on the translocation of these C2-less isozymes.

While the chronotropy of myocytes is not affected by .beta.PKC isoenzymes, the insulin-induced maturation of Xenopus oocytes is mediated by the .beta. form. Insulin treatment of these oocytes results in translocation of .beta.PKC and maturation is delayed by the PKC-specific catalytic inhibitor pseudosubstrate peptide. PKC translocation is blocked by injection of purified RACKs or a peptide corresponding to the PKC binding site on RACKs. (Smith, B. L. el al., Biochem Biophys Res Commun (1 992) 188:1235-1240; Ron, D. et al., J Biol Chem (1 994) 269:21395-21398).

Accordingly, the maturation of Xenopus oocytes was used as an alternative assay system to test the function of the peptides derived from the C2 region described above. In this assay, oocytes were injected with 50 .mu.M of the test peptide one hour before insulin treatment (8.25 .mu.g/ml). Insulin-induced oocyte maturation was then determined by monitoring the appearance of a white spot in the animal pole of the oocyte that is indicative of germinal vesicle breakdown in maturation. 10-15 oocytes were included per assay and oocytes were scored for 35 hours after treatment. As expected, .beta.C2-1, .beta.C2-2 and .beta.C2-4 supplied in the range of 5 .mu.M-500 .mu.M significantly delayed oocyte maturation in a dose-dependent manner. The control peptides did not. The association of this effect with the prevention of translocation of .beta.PKC to the particulate fraction in Xenopus oocytes was confirmed in a separate experiment. The peptide .beta.C2-4 inhibited .beta.PKC translocation but not .theta.PKC in Jurkat T-cells. FIG. 4 shows the effect of these various peptides on Xenopus oocyte maturation.

EXAMPLE 3

Agonist Effect of Interacting Peptides

The oocyte maturation assay described above was also used to test the effect of various peptides derived from the PKC/RACK1 pair. Peptide I, derived from RACK1, as expected, inhibits the maturation of Xenopus oocytes presumably by interfering with the binding of .beta.PKC1 to RACK1. On the other hand, a short peptide, rVI derived from the sixth WD-40 repeat in RACK1 enhances maturation, both in the presence and absence of insulin. Ron, D., Mochly-Rosen, D., J. Biol. Chem. (1994) 269:21395-21398 These results are shown in FIGS. 5 and 5b. The rVI peptide is believed to interfere with the RACK-mimicking site on PKC which normally covers the RACK-binding site in the absence of activation.

EXAMPLE 4

Interaction Peptides Derived from OPKC

.theta.PKC is a member of the NPKC family and lacks a C2 region. Comparison of variable sequences with .epsilon.PKC and other isozymes reveals regions of maximum disparity. Of these regions, some are strongly conserved across vast phylogenetic spans, e.g., from mammals to the invertebrate Aplysia. Isozyme specific sequences that are strongly conserved by evolution are probable sites for binding cognate proteins. Comparing .delta.PKC to .theta.PKC in the analogous region allowed identification of a .theta.-specific peptide expected to interfere with PKC binding to a RACK. Peptides with these characteristics from the V1 region of .theta.PKC were prepared and tested for their ability to inhibit the binding of .theta.PKC V1 fragment to RACK1 in vitro. The results are shown in FIG. 6. Of a multiplicity of peptides tested, both from other regions of the .theta.PKC isoenzyme and from alternative isoenzymes in the family, only .theta.V1 derived peptides T1 and T2, having the amino acid sequences GLSNFDCG (SEQ ID NO:24) (.theta.PKC residues 8-15) and YVESENGQMYI (SEQ ID NO:25) (.theta.PKC residues 36-46), respectively, were able to affect the interaction negatively. As expected, peptides rIII and rVI derived from the WD-40 regions of RACK1 were also effective.

The peptides T1 and T2 could inhibit the translocation of .theta.PKC to the cellular particulate fraction in Jurkat T-cells. The specificity of this interaction is shown in FIG. 11A which shows inhibition by the .theta.PKC V1 peptide (T2) of the translocation of .theta.PKC but not of .beta.PKC.

Sequences derived from other PKC regions.

V3 region. The V3 or hinge region separates the regulatory and catalytic domains. This region contains the sites of proteolysis by trypsin and calpain. The lack of conservation of the V3 domain between the different PKC isozymes suggests that this section may also be at least, in part, involved in targeting the PKC isozymes to their anchoring proteins. Furthermore, it has been demonstrated that regions within the V3 of .alpha.PKC mediate the translocation of that isozyme to the nucleus (James G., and Olson E. J. Cell Biol. 116:863-873, 1992). The V3 region of .theta.PKC was found to bind to RACK1 in vitro. Therefore, the V3 region could affect not only the targeting of the activated isozymes (in which the V3-region is exposed) but could also regulate the enzyme susceptibilities to proteolysis. Antibodies made against theta V3 fail to precipitate fyn, suggesting competition for the same sites on V3.

V5. The amino acid sequences of the .beta.PKC isoforms .beta.IPKC and .beta.IIPKC are identical except for variability within the V5 region (35 amino acids for .beta.IPKC and 38 amino acids for .beta.IIPKC). Upon activation, .beta.IPKC and .beta.IIPKC translocate to different localization sites in the cell (Disatnik M. -H., Buraggi G., Mochly-Rosen D. Exp. Cell Res. (1994) 210:287-297). This difference in localization of isozymes that are almost identical can be explained by the importance of the V5 region in mediating their targeting. Moreover, .beta.IIPKC was found to selectively translocate to the nucleus upon proliferative stimulation where it selectively phosphorylated the nuclear envelope protein Iamin B.sub.1. (Murray N. R. et al. J. Biol. Chem. (1994) 269:1385-2191).

PKC-related proteins. A distant homolog has been designated PKC-mu. It is unusual in that is has an N-terminal transmembrane domain. Recently the human cDNAs encoding two novel protein kinases have been cloned. These proteins termed PRK1 and PRK2 (protein kinase C related kinase I and 2) show high homology to each other and some homology to the regulatory region of PKC (Palmer R. H., et al. Eur. J. Biochem. (1995) 227:344-351. Since the sequences within the regulatory domain of PKC are responsible for the interaction between a PKC and its anchoring proteins, sequences from PRK1 and PRK2 which show homology to functionally important sequences within the regulatory domain of PKC, are likely to be of biological importance.

Sequences from other isozymes and related proteins that meet the same isozyme selectivity/evolutionary conservation criteria include the following:

.mu.V1-1, .mu.V 1-2 derived from .mu.PKC were picked because they aligned with .epsilon.V1-2 and .theta.V1-2 and part of .theta.V1-1 respectively. .lambda.V1-1 and .lambda.V1-2 from .lambda.PKC were picked based on their alignment with .epsilon.V1-2 and part of .epsilon.V1-3 and .theta.V1-2 respectively. .zeta.V1-1, .zeta.V1-2, .zeta.V1-3 derived from .zeta.PKC were picked according to their homology to: .epsilon.V1-2, .theta.V1-2, and .epsilon.V1-3 respectively. PRK1-1 and PRK2-2 were identified according to their homology to .beta.C2-1. PRK1-2 and PRK2-2 were identified according to their homology to the biologically active .epsilon.PKC-derived peptide .epsilon.V1-3 and part of .epsilon.V1-2. PRK 1-3 and PRK2-3 were picked according to their alignment with the peptide .epsilon.V1-5.

The peptide sequences were generated by aligning the human PKC sequences and the human PRK1 and PRK2 sequences using the MegAlign DNASTAR Inc. program. The sequences were aligned by using the clustal method. The algorithm groups sequences into clusters by examining the distances between all pairs. The clusters are aligned, first individually, then collectively to produce an overall alignment. (Higgins D. G. et al. (1989). CABIOS 5(2):151-153). The matrix for the alignment was PAM250 (percent accepted mutation 250-2.5 mutations per residue). This matrix allows only high stringency alignments.

EXAMPLE 5

fyn as a PKC-theta Cognate

In this example, evidence is provided which identifies a specific protein as a cognate binding partner to PKC-theta in T-cells. This protein is a tyrosine protein kinase called fyn, which was previously known and believed to play an important role in T-cell function. Evidence is further provided to demonstrate that disrupting the localization of PKC-theta, using peptides from the PKC domain which interacts with fyn, depresses T-cell function.

The T-cell receptor (TCR) complex comprises at minimum the CD3 and CD4 complexes of proteins, to which several tyrosine kinases are associated. PLC-gamma, important in generating second messengers such as diacylglycerol and inositol triphosphate, is also a substrate for tyrosine phosphorylation. Among the non-receptor tyrosine kinases are ZAP-70 and the src-related proteins fyn and lck, believed to interact with CD3 and CD4 respectively; another tyrosine kinase, csk, is also associated with the TCR supercomplex of proteins.

In some reports, fyn is only included in 1% of the CD3 complexes. A transient association makes sense, however, in the context of what is known about other localization factors, such as RACK1 for PKC, which are similarly present in particular places in the cell only during particular signal transduction episodes. A role for fyn in T-cell signaling is well documented, including association with at least half a dozen other proteins which are also associated under some experimental conditions with TCR (Penninger el al., Immunol. Rev. 135:183-214 (1993). An association of fyn with PKC has not been previously observed.

The evidence that fin interacts with PKC-theta comes from several independent and mutually supportive lines of experimentation. In most cases, the region of PKC-theta used to define the binding specificity was the V1 region (.about.140 amino acids), which is from the regulatory domain of PKC and is a sequence unique to this isozyme; some experiments also used the V3 domain, FIG. 1. Proteins that interact with these regions fulfill the criteria defined in Examples 1-4. In the data provided in Example 1-4, RACK1 was shown to have some degree of binding to PKC-theta, which could be partially blocked with the T1 or T2 peptides (derived from the V1 region). Both peptides were also able to inhibit normal subcellular translocation of PKC-theta following treatment with activators of the signal transduction network; the latter experiments indicate the importance of the V1 region but do not suffice to identify the physiologically relevant cognate binding partner.

To identify the physiologically relevant cognate binding partner, a Triton (non-ionic detergent) cell extract was prepared from Jurkat T-cells (a human T-cell lymphoma line) using standard procedures. Based on the prior experience that physiologically relevant cognate binding partners for PKC may be associated with the particulate fraction, the "Triton extract" included both soluble and some particulate fraction proteins and is referred to herein as the Triton extract. A V1-his tail construct was also engineered; six histidine residues were attached to the N-terminus of V1. The six histidine residues bind to nickel agarose affinity beads.

Using these beads as an affinity extraction medium, the V1 region was immobilized and incubated with the Triton extract. After washing (by centrifugation in an Eppendorf tube), the bound proteins were eluted with strongly denaturing SDS gel sample buffer. After the eluate was separated by gel electrophoresis and transferred to a membrane, fyn was detectable at the correct molecular weight using a fyn specific antibody (Santa Cruz Biotechnology, Santa Cruz, Calif.). A band at the same molecular weight was also detected by antibodies to phosphotyrosine (Transduction Laboratories, Lexington, Ky.). Antibodies to a related tyrosine kinase, csk, did not indicate any binding to the PKC domain.

Antibodies prepared against the V1/C1 boundary (isozyme-specific) region of PKC-theta were used to immunoprecipitate PKC-theta and any associated proteins from the particulate fraction of an extract from Jurkat T-cells (overnight at 4.degree.). These associated proteins were separated by gel electrophoresis and blotted to a membrane. When the blot was probed with an antibody against phosphotyrosine, several bands were identified including a prominent one at the MW of fyn, 59 kDa, FIG. 7A. By contrast, a band at the MW of lck, another fyn-related tyrosine kinase, was not identified as shown by probing with the appropriate antibody. Direct evidence that the coprecipitated protein is fyn was provided by staining with a commercially available labeled antibody prepared against fyn, FIG. 7B. As a control for specificity, antibodies to PKC-beta were successfully used to coprecipitate RACK1, the known cognate binding protein for this isozyme; fyn was not coprecipitated in this case.

To define the fyn subsequence specificity of interaction, a modified yeast two hybrid system was used (U.S. Pat. No. 5,283,173, Vojtek et al. Cell 74:205-214 (1993). Gene sequences for PKC-theta V1 and fyn were cloned as fusions to complementing halves of a transcription factor. In the particular system used, two reporter genes become activated as a consequence of the association between the two hybrid proteins, which thereby restores the holoenzyme status of the transcriptional activator. The first reporter is a histidine auxotroph repair enzyme, allowing growth selection; the second is beta-galactosidase, whose activity can be visualized with the substrate x-gal which turns blue following enzymatic cleavage. Visual inspection of the colonies allows scoring of interactions as strong (full color development within 2 hours), weak (by 12 hours), null (no signal at 24 hours). Liquid phase assays can also provide quantitative data. Miller, J. H. Experiments in Molecular Genetics (1972) Cold Spring Harbor Laboratory Press.

The catalytic domain and the regulatory domain of fyn were therefore tested separately as binding partners to the V1 region of PKC-theta. The catalytic domain gave a strong signal, and the regulatory domain gave a medium signal. It is not known if each domain folds equally well into a stable structure, so this difference is not conclusive as to what portion of fyn has the best binding to V1. The V3 region of PKC-theta has also been tested against the fyn regulatory, and catalytic domain constructs; both interact as measured by his selection, with beta-galactosidase experiments all showing strong interaction. It thus appears that PKC-theta interacts within over a substantial contact surface, in contrast to coprecipitation experiments using antibodies to the V1 region. In these experiments, antibodies to the V3 region failed to immunoprecipitate fyn, suggesting that the V3 antibodies and fyn are binding in part to the same site.

FIG. 8 thus shows the results of the two-hybrid assay wherein the fyn peptide to be tested is fused to a DNA encoding the polymerase activating domains (of the VP 16 transcription factor) and the relevant portion of the PKC-theta protein is fused to the DNA binding region of the transcription factor (LexA). Beta galactosidase was used as a reporter gene for the results in this Figure.

FIG. 8A shows the constructs and FIG. 8B summarizes the results. In addition to full-length fyn, the regulatory regions represented by fyn positions 1-259 interact strongly with the V1 and V3 regions of PKC-theta as does the kinase region from positions 259-534. fyn.DELTA.SH2, representing positions 1-143 shows diminished interaction with both V1, although interaction is still present. A peptide representing positions 1-88 interacts only weakly with PKC-theta V3 and does not appear to interact with V1. Only very weak interaction between positions 88-143 (SH3) and 143-259 (SH2) with PKC-theta V1 is observed.

To test the specificity of the interaction of PKC-theta V1 with the fyn domains as compared to other proteins, the two hybrid system constructs incorporating the fyn portions were diluted into a large excess of random cDNA clones, prepared from murine T-cells, in the analogous vector. For the kinase domain, 10 of 12 positive clones picked at random were the fyn construct which had been spiked into the library; for the regulatory domain, 3 of 6 were from the spiked fyn construct. The other positive clones are further described in Example 8. The interaction of PKC-theta with fyn appears to be specific since an analogous construct using the V1 region of the most closely related isozyme, PKC-delta, did not appear to bind to fyn.

FIG. 9 represents a diagram and sequence of the fyn splice variant detected in this example, p59 fyn. It is possible that PKC binds at the interface between the two fyn domains, which are both exposed upon activation. A PKC consensus phosphorylation site has been identified within the primary sequence of fyn, centered on threonine-297, in a short stretch of sequence that also scores high as part of an ATP binding consensus site. It is known that fyn has alternative splicing forms; the form found in T-cells includes the consensus PKC site, although this site is not unique to T-cells. Interestingly, addition of ATP to Jurkat T-cell extracts reduces the association of PKC-theta and fyn, measured by immunoprecipitation further suggesting a physiological interaction between the two proteins. By weaker criteria for a PKC consensus site, fyn has 13 additional potential phosphorylation sites. Finally, it is further possible that PKC is a substrate for fyn, since there are 5 tyrosine residues in the V1 region; tyrosine-36 looks particularly reasonable in this regard.

The first PKC cognate binding protein discovered, RACK1, was a clear member of the WD40 family of proteins, characterized by having multiple tandem copies of a sequence of .about.40 amino acids with a conserved WD pair towards the C-terminus. Fyn also has three repeats with weak WD40 homology (aa 51-270), beginning in the middle of the Unique region and ending early in the catalytic domain. At a more detailed level, RACK1 shares several other short sequences of homology with fyn, both in its regulatory and catalytic domains.

Sequence variation between three alternatively spliced human fyn forms is found between amino acids 242-270 (the end of the regulatory domain and the beginning of the catalytic domain, which includes the N-terminal of the third WD-40 repeat). Interestingly, the WD-40 motif landmarks are conserved in all the alternative spliced forms, including S-253, KD-257-258, and WEV 260-262.

Since fyn has an SH3 domain, other homologs of which are known to bind proline rich domains, it is noteworthy that PKC-theta has a moderately proline rich region of 50 residues (includes 10 prolines), accounting for a substantial part of the V3 region. Antibodies to the V3 region failed to immunoprecipitate fyn, suggesting that the antibody and fyn are binding in part to the same site. These results are not conclusive since the V3 region gave positive results in the two-hybrid system. Most other PKC isozymes do not contain proline rich regions, including the most closely related isozyme PKC-delta. The only other isozyme in which a moderately proline-rich domain is found is PKC-mu, a recently described member of the class with less overall PKC homology.

We have also observed that the T2 peptide, which blocks PKC-theta translocation as shown in FIG. 10A and hence blocks association with the cognate localization factor, causes measurable suppression of IL-2 production from activated Jurkat T-cells, FIG. 10B.

In summary, the key independent lines of evidence indicating that fyn is a physiologically relevant binding partner for PKC-theta in T-cells are: (i) a PKC-theta variable domain pulls fyn out of a cell extract in an affinity binding mode; (ii) antibodies to PKC-theta variable domain immunoprecipitate a complex of PKC and fyn from cell extracts; (iii) PKC-theta variable domains and fyn interact in the yeast two hybrid system. In all cases, appropriate controls using closely related proteins show specificity of the interaction.

The existence of cognate binding proteins has been previously described, as have methods for their identification and their utility in drug discovery (for example see Fields, U.S. Pat. No. 5,283,173 and U.S. Pat. No. 5,352,660). The actual identification of fyn as a partner for PKC-theta can now be used in such art-known methods to identify and isolate compounds which block fyn/PKC-theta interactions. Such agents can be used to modulate biological activities which are mediated by fyn/PKC-theta binding; these include activities associated with the immune system.

EXAMPLE 6

Assay for Immune Modulating Compounds Using PKC-theta Translocation as an Index

To assess translocation, suitable cells, preferably nontransformed human T-cells, are cultured to a density of 10.sup.6 /ml and then incubated in cytokine-free medium for overnight. Ten-milliliter aliquots are used for each assay.

The substance to be tested is added to the appropriate samples and incubated for 15 minutes at 37.degree. C. Substances known to stimulate the translocation of PKC-theta are then added: typically PMA at 20-80 nM.+-.PHA at 1 .mu.g/ml are added, and the culture is incubated for 15 minutes at 37.degree. C.

After the incubation period, the samples are spun at 1000 rpm for 10 minutes and the cell pellet is washed with cold PBS. The cells are then resuspended in homogenization buffer and sonicated. They are then centrifuged at 55000 rpm for 30 minutes to obtain a supernatant cytosolic fraction and a pellet which is resuspended in homogenization buffer with a 27-gauge needle to obtain the particulate fraction. After normalization of total protein, the content of PKC in each of the supernatant and particulate fraction is then determined using SDS-PAGE and detection using appropriate antibodies.

FIG. 11 shows the results of SDS-PAGE of the soluble and particulate fraction stained with anti-PKC-theta antibodies (FIG. 12A) or anti-PKC-beta antibodies (FIG. 12B). As shown, PKC-beta is essentially all in the soluble fraction in unstimulated cells, while PKC-theta is more or less evenly distributed between these two fractions. Presumably, this is the case because proliferation of these cells requires the presence of added cytokines, which may effect partial activation of the T-cells. When deprived of cytokines in the medium for, for example, overnight, the levels of PKC-theta in the soluble fraction are increased. In any case, after stimulation with either PMA or a combination of PMA/PHA, which directly cause PKC activation, both PKC-theta and PKC-beta are translocated to the particulate fraction. However, stimulation with OKT-3, an antibody immunoreactive with the CD3 component of T-cell receptors, which provides a more focused physiologically based activation, results in translocation of PKC-theta, but not PKC-beta. The ability of OKT-3 to effect, specifically, translocation of PKC-theta is further evidence of the involvement of PKC-theta in transduction of T-cell receptor mediated stimulation.

The use of the foregoing as an assay for identifying an immune system modulator is shown in FIG. 12. In the assay conducted as described above, the cells were initially incubated with 20 .mu.M of a candidate compound TER14687. As shown in FIG. 12, TER14687 was able to inhibit the translocation of PKC-theta to the particulate fraction when either OKT-3 or PMA/PHA was used as a stimulant. (The assay performed with OKT-3 involved one hour of incubation with OKT-3, rather than the 15-minute period described above.)

TER14687 was also shown to increase the proportion of PKC-theta in the soluble fraction in resting T-cells. T-cells are generally grown in the presence of exogenously added cytokines and are thus in a partially activated state. When cytokine stimulation is withdrawn for a period of time, PKC-theta concentration in the soluble fraction normally increases. TER14687 thus facilitates decay to the resting state.

The effect of TER14687 on the PKC-theta/cognate interaction can also be demonstrated by showing that TER14687 prevents tyrosine phosphorylation of a 21 kD protein after OKT-3 stimulation of T-cells, in which the foregoing protocol was followed except that phosphorylation was measured directly, with an Ab to phosphotyrosine. This phosphorylation is a known downstream event of T-cell activation.

EXAMPLE 7

Assay for Modulation Indexed by Binding

To visualize endogenous PKC cognates T-cell proteins are separated by SDS gel electrophoresis and transferred to nitrocellulose membranes. A large number of proteins are thus distributed on the nitrocellulose. The nitrocellulose is then incubated with PKC-theta and washed, and the bound PKC-theta is visualized using labeled anti-PKC-theta. PKC-theta protein was produced Sf9 insect cells using a baculovirus expression system, Invitrogen, San Diego, Calif. In the absence of substances known to activate PKC, essentially no bands appear. In the presence of PKC activators, a large number of bands is seen. Some of these bands may be true endogenous cognates; others may mimic cognate binding abilities in the denatured state characteristic of proteins transferred to nitrocellulose membranes. As such, these proteins are substitutes for the endogenous cognates.

In the presence of TERI4687 during PKC-theta incubation on the blot, binding is diminished, thus showing the modulating effect of TER14687 on cognate binding.

In addition to TER14687, the following peptides derived from the regulatory region of fyn were tested: fyn2, representing residues 111-118 underlined in FIG. 9, showed partial inhibition of PKC-theta binding to fyn; and fyn3, residues 188-195 underlined in FIG. 9, showed complete inhibition of PKC binding. fyn2 and fyn3 are derived from the portion of fyn which shows homology to WD40 repeats. An additional peptide, fyn1 was not capable of inhibiting binding; this peptide was derived from a fyn splice variant different from that shown in FIG. 9. Another inactive peptide was fyn4, representing residues 441-449 from the catalytic domain.

EXAMPLE 8

Additional Cognates Identified by the Yeast Two-Hybrid System

The intracellular method to detect peptide/peptide binding, described in Example 5 hereinabove was used a) to find additional peptides which bind to PKC-theta and b) to provide an assay system for the effect of candidate substances on interaction with PKC-theta and its cognate.

First, the yeast two-hybrid system was used to retrieve cDNAs encoding peptides that interact with PKC-theta.

In this approach, plasmids containing the polymerase activating domain fused to the kinase or regulatory domains of fyn were mixed with similar vectors containing a cDNA library from the murine T-cell line HT2 in a ratio of 1:500. Yeast harboring plasmids containing PKC-theta V1 fused to the DNA binding domain were transformed with this mixture of plasmids at one-tenth of the regular protocol. The yeast were plated onto media selecting for presence of the plasmids (THULL) plates and assayed for .beta.-galactosidase on filter lifts. Positive colonies were picked on THULL grids and retested. DNA was extracted from colonies which remained positive, and then amplified using 5' and 3' plasmid-based polylinker flanking primers and the products were analyzed by Southern blot using fyn as a probe.

Of 12 positives found when fyn kinase domain was mixed into the cDNA library, 10 were the fyn kinase itself However, two contained sequences other than those derived from fyn.

Of six positives obtained from the mixture when the cDNA library had been mixed with the fyn regulatory domain, three contained DNA sequences other than fyn. Two of these, designated 2-10 and 2-40 have identical nucleotide sequences not yet identified with any known sequence and the other, 2-32 appears to be a mouse homolog of a recently isolated human gene, ELL. The cDNA insert in 2-10 and 2-40 contained 1083 base pairs with an open reading frame encoding a previously unknown 335 amino acid peptide with several potential zinc-finger motifs. The cDNA hybridizes with a 2.4 kD transcript present in lymphoid tissue and a 6 kD transcript present in heart and skeletal muscle as shown by Northern blot.

Both clones 2-10 and 2-32, when tested in a liquid .beta.-gal assay, were able to interact with PKC-theta V1 in the two-hybrid system described above. When transformed into the parental yeast strain L40, lacking PKC-theta V1, no activation was observed by a reporter assay, nor was activation observed when theta V1 was replaced by delta V1 or an unrelated protein, lamin.

Using the same assay, the effect of TER14687 on the interaction of PKC-theta V1 with either fyn or with clones 2-10 and 2-32 was tested. In this assay, yeast containing the above-described two-hybrid system were inoculated 1: 10 dilutions into 1 ml THULL containing 40 .mu.M TER14687 or a control compound in DMSO. Various other controls were included; for example, TER14687 did not inhibit expression of the reporter gene when the transcription factor (LexA/E2A) was supplied in covalently bound form. The activity of .beta.-galactosidase produced was measured as OD at 420 nm. TER14687 markedly diminished binding theta V1 to fyn or the substitute cognates generated from 2-10 and 2-32, as shown in FIG. 13.

In addition, several clones encoding apparent cognate proteins were obtained from a human's CD4.sup.+ T-cell cDNA library using the same yeast two-hybrid protocol. The cDNA library was constructed using T-cells isolated from donor blood and had a complexity estimated between 10.sup.5 to 5.times.10.sup.6 independent clones. Out of 1.4.times.10.sup.7 clones screened, 63 remained positive after a secondary .beta.GAL filterlift assay.

Thirty-nine of the clones appear to have identical sequence and the same approximate cDNA insert size. The sequence contains an open reading frame and is clearly a partial sequence which does not match any data base sequence. One of these clones, No. 10, was retested in the yeast two-hybrid system and found to bind .theta.V1, but not .delta.V1 or lamin.

In addition, two clones from this library, 1-22 and 1-23 share sequence with human elongation factor 1-.gamma.. 1-23 overlaps with 1-22, but both are partial sequences. Both bind specifically to .theta.V1 in the yeast two-hybrid assay.

One additional clone, 2-18 also shows specific binding to .theta.V1 in the yeast two-hybrid assay and shows high homology with SH3-containing c-abl binding proteins which inhibit c-abl transformation. U17698 murine abl philin-1 is 89% homologous; U31089 human Abl binding protein 3 is 94% homologous; and U23435 human Abl interactor 2 is 88% homologous.

Two additional clones were obtained in a similar manner. Clone 2-20 contains 729 bp as a partial cDNA matched to human elongation factor 2. Clone 3-1 contains 839 bp with high homology to HSV1 transducing factor .alpha..

The nucleotide sequences and the amino acid sequences encoded for clones 2-10, 2-32, #10, 1-22, 2-18, 2-20 and 3-1 are shown in FIG. 15.

EXAMPLE 9

Association of PKC-theta Activation with Allergic Reactions

A human T-cell line, TT7.5, is physiologically activated with OKT3 coated on tissue culture plates at 10 .mu.g/ml as shown by enhanced proliferation using a tritiated thymidine assay. Activation is also characterized by enhanced secretion of interferon-.gamma., and of interleukins 4 and 5, but not of IL-2, as assayed in appropriate ELISAs. This pattern characterizes a Th2-like cell. Th2 cells have been shown to mediate allergy via the immunoregulatory effects of IL-4 and IL-5 and can thus be used as an in vitro model of T-cell function in mediating allergy.

As was shown above, OKT3 stimulation of T-cell lines, including TT7.5 cells, results in translocation of PKC-theta. In a similar assay that uses cytokine production as a measure of PKC-theta interaction with cognate, TER14687, administered during OKT3 activation, inhibits the production of IL-4 and IL-5, but not the induced production of interferon-.gamma., as shown in FIG. 14. This pattern of cytokine production is associated with reduced stimulation of IgE and eosinophil production. Similar results are obtained with the Th2-like T-cell line TT3.6. In a Th1-type cell line (TT6.4) TER14687 inhibits OKT3-induced IL-2 production, but not interferon-.gamma. production.

TER14687 also strongly inhibits expression, in TT7.5 cells, of the surface markers of T-cell activation, such as CD69, CD25 and CD40L following overnight OKT3 stimulation. CD69 is expressed exclusively on activated T-cells, CD25 is a low-affinity IL-2 receptor expressed on activated T-cells and CD40L is a ligand for CD40 which is also expressed exclusively on activated T-cells. These markers were assayed using appropriately labeled fluorescent antibodies and a flow cytometer.

Similar results were obtained in Jurkat cells following activation by PMA rather than OKT3. In the presence of 50 .mu.M fyn2 peptide, enhanced expression of CD69 is reduced.

Claims

We claim:

1. A method to identify a substance that has immunomodulating activity which method comprises providing an environment containing PKC-theta or a fragment thereof and a cognate of said PKC-theta under conditions wherein said PKC-theta or fragment interacts with said cognate; adding a candidate substance to said environment; determining the interaction of PKC-theta or fragment with cognate in the presence and absence of said candidate; comparing said interaction in presence and absence of candidate; wherein a candidate which modulates said interaction is identified as having immunomodulating activity, and wherein said cognate is selected from the group consisting of fyn; the fyn fragment fyn-3; the fyn fragment fyn-2; the protein encoded by clone 2-10 (SEQ ID NO:3); the protein encoded by clone 2-32 (SEQ ID NO:5); and a protein which inhibits the binding of fyn to PKC-theta.

2. The method of claim 1 wherein said environment is an intracellular environment.

3. The method of claim 2 wherein determining said interaction is by measuring translocation of said PKC-theta or fragment.

4. The method of claim 2 wherein determining said interaction is by measuring tyrosine phosphorylation of a 21 kD protein after OKT-3 stimulation of T-cells.

5. The method of claim 2 wherein determining said interaction is by measuring diminution of IL4 and/or IL5 production without affecting IFN.gamma. production.

6. The method of claim 1 wherein determining said interaction comprises determining binding of said PKC-theta or fragment to its cognate.

7. The method of claim 6 wherein said environment is an intracellular environment.

8. The method of claim 7 wherein said binding is determined by means of a "two-hybrid" system as defined herein.

Patent Citations (3)

Patent	Date	Inventor
US5283173	1994-02-01	Fields et al.
US5352660	1994-10-01	Pawson
WOXWO95/21252	1995-01-01

Non-Patent Literature (22)

Derwent Publications Ltd., AN 94-026226 (1994).
Meller et al., Mol. Cell. Biol. 16:5782-5791, 1996.
Dalrymple, M.A. et al., "The Product of the PRP4 Gene ofS. cerevisiae Shows Homology to .beta. Subunits of G Proteins," Cell 58:811-812 (1989).
Dynlacht, B.D. et al., "The dTAF . . . 80 subunit of Drosphila TFIID contains .beta.-transducin repeats," Nature 363: 176-179 (1993).
Fong, H.K.W. et al., "Repetitive segmental structure of the transducin .beta. subunit: Homology with the CDC4 gene and identificationof related mRNAs," Proc. Natl. Acad. Sci. USA 83: 2162-2166 (1986).
Guillemot, F. et al., "Physical linkage of a guanine nucleotide-binding protein-related gene to the chicken major histocompatibility complex," Proc. Natl. Acad. Sci. USA 86: 4594-459 (1989).
Keleher, C.A. et al., "Ssn6-Tup 1 Is a General Repressor of Transcription in Yeast," Cell 68: 709-719 (1992).
Mochly-Rosen, D. et al., "Identification of intracellular receptor proteins for activated protein kinase C," Proc. Natl Acad. Sci USA 88: 3997-4000 (1991).
Mochly-Rosen, D. et al., "Intracellular Receptors for Activated Protein Kinase C," J. Biol. Chem 266(23): 14866-14868 (1991).
Peitsch, M.C. et al., "Sequence similarity of phospholipase A2 activating protein and the G protein .beta.-subunits: a new concept of effector protein activation in signal transduction?," TIBS 18(8): 292-293 (1993).
Ron, D. et al., "An Autoregulatory Region in Protein Kinase C: The Pseudoanchoring Site" Proc. Natl. Acad. Sci. USA 92:492-496 (1995).
Ron, D. et al. "Agonists and Antagonists of Protein Kinase C Function, Derived from its Binding Proteins" J. Biol. Chem. 269:21395-21398 (1994).
Ron, D. et al. "Cloning of an Intracellular Receptor for Protein Kinase C: A Homolog of the Beta Subunit of G Proteins" Proc. Natl. Acad. Sci. USA 91:839-843 (1994).
Ruggieri, R., et al. "MSI1, a Negative Regulator of the RAS-cAMP Pathway in Saccharomyces Cerevisiae" Proc. Natl. Acad. Sci. USA 86:8778-8782 (1989).
Smith, B.L. and Mochly-Rosen, D., "Inhibition of Protein Kinase C Function by Injection of Intracellular Receptors for the Enzyme," Biochem. Biophys. Res. Comm. 188(3): 1235-1240 (1992).
Takagaki, Y. and Manley, J.L., "A Human Polyadenylation Factor Is a G Protein .beta.-subunit Homologue," J. Biol. Chem 267(33):23471-23474.
Tamaki, M. et al. "Rat Lipocortin I cDNA" Nucleic Acids Res. 15:7637 (1987).
van der Voorn, L. and Ploegh, H.L., "The WD-40 repeat," FEBS Lett. 307(2): 131-134 (1992).
Wallner, B., et al. "Cloning and Expression of Human Lipocortin, a Phospholipase A2 Inhibitor with Potential Anti-Inflammatory Activity" Nature 320:77-81 (1986).
Weinstat-Saslow et al., "A Transducin-like Gene Maps to the Autosomal Dominant Polycystic Kidney Disease Gene Region" Genomics 18:709-711 (1993).
Williams, F.E. and Trumbly, R.J., "Characterization of TUP1, a Mediator of Glucose Repression in Saccharomyces cerevisiae," Mol. Cell. Biol. 10(12): 6500-6511 (1990).
Williams, F.E. et al., "The CYC8 and TUP1 Proteins Involved in Glucose Repression in Saccharomyces cerevisiae Are Associated in a Protein Complex," Mol. Cell. Biol. 11(6): 3307-3316 (1991).

Abstract

Description