CA2081807A1 - Protein partner screening assays and uses thereof - Google Patents

Abstract

Abstract A rapid, simple and inexpensive method to screen and classify proteins as partners of dimeric proteins is described. The method utilizes fusion protein constructs containing a DNA binding domain and complementary dimerization domains from a different protein. According to the method of the invention, protein partner heterodimer formation is detected by the ability of the protein partner to displace formation of DNA binding domain homodimers, and thus reveal a phenotypic change in a bacterial host which was dependent upon maintenance of the DNA binding domain homodimer configuration. The method of the invention may further be used to identify compounds of interest which inhibit such heterodimer formation, and especially to identify compounds which prevent heterodimer formation and activation of oncogenic transcriptional regulatory proteins.

A190-03.WP5 102892

Description

2~8~
I'itle of the lnvention Protein Partner Screening Assays and Uses Thereof Cross-References to Related Applications This application is a continuation-in7art of U.S. Application No.
07/915,745 filed July 21, 1992, which is a continuation of U.S. Application No. 07/815,880, filed January 7, 1992, which is a continuation of U.S.
Application No. 07/510,254 filed April 19, 1990.

F'ield of the Invenfion This invention is in the field of molecular biology and is directed to a method of identifying a peptide capable of associating with another peptide in a heterodimeric complex. The invention is also directed to a method of identifying inhibitors of such heterodimeric complex formation.

BackgrQund of the Invention Many regulatory proteins are heterodimers, that is, they are composed of two different peptide chains which interact to generate the native protein.
Among such regulatory proteins are DNA binding proteins which are capable of binding to specific DNA sequences and thereby regulating transcription of DNA into RNA. The dimerization of such proteins is necessary in order for these proteins to exhibit such binding specificity. A
large number of transcriptional regulatory proteins have been identified: Myc, Fos, Jun, Ebp, Fra-1, Jun-B, Spl, H2TF-1/NF-KB-like protein, PRDI, TDF, GLI, Evi-1, the glucocorticoid receptor, the estrogen receptor, the progesterone receptor, the thyroid hormone receptor (c-erbA) and ZIF/268, OTF-1(OCT1), OTF-2(0CT2) and PIT-1; the yeast proteins GCN4, GAL4, HAPl, ADR1, SWI5, ARGRII and LAC9, mating type factors MATa1, .. .

- .
~- ' - 2~8~80~

MATa2 and MATal; the Neurospora proteins cys-3 and possibly cpc-l; and the Drosophila protein bsg 25D, kruppel, snail, hunchback, serendipity, and suppressor of hairy wing, antennapedia, ultrabithorax, paired, fushi tarazu, cut, and engMiled. Eukaryotic transcriptional regulatory proteins, and the methods used to characterize such proteins, have been recently reviewed (Pabo, C.O. et al., Ann. Rev. Biochem. 61:1053-1095 (1992); Johnson, P. F.
et al., Ann. Rev. Biochem. 58:799-839 (1989)).
Members of the mammalian transcriptional regulatory protein families Jun/Fos and ATF/CREB only bind to DNA as dimers. The proteins in these families are "leucine zipper" proteins which contain a region rich in basic amino acids followed by a stretch of about 35 amino acids which contains 4-S
leucine residues separated from each other by 6 amino acids (the "leucine zipper" region). Collectively, the combination of a basic region and the leucine zipper region is termed the bZlP domain.
Generally, it is the basic region which has been found to be predominantly involved in contacting DNA whereas the zipper region mediates the dimerization. Many dimeric combinations are possible, however, the particular nature of the zipper specifies which partnerships are permissible (Abel, T. et al., Nature 341:24-25 (1989)).
Another large family of proteins contains the DNA
binding/dimerization motif known as the basic helix-loop-helix motif (bHLH) (Jones, N., Cell 61:9-11 (1990)). A bHLH protein generally contains a basic N-terminus followed by a helix-loop-helix structure; two short amphipathic helices containing hydrophobic residues at every third or fourth position. The sequence of the basic region characteristically reveals no indication of an amphipathic helix. The intervening loop region usually contains one or more helix-breaking residues.
The bHLH motif was first detected in two proteins, E12 and E47, that bind to a specific "E box" DNA enhancer sequence found in immunoglobulin enhancers (Murre C. et al.~ Cell 56:777-783 (1989)). E motifs generally are double stranded variants of the 5'-CAGGTGGC-3' consensus sequence. For . .
.

-3- 2 0 ~

example, the ~E1 motif is GTCAAGATGGC [Seq. ID NO. 1], ~4E2 motif is AC~CAGCTGGC[SEQID NO. 2], ~E3 is GTCATGTGGC [Seq. ID NO. 3~, ,uEisTGCAGGTGT (Murre, C. et al., Cell 56:777-783 (1989)). Like many transcriptional factors, peptides containing the bHLH motif often dimerize with each other, either as a homodimer which contains two identical peptides or as a heterodimer which contains two different peptides. Examples of heterodimeric complex of two bHLH proteins binding DNA with a greater efficiency than homodimeric complexes of either peptide in the heterodimer are known (Murre C. et al., Cell 56:777-783 (1989); Murre, C e~ al., Cell 58:537-544 (1989)).
Identification of partners which direct protein-DNA binding and compounds which inhibit such activity by inhibiting such protein partner interaction could be very useful. For example, identification of partners of themyc protein and inhibitors of myc-partner interactions could provide a means for treating diseases in which expression and activity of myc is a factor in promoting cell growth or in maintaining the cell in a transformed state.
Myc is a bHLH protein and the bHLH domain of c-myc is encoded in c-myc amino acids 255-410. The sequence homology between the proteins expressed by the three n~yc genes (human N-myc 393437, human c-myc 34~
401, and human ~myc 289-338) and other genes which contain a bHLH
domain have been compared (Murre C. et al., Cell 56;777-783 (1989)).
Proteins such as myc which contain the bHl,H motif also possess the ability to dimerize with other bHLH motif proteins. Such interactions among bHLH proteins may play a critical role in their function andtor regulation.
ldentification of these protein partners would be useful not only in understanding how these proteins function, but also in developing or identifying inhibitors of these proteins. For example, identification of myc-partners would make it possible to identify inhibitors of myc-partner interactions. By inhibiting such interactions, inhibition and/or control of myc-induced cell growth may be achieved.

To date, no myc inhibitors have been identified. The identification of such inhibitors has suffered for lack of a simple, inexpensive and reliable screening assay which could rapidly identify potential inhibitors and active derivatives thereof. Thus a need still exists for rapid, economical screening S assays which identify specific inhibitors of oncogene activity.

Sl~mnza~y of the Invention ~ecognizing the potential importance of inhibitors of oncoproteins in the therapeutic treatment of many forms of cancer, and cognizant of the lack of a simple assay system in which such inhibitors might be identified, the inventors have investigated the use of chimeric oncogene constructs in in vitro assays in prokaryotic hosts as a model system for identifying agents which alter oncogene expression.
These efforts have culminated in the development of a simple, inexpensive assay which can be used to identify protein partners in general, and partners of transcriptional regulatory proteins in particular.
The methods of the invention are especially useful for the identification of partners which influence transcriptional regulatory proteins, and especially oncoprotein activity.
The method of the invention further provides a method of identifying, isolating and characteri~ing inhibitors of such partner formation and especiallyinhibitors of oncoprotein activity.
The invention further provides a quick, reliable and accurate method for objectively classifying compounds, including human pharmaceuticals, as inhibitors of oncogene activity.
The invention further provides a method of identifying protein partners by their ability to disrupt ~cl induced repression of phage promoters in bacterial hosts which express fusion proteins containing the cl DNA binding domain and a dimerization domain from a protein of interest. Proteins identified by this method are partners of the protein from which the . -- . .

dimerization domain was obtained. Protein partners thus identified are already in a cloned form, amenable to further characterization.

Brief Description of the Drawings Figure 1 shows the DNA sequence (Seq. ID No. 4) and protein sequence (Seq. ID No. S) of human c-myc exon 3 and the sites used to synthesize the HLH/LZ and HLH fragments of c-myc.

Description of the Preferred Embodiments In the description that follows, a number of terms used in recombinant DNA technology are extensively utiliæd. In order to provide a clearer and more consistent understanding of the specification and claims, including the scope to be given such terms, the following definitions are provided in alphabetical order.
Bioactive Compound. The term "bioactive compound" is intended to refer to any compound which induces a measurable response in the assays of the invention.
Clonin~ vehicle. A "cloning vehicle" is any molecular entity which is capable of providing a nucleic acid sequence to a host cell for cloning purposes. Examples of cloning vehicles include plasmids or phage genomes.
A plasmid which can replicate autonomously in the host cell is especially desired. Alternatively, a nucleic acid molecule which can insert into the host cell's chromosomal DNA is especially useful.
Cloning vehicles are often characterized by one or a small number of endonuclease recognition sites at which such DNA sequences may be cut in a determinable fashion without loss of an essential biological function of the vehicle, and into which DNA may be spliced in order to bring about its replication and cloning.

.

, The cloning vehicle may further contain a marker suitable for use in the identification of cells transforrned with the cloning vehicle. Markers, for example, are tetracycline resistance or ampicillin resistance. The word "vector" is sometimes used for "cloning vehicle. "
Compound. The term "compound" is intended to refer to a chemical entity, whether in the solid, liquid, or gaseous phase. The term should be read to include synthetic compounds, natural products and macromolecular entities such as polypeptides, polynucleotides, or lipids, and also small entities such as neurotransmitters, ligands, horrnones or elemental compounds.
Dimeric Protein. The term "dimeric protein" is intended to refer to a protein which contains two polypeptide chains that associate with one another, but which are not bound to one another by an amino acid linkage. Association of the polypeptide chains may be due to, for example, hydrogen bonding, ionic interactions, hydrophobic interactions, disulfide bonds, and the like.
Dimerization Domain. The term "dimerization domain" is intended to refer to that portion of each polypeptide chain of a dimeric protein which is necessary for the polypeptide chains to associate with one another. The dimerization domains of a dimeric protein, which may be identical or different, are referred to herein as complimentary to each other.
Expression. Expression is the process by which the information encoded within a gene is transcribed and translated into protein.
A nucleic acid molecu1e, such as a DNA or gene is said to be "capable of expressing" a polypeptide if the molecule contains the sequences which code for the polypeptide and the expression control sequences which, in the appropriate host environment, provide the ability to transcribe, process and translate the genetic information contained in the DNA into a protein product, and if such expression control sequences are operably-linked to the nucleotide sequence which encodes the polypeptide.
Expression vehicle. An "expression vehicle" is a vehicle or vector similar to a cloning vehicle but is especially designed to provide sequences capable of expressing the cloned gene after transformation into a host.

2~8~8~7 In an expression vehicle, the gene to be cloned is operably-linked to certain control sequences such as promoter sequences.
Expression control sequences will vary depending on whether the vector is designed to express the operably-linked gene in a prokaryotic or eukaryotic host and may additionally contain transcriptional host specific elements such as operator elements, upstream activator regions, enhancer elements, termination sequences, tissue-specificity elements, andlor translational initiation and termination sites.
Functional Derivative. A "functional derivative" of a fusion protein is a protein which possesses an ability to dimerize with a partner protein, and/or an ability to bind to a desired DNA target, that is substantially similar to theability of the fusion protein constructs of the invention to dimerize. By "substantially similar" is meant that the above-described biological activities are qualitatively similar to the fusion proteins of the invention but quantitatively different. For example, a functional derivative of a fusion protein might recognize the same target as the fusion protein, or form heterodimers with the same partner protein, but not with the same affinity.
As used herein, for example, a peptide is said to be a "functional derivative" when it contains the amino acid sequence of the fusion protein plus additional chemical moieties not usually a part of a fusion protein. Such moieties may improve the derivative's solubility, absorption, biological half-life, etc. The moieties may alternatively decrease the toxicity of the derivative, or eliminate or attenuate any undesirable side effect of the derivative, etc. Moieties capable of mediating such effects are disclosed in Remington's Pha~7naceuhcal Sciences (1980). Procedures for coupling such moieties to a molecule are well known in the art.
A functional derivative of a fusion protein may or may not contain post-translational modifications such as covalently linked carbohydrate, depending on the necessity of such modifications for the performance of the methods of the invention.

~ .

: ~ .

-8- 2 ~ J

The term "functional derivative" is intended to encompass functional "fragments, " "variants, " "analogues, u or "chemical derivatives" of a molecule.
Fusion protein. As used herein, "fusion protein" is a hybrid protein which has been constructed to contain domains from two different proteins.
The term "fusion protein gene" is meant to refer to a DNA sequence which codes for a fusion protein, including, where appropriate, the transcrip-tional and translational regulatory elements thereof.
Heterodimer. The term "heterodimer" or "heterodimeric protein" is intended to refer to a protein which contains two different polypeptide chains that associate with one another, but which are not bound to one another by an amino acid linkage.
Homodimer. The term "homodimer" or "homodimeric protein" is intended to refer to a protein which contains two identical polypeptide chains that associate with one another, but which are not bound to one another by an amino acid linkage. This term may be modified to refer only to a particular portion of a dimeric protein. For instance, a DNA binding domain homodimer is intended to refer to any dimeric protein containing identical DNA binding domains on its separate polypeptide chains.
Host. By "host" is meant any organism that is the recipient of a cloning or expression vehicle as deflned herein. Appropriate hosts for use in the method of the invention include, but are not limited to, bacteria, yeast, and mammalian cells.
Marker Gene. The term "marker gene" is intended to refer to a gene whose expression in a host cell produces a readily observable, assayable, or selectable phenotype. Examples of marker genes which may be useful in the method of the invention include, but are not limited to, lacZ, aada (which confers spectinomycin and streptomycin resistance), and ble-1 (which confers bleomycin and phleomycin resistance).
Operably-linked. As used herein, two macromolecular elements are operably-linked when the two macromolecular elements are physically 9 2~

arranged such that factors which influence the activity of the first element cause the first element to induce an effect on the second element.
Promoter. A "promoter" is a DNA sequence located proximal to the start of transcription at the 5' end of the transcribed sequence, at which RNA polymerase binds or initiates transcription. The promoter may contain multiple regulatory elements that interact in modulating transcription of the operably-linked gene.
Protein Partner. The term "protein partner" is intended to refer to a polypeptide chain capable of associating with a heterologous polypeptide chain to form a heterodimeric protein. The two polypeptide chains of a heterodimeric protein are herein referred to as "partners" of one another. A
polypeptide chain of a homodimeric protein may act as a partner in a heterodimeric protein.
Response. The term "response" is intended to refer to a change in any parameter which can be used to measure and describe the effect of a compound on the activity of a protein. The response may be revealed as a physical change (such as a change in phenotype) or a molecular change (such as a change in a reaction rate or affinity constant). Detection of the response may be performed by any means appropriate.
Variant. A "variant" of a fusion protein is a protein which contains an amino acid sequence that is substantially similar to, but not identical to, the amino acid sequence of a fusion protein constructed from naturally-occurring domains, that is, domains containing the native with the amino acid sequence.
By a "substantially similar" amino acid sequence is meant an amino acid sequence that is highly homologous to, but not identical to, the amino acid sequence found in a fusion protein. Highly homologous amino acid sequences include sequences of 80% or more homology, and possibly lower homology, especially if the homology is concentrated in domains of interest.
Transcription regulatory proteins, which normally function as dimeric proteins, have been found to possess discrete dimerization domains and DNA

~' ~ .

-lo- 2~

bincling domains. The inventors have used these findings to develop the method of the invention for identifying a partner of a dimeric protein. This method involves construction of chimeric peptides with (1) known complementary dimerization domains and (2) DNA binding domains which, when present in homodimer form, are capable of conferring a detectable phenotype upon a host cell (preferably a bacterial host cell, such as E. coli).
This detectable phenotype is a marker other than resistance to phage infection, such as infection by lambda phage. It has been discovered that this phenotype may be detected by methods which do not depend upon phage resistance, or phage-induced cell Iysis.
In the host cell, the chimeric peptides form DNA binding domain homodimers by association of the known complementary dimerization domains. Protein partners capable of associating with the chimeric peptides to form heterodimeric proteins will interfere with formation of the chimeric peptides into DNA binding domain homodimers. By monitoring the homodimer-conferred phenotype in the host cell, formation of interfering heterodimers may be detected and protein partners thus identified.
This method of the invention is generally useful to identify partners for any homodimer or heterodimer. For a homodimer, a single chimeric peptide containing the dimerization domain of the homodimer is used. For a heterodimer, two separate chimeric peptides are used; each containing one of the complementary dimerization domains of the heterodimer. The chimeric peptides also contain a DNA binding domain that confers a detectable phenotype in homodimer form.
DNA binding domains useful in construction of chimeric peptides of the invention may be obtained from proteins where they have been identified.
For example, DNA binding domains may be obtained from bacteriophage repressors, such as bacteriophage lambda (~) repressor. In particular, the lambda repressor protein cI is useful as a source of a DNA binding domain.
cl represses lambda gene expression in its homodimeric form (Lambda 11, :: :

2~8~Q~

Hendrix, R.W. et al., eds., Cold Spring Harbor Laboratory, New York, (1983).
Other DNA binding domains may be identified by a variety of techniques known in the art and previously used to identify such domains (see Pabo, C.O. etal., Ann. Rev. Biochem. 61:1053-1095 (1992); Johnson, P. E.
et al., Annu. Rev. Biochem. ~8:799-839 (1989) for a review of such domains).
DNA binding proteins, and DNA binding domains in such proteins, are identified and purified by their affinity for DNA. For example, DNA binding may be revealed in filter hybridization experiments in which the protein 10 (usually labelled to facilitate detection) is allowed to bind to DNA immobilized on a filter or, vice versa, in which the DNA binding site (usually labelled) is bound to a filter upon which the protein has been immobilized. ~he sequence specificity and affinity of such binding is revealed with DNA protection assays and gel retardation assays. Purification of such proteins may be performed 15 utilizing sequence-specific DNA affinity chromatography techniques, that is, column chromatography with a resin derivatized with the DNA to which the domain binds. Proteolytic degradation of DNA binding proteins may be used to reveal the domain which retains the DNA binding ability.
Dimeric proteins for which protein partners are desired to be identified 20 sene as the source of dimerization domains useful in the construction of chimeric peptides of the invention. Dimerization domains may be currently known dimerization domains or those recognized by their homology to known dimerization domains. Other dimerization domains may be predicted by analysis of the three-dimensional structure of a protein using the amino acid 25 sequence and computer analysis techniques commonly known in the art, for example, the Chou-Fasman algorithm. Such techniques allow for the identification of helical domains and other areas of interest, for example, hydrophobic or hydrophilic domains, in the peptide structure.
One class of known dimerization domains are the HLH domains, which 30 share a common helix-loop-helix amino acid structure. The bHLH region of the c-myc protein is one such dimerization domain. This domain is :. :

cornplementary to itself and is therefore useful in the construction of chimericpeptides that form homodimers.
An HLH dimerization domain in a protein can be identified by comparison of an amino acid sequence with that of ten known HLH
S dimerization domains (amino acids 336-393 in E12, 33~393 in E47, 55~613 in daughterless, 357-407 in ~st, 393-437 in human N-myc, 289-338 in human ~myc, 346~01 in human c-myc, 108-164 in MyoD, and genes of the achaete-scute locus: 101-167 of T4, 26-9S of T5 (Murre, ~. et al., Goll 56:777-783 (1989)). The HLH dimerization domain contains two amphipathic helices separated by an intervening loop. The first helix contains 12 amino acids and the second helix contains 13 amino acids. Certain amino acids appear to be conserved in the HLH format, especially the hydrophobic residues which are present in the helices. Comparisons of the two sequences named above shows that there are five virtually identical hydrophilic residues within the 5' end of the homologous region and a set of mainly hydrophobic residues located in two short segments that are separated form one another by a sequence that generally contains prolines or clustered glycines.
Another class of known dimerization domains are the leucine zipper domains. This domain is typically about 35 amino acids long and contains a repeating heptad array of leucine residues and an exceedingly high density of oppositely charged amino acids (acidics and basics) juxtaposed in a manner suitable for intrahelical ion pairing. It is thought that the leucines extendingfrom the helix of one polypeptide interdigitate with those of the analogous helix of a second peptide (the partner) and form the interlock termed the leucine zipper.
The DNA binding domain and the dimerization domain are engineered into the fusion protein in a manner which does not destroy the function of either domain; that is, the DNA binding domain, when properly dimerized, can recognize the DNA element to which it naturally binds and the dimerization domain retains the ability to dimerize with its partners. One of skill in the art, by running control assays, will be able to establish that the fusion protein functions in the proper manner.
The DNA sequence encoding the fusion protein may be chemically constructed or constructed by recombinant means known in the art. Methods of chemically synthesizing DNA are well known in the art (Oligonucleot~de Synthesis, A Practical Approach, M.J. Gail, ed., IRL Press, Washington, D.C., 1094; Synthesis and Applicahons of DNA and RNA, S.A. Narang, ed., Academic Press, San Diego, CA, 1987). Because the genetic code is degenerate, more than one codon may be used to construct the DN~ sequence encoding a particular amino acid (Watson, J.D., In: Molecular Biology of the Gene, 3rd edition, W.A. Benjamin, Inc., Menlo Park, CA, 1977, pp. 35~-357).
To express the recombinant fusion constructs of the invention, transcriptional and translational signals recognizable by the host are necessary.
A cloned fusion protein gene, obtained through the methods described above, and preferably in a double-stranded form, may be operably-linked to sequences controlling transcriptional expression in an expression vector, and introduced, for example by transformation, into a host cell to produce the recombinant fusion proteins, or functional derivatives thereof, for use in the methods of the invention.
Transcriptional initiation regulatory signals can be selected which allow for repression or activation of the expression of the gene encoding the fusion protein, so that expression of the fusion construct can be modulated, if desired. Of interest are regulatory signals which are temperature-sensitive so that by varying the temperature, expression can be repressed or initiated, or are subject to chemical regulation, for example, by a metabolite or a substrate added to the growth medium. Alternatively, the fusion construct may be constitutively expressed in the host cell.
It is necessary to express the proteins in a host wherein the ability of the protein to retain its biological function is not hindered. Expression of ' . ''': ' .

~08~7 proteins in bacterial hosts is preferably achieved using prokaryotic regulatory signals.
Expression vectors typically contain discrete DNA elements such as, for example, (a) an origin of replication which allows for autonomous replication of the vector, or elements which promote insertion of the vector into the host's chromosome in a stable manner, and (b) specific genes which are capable of providing phenotypic selection in transformed cells. Many appropriate expression vector systems are commercially available which are useful in the methods of the invention.
Once the vector or DNA sequence containing the construct(s) is prepared for expression, the DNA construct(s) is introduced into an appropriate host cell by any of a variety of suitable means, for example by transformation. After the introduction of the vector, recipient cells are grown in a selective medium, which selects for the growth of vector-containing cells.
Expression of the cloned gene sequence(s) results in the production of the fusion protein.
If the fusion protein DNA encoding sequence and an operably-linked promoter is introduced into a recipient host cell as a non-replicating DNA (or RNA) molecule, which may either be a linear molecule or, more preferably, a closed covalent circular molecule which is incapable of autonomous replica-tion, the expression of the fusion protein may occur through the transient expression of the introduced sequence.
Genetically stable transformants may be constructed with vector systems, or transformation systems, whereby the fusion protein DNAis integrated into the host chromosome. Such integration may occur de noYo within the cell or be assisted by tMnsformation with a vector which functionally inserts itself into the host chromosome, for example, with bacteriophage, transposons or other DNA elements which promote integration of DNA sequences in chromosomes.
Cells which have been transformed with the fusion protein DNA
vectors of the invention are selected by also introducing one or more markers ' :

2 ~

which allow for selection of host cells which contain the vector. Markers incorporated in the vector may provide, for example, biocide resistance, e.g., resistance to antibiotics, or the like.
The transformed host cell can be fermented according to means known in the art to achieve optimal cell growth, and also to achieve optimal expression of the cloned fusion protein sequence fragments. Optimal expression of the fusion protein is expression which provides no more than the same moles of fusion protein subunit as the moles of the partner protein which are being expressed. However, variations in this amount are acceptable if they do not prevent the partner from forming heterodimers with the fusion protein, thereby interfering with fusion protein homodimer activity.
Any protein that possesses a binding domain which can form a heterodimer with the fusion protein will impair or prevent the formation of fusion protein homodimers. Such proteins can thus be identified by their ability to interfere with the phenotype conferred by the fusion protein homodimer.
In one embodiment the bacterial host, which is expressing a fusion protein as described above, is transformed with a ~ expression library capable of expressing cloned eukaryotic genes. Those cells transformed with a eukaryotic gene expressing a protein which is a partner of the fusion protein can then be detected due to loss of the phenotype conferred by the fusion partner homodimer.
~gtl 1 packaging systems for the creation of expression libraries from mRNA, which are useful in the methods of the invention, are known in the art and may be obtained commercially (for example, through Promega Corporation, Madison, Wisconsin). Further, custom genomic expression libraries may also be obtained commercially. Using the commercial kits, an oligo(dT)-primed cDNA library in ~gtll may be generated with the use of cytoplasmic poly(A)-containing mRNA from any desired mammalian source.
To induce expression of the cloned proteins contained therein, 10 mM IPTG
(isopropyl-thiogalactoside) may be added.

,. ~
, -16- 2~

A particular advantage of the method of the invention for the identification of protein partners is that, where approximately equal amounts of the fusion protein(s) and the protein partner are present in the host cell, the partner which is identified will have a higher affinity for the fusion protein(s) than the fusion protein(s) has to itself. If the disrupted dimeriza~ion is normally associated with a biological activity, such a protein partner is highly likely to be an important regulator of that biological activity. Further, the partner which is identified is already in a cloned, expressing form which may be utilized to obtain larger quantities of the protein for its isolation and further characterization by protein and molecular biology techniques known in the art.
Utilizing the above techniques, a chimeric peptide containing the bHLH
dimerization region of c-myc and the DNA binding domain of cI was constructed (see Example 1). In the appropriate host cell, this chimeric peptide formed homodimers and repressed expression of the l~acZ gènè under the control of a lambda PL promoter and repressed phage Iysis (see Example 2). Introduction of a partner protein into the host cell interfered with homodimer formation and de-repressed exp~ession of the lacZ gene (see Example 3). The inventors used this method to screen a cDNA expression library and discovered a specific partner protein which associates with c-myc in ViYo (see Example 3).
Compounds which inhibit the ability of protein partners to form interfering heterodimers, but which do not interfere with homodimer formation, may be identified by screening for the ability of a compound to reverse the interfering effect of the heterodimers and restore the homodimer-conferred phenotype.
For example, for partners identified by de-repression of the lacZ gene as described above (see also Example X), compounds which prevent or otherwise interfere with heterodimer formation of the protein partners can be identified by screening for the ability of such compounds to restore repression of the lacZ gene and cause partner-containing cells to remain white when grown on X-gal plates. A compound which is found to restore lacZ gene ,~ .

-17- ~18~07 repression in this example would be a compound which (a) prevents the fusion protein from associating with the partner peptide which is also being expressed in tEle host, (b) does not prevent homodimer formation and (c) does not inhibit cell growth.
The methods of the invention can be used to screen compounds in their pure form, at a variety of concentrations, and also in their impure form. The methods of the invention can also be used to identify the presence of such inhibitors in crude extracts, and to follow the purification of the inhibitors therefrom. The methods of the invention are also useful in the evaluation of the stability of the inhibitors identified as above, to evaluate the efficacy ofvarious preparations.
Analogs of such compounds which are more permeable across bacterial host cell membranes may also be used. For example, dibutyryl derivatives often display an enhanced permeability.
Partners, and compounds which inhibit the association of such partners, of any type of transcriptional regulation protein which associates into dimers may be identified by the bacterial methods of the invention. The methods of the invention can also be used to identify partners, and compounds which interfere with such partners, of membrane-localized and/or cytoplasmically-localiæd proteins which associate into dimers.
It may be desired to fur~her characterize the partner proteins of c-myc which are identified by the methods of the invention in a eukaryotic expression system. Such characterization may be performed according to the methods described in the inventor's copending U.S. patent application entitled "C-Myc Screening Assays," Serial 07/785,567 filed Oct. 30, I99I and incorporated herein by reference.
The following examples further describe the materials and methods used in carrying out the invention. The examples are not intended to limit the invention in any manner.

-18- 2~ 7 Examples E~cample I
Constr~ction of cl/c-myc Fusion Proteins Chimeric genes capable of expressing fusion pr~teins containing the DNA binding domain of the lambda repressor d and either 1) the c-myc basic helix-loop-helix (bHLH) dimerization domain or 2) the c-myc bHLH and leucine zipper (LZ) dimerization domains were constructed.
The promoterloperator region used consists of the ,B-lactamase promoter, lac operator and Shine-Delgarno (S.D.) sequence. The sequence is as follows:
GGA TCC TCT AAA TAC ATT CAA ATA AGT ATC CGC TCA TGA
BamHI - 3 5 CAC AAT AAC GGT AAC CAG AAT TGT GAG CGC TCA CAA TTT TG
-10 BstEII
ATC GAT AGC AAA CTC GAG ATG...... [Seq. ID No. Ç]
ClaI S.D. XhoI +l cI

The N-terminal 336 bp (112 amino acids) of d, which contains the DNA binding domain of this protein, was incorporated into this construct.
This portion was amplified for cloning using polymerase chain reaction with primers adding ,~oI and X~al sites on the 5' and 3' ends, respectively. The promoter/operator and cI DNA were cloned into pUC18 digested with BamHI
and XbaI to generate pUC3d.
The sequence around the XbaI site is as follows:
5 CAG GCA GGG TCT AGA . . . [ SEQ ID NO . 7 ]
Gln Ala Gly XbaI
cI coding seq.
The bHLH/LZ and bHLH fragments of c-myc were generated by PCR
using a human c-myc cDNA as a template. The bHLH/LZ fragment used was a 258 bp fragment synthesized with primers starting at sites #2 and #9 (Figure . .
,, ~ :
. , - . . :
, .:

-l9-1) with Xbal and SalI sites added at the S' and 3' ends, respectively. The -bHLH fragment used is a 165 bp fragment with Xbal and PstI sites added on the 5' and 3' ends, respectively. The boundaries of bHLH are at sites marked #2 and #10 (Figure 1). The primer used at site #10 included a termination codon, as does that used at site #9. Insertion of the c-myc sequences into pU3cl was at the restriction sites corresponding to those added by the indicated PCR primers. The resulting constructs containing c-myc bHLH/LZ
and bHLH were referred to as pU3.29 and pU3.210, respectively. As a result of the cloning procedure used, an X~7al site (TCT AGA) encoding amino acids Serine and Arginine was incorporated in-between the cl and c-myc sequences.
The chimeric cl/c-myc gene constructs in pUC18 were subcloned into pACYC177 (Chang, A.C.Y. et al., J. Bacteriol. 134: 1141-1156 (1978)) as follows. Both chimeric genes were excised from pUC18 by digestion with Hindlll, fill-in of the Hindlll overlap with Klenow, and subsequent BamHI
digestion. The chimeric gene fragments were then cloned into pACYC177 digested with Bgn (filled in with Klenow) and BamHI. The resulting constructs were designated pYC188 which contains cI-bHLH/LZ and pYC192 which contains cI-bHLH. These pYC-constmcts confer kanamycin resistance upon transformed E. coli host cells and are normally maintained in low copy number (5-20 copies/cell).

Example 2 Assaying transformed bacteria for the phenotype conferred by the cl/c-myc fusion protein in homodimer form The DNA binding domain of the d protein must be present in dimer form to function as a repressor of lambda transcription/infection. Native d protein is unable to form dimers at physiological levels and is therefore functionally inactive. In contrast, fusion proteins containing a functional DNA
binding domain from cI and a functional dimeriza~ion domain from c-myc should be able to form functional homodimer repressors. To detect the - 2~8~8~7 repressor phenotype in bacterial cells transformed with the cl/c-myc fusion constructs described in Example 1, two different assays were used.
In the "dot plaque assay" (DPA), transformed E. coli cells were tested for susceptibility to lambda phage infection. These cells were predicted to be S resistant to infection if the cl/c-myc fusion protein was adequately expressed and formed functional homodimers. Cell strains carrying fusion protein constructs were grown in L-broth media containing 30 ~g/ml Kanamycin, 10 mM MgSO4, and 0.2% maltose at 37C. 0.25-0.5 ml of culture at an OD60"
of 1.0 to 2.0 was added to 3mls of 48C top agar, mixed by vortexing and plated on pre-warmed L-broth/Kanamycin plates. The top agar was allowed to solidify for 2-3 min. at room temperature and then 5~41 aliquots of lambda phage KH54 (provided by J. Hu and R. Sauer of the Massachussetts Institute of Technology) of titer 5x101-5x106 plaque forming units (pfu) were dotted onto the top agar. Lambda phage KH4i434 (provided by J. Hu and R. Sauer of the Massachussetts Institute of Technology), which carries the immunity region of phage 434 and is therefore not affected by lambda cI, was also dotted on as a control. Phage aliquots were allowed to dry and then the plates were incubated overnight at 37C.
In this assay, the titer of phage required to create a clear spot is used as a measure of phage resistance. Bacteria that express native cI protein (which is unable to form dimers) from pACYC177 (Chang, A.C.Y. et al., J.
Bacteriol. 134: 1141-1156 (1978)) are not resistant and clearing can be seen at < 102 pfu. In contrast, bacterial strains containing pYC188 and expressing the cl-bHLH/LZ fusion protein are resistant up to 105-106 pfu and those containing pYC192 and expressing the cl-bHLH are resistant up to 107 pfu.
This resistance demonstrates the ability of the cl/c-myc fusion proteins to dimerize and effectively repress phage transcription/infection.
In the second assay, referred to herein as the X-gal assay, cells transformed with the cl/c-myc fusion construct pYC192 also contained a chimeric lacZ gene under the control of the lambda PL promoter. In these host cells, expression of functional cl/c-myc fusion protein would be expected .

..

2 ~

to repress lacZ expression via repression of the lambda PL promoter.
Expression of the lacZ gene is easily detectable by growth of cells on X-gal corltaining media. Cells expressing lacZ become blue on this media while nonexpressing or poorly expressing cells become white or pale blue, respectively.
As expected, those cells transformed with pYC192 grew as white or pale blue colonies due to repression of the PL-IacZ gene while nontransformed cells grew as blue colonies due to expression of the PL-lacZ gene.

Exan~ple 3 Screening a cDNA expression library for protein partners able to form heterodimers with the cl/c-myc fusion protein Interference with dimerization by direct protein-protein interaction between the dimerization domain of the chimeric repressor and a cDNA-encoded protein is the basis for the screening system of the invention. Upon dimerization of a repressor monomer with a heterologous protein partner, which is not part of a cl fusion, the repressor chimera will be inactivated, as it is unable to bind DNA as a monomer.
The dot plaque assay (DPA) and X-gal assay, described in Example 2, were used in the screening system. Bacteria expressing cl/c-myc fusion proteins and exhibiting the homodimer conferred repression phenotype (either phage resistance or repression of PL-IacZ expression) were used.

Screening with the Dot Plaque Assay For the DPA, E. coli strain Y1090 (available from Promega Corpora-tion, Madison, Wisconsin) expressing the chimeric repressor cllc-myc, which are resistant to infection by )~gtll, were used. Using this bacterial stMin, ~gtl 1 phage cDNA libraries, expressing cDNA encoded proteins as C-terminal fusions with lacZ, were screened. Only those phage containing a cDNA

.

encoding a protein partner were predicted to form plaques due to interference with cl/c-myc repressor/homodimer formation.
~gtll libraries were screened as follows. Bacterial strain Y1090 containing pYC192 and expressing cl-mycbHLH was grown in L-broth media containing Kanamycin, Mg2+, and maltose essentially the same as for the DPA described in Example 2. 0.6 mls of culture were exposed to lx106-5X106 pfu of i~gt11 library phage in liquid for 20 min. at 30C and then mixed with 7 mls of top agar and poured on 150 mm L-broth/Kanamycin plates.
Plates were incubated overnight at 42C. Four cDNA libraries were screened:
one from HeLa cells, one from T cell line EL4, one from the pre-B cell line 38B9, and one from primary tonsil cells which are almost exclusively B cells.
These libraries were respectively obtained from T. Kadesch at the University of Pennsylvania, K. Georgopolas of Massachussetts General Hospital, D.
Weaver of the Dana Farber Cancer Institute (DFCI), and T. Tedder of DFCI.
The lambda cDNA libraries were also plated onto a bacterial strain expressing a chimeric repressor from plasmid PJH370 (Hu, l.C. etal., Science 250:1400-1403 (199~)) containing the d DNA binding domain and the leucine zipper dimeriMtion domain of the yeast transcription factor GCN4.
For initial screenings, this strain acted as a comparison control to provide a baseline number of false positive plaques to be expected from the libraries resulting from phage mutations rendering them insensitive to the d repressor.
This strain also acted as a control for subsequent screening of put~tive positive plaques to determine if interference was due to a specific interaction with the cI/c-myc fusion protein or if the interference was of a more general nature, affecting the GCN4 dimeriMtion domain as well.
For all libraries screened, essentially equal numbers of plaques were observed with the strain containing cI-GCN4 vs. the strain containing pYC192 (cl-myc), indicating that these plaques were likely to be false positives. The number of plaques obtained varied from 5 to approximately 250, depending on the library used. Ninety phage which formed plaques on the strain containing pYCl92 were plaque purified and subsequently screened on the cl-., ' ~ "' 2~g~

GCN4 containing strain. All these phage again formed pla~ues, indicating that they did not specifically interact with the cl/c-myc fusion protein.
In light of these results, a subsequent experiment was performed to determine if a known protein partner could be detected with this screeening procedure. In this experiment a bacterial strain expressing a cl/c-myc fusion protein was challenged with a ~gtl 1 phage expressing a Max DNA. Max is a bHLH/LZ protein known to interact with c-myc. The challenged cells exhibited full resistance to the Max ~gtl 1 phage.
In contrast to these results, DPA screening for a predicted protein partner introduced before phage infection was succesful. In this experiment, a pUC18 plasmid capable of expressing a protein containing the bHLH/LZ
domains of c-myc, but not the DNA binding domain of d, was introduced into a bacterial strain which already contained a pACYC177 plasmid capable of expressing a cl/c-myc fusion protein. The protein containing the bHLH/LZ
domains was predicted to function as a partner to the cI/c-myc fusion protein and interfere with the repression of phage infection. As predicted, cells expressing cl/c-myc and the bHLH/LZ protein were approximately 100-fold less resistant to phage infection than cells expressing cl/c-myc only, as measured by the DPA.
These results indicate that the DPA can be used to screen for protein partners, but that the protein partner must be expressed in the bacteria before it is challenged with phage. Simultaneous introduction of the protein partner gene with the challenging phage, as occurs in the direct screen, probably does not work because the phage is effectively repressed before the protein partner gene is given a chance to express and interfere with the cl/c-myc fusion protein repressor.

Screening with the X-Gal Assay As described above, in cells with an active cltc-myc repressor the PL-lacZ gene is turned off resulting in the generation of white colonies on X-gal , 2~8180~

indicator plates. Interference with repressor dimerization is predicted to yieldblue colonies as the lacZ gene would be expressed (de-repressed).
Screening using the X-gal assay was performed as follows. The strain Y1090 was transformed with the lacZ target plasmid pNNPL387 and pYC188 which expresses cl-bHLH/LZ. The plasmid pNNPL387 was constructed by inserting a PCR generated DNA fragment containing the left promoter of phage lambda upstream of lacZ in pNN387 (provided by S. Elledge, Baylor University, See Elledge, S.J. et al., Genes & Develop. 3: 185-1g7 (1989)).
These cells, when plated on L-broth/Kanamycin/Chloramphenicol/X-gal, form white to pale blue colonies. This strain, referred to as 10B18, was made competent for electroporation (see Current Protocols in Molecular Biology,sec.
1.8.4, Wiley Interscience, ed. by Ausubel et al. (1987)) and transformed with a plasmid-based cDNA library made from human peripheral blood Iymphocytes which had been transformed with Epstein-Barr Virus (provided by S. Elledge, See Elledge, S.J. et al., Proc. Natl. Acad. Sci. USA 88: 1731-1735 (1991)). There were about 107 recombinants in this once-amplified library.
10B18 was electropoMted on two separate occasions with 500 ng of library DNA. Cells were allowed to recover from electroporation for 45 min.
at 37C in SOC media and then plated on M9/0.2 % mannitol with Chloramphenicol (20~g/ml), Kanamycin (30~g/ml), IPrG (2mM), Ampicillin (50~g/ml), and X-gal (0.004%). The electroporations yielded 2.8x106 and 5.6x105 tsansformants, of which approximately 500 and 29, respectively, were blue. A total of 322 blue colonies were picked and restreaked to isolate single colonies. From these, 97 blue clonal colonies were isolated and plasmid DNAs were prepared. Plasmid DNA from each clone was then retransformed into 10B18 and plated as above. Only one clone consistently produced blue colonies.
This clone was shown to be Tecific for c-myc by comparing the phenotypes it produced in different repressor chimera backgrounds. Bacterial strains similar to 10B18 which contain different cl-dimerization domain fusion 2~81~7 constructs were used. These strains express d fusions with the c-myc bHLH
domain (lOB19), the transcription factor E2/SbHLH domain (lOBE2/5), or thyroid hormone receptor ,B (lOB,B). The positive clone isolated in the originalscreen produced blue colonies only in 10B18 and 10B19, where dimerization S was mediated by a c-myc domain. Strains 10BE2/S and 10B,6' remained white on X-gal plates after transformation with this clone.
The high number of false positives obtained during the initial rounds of screening could be due to the instability of the plasmid containing the chimeric repressor gene in the screening strain. Alternatively, blue colonies could result from an increase in the copy number of the PL-lacZ containing plasmid or increased expression of the PL-IacZ gene which titrates out repressor dimers. Whatever the cause, repeated passages through the 10B18 strain was effective in screening out false positives.

Example 4 Identification of compounds which prevent c-myc partnerfor~na~ion To identify compounds which inhibit c-myc partner heterodimerization without interfering with c-myc homodimerization, cells identified according to the method described in Example 3 which contain the cllc-myc fusion protein and a partner protein are used along with cells containing only the cl/c-myc fusion protein as described in Example 2. These cells are further exposed to experimental compounds W, X, Y, and Z and the effect of such compounds on the homodimer/heterodimer dependent phenotype is determined.
Typical results from such an experiment are shown in Table 1.

.

.

~Q8~07 Table 1. Identification of C-myc-protein Partner Inhibitors Compound ¦ Protein Partner ¦ Assay Phenotype none no plaques/white _ + plaques/blue W no plaques/white + plaques/blue X ~ plaques/blue + plaques/blue Y no plaques/white _ + no plaques/white The results of the above table indicate that, in the absence of the partner protein, compound W had no effect on the ability of the cIlc-myc protein to form homodimers and exhibit the corresponding phenotype.
Compound W also had no effect on the ability of the partner to form heterodimers with the myc fusion protein and reverse the homodimer-conferred phenotype. Therefore, compound W will not be a compound of interest.
Compound X interfered with homodimer formation and therefore will not be a compound of interest.
Compound Y is an inhibitor of heterodimer formation. Compound Y
did not interfere with homodimer formation but did interfere with hete.rodimer formation. Therefore, compound Y is a compound of interest as it may disrupt c-myc action in vivo.

All references cited herein are fully incorpoMted by reference. Having now fully described the invention, it will be understood by those with skill in the art that the scope may be performed within a wide and equivalent range of conditions, paMmeters and the like, without affecting the spirit or scope of the invention or any embodiment thereof.

, .
:.

-27- 2 0 ~ 7 SEQUENCE LISTING

(1) GENERAL INFORMATION:
(i) APPLICANT: Kinqston, Robert E.
Bunker, Christopher (ii) TITLE OF INVENTION: Protein Partner Screening Assays and Uses Thereof (iii) NUMBER OF SEQUENCES: 7 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: Sterne, Ressler, Goldstein and Fox (B) STREET: 1225 Connecticut Avenue (C) CITY: Washington (D) STATE: D.C.
(E) COUNTRY: U.S.A.
(F) ZIP: 20036 (v) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) CONPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release #1.0, Version #1.25 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NVMBER: US (to be assigned) (B) FILING DATE: (herewith) (C) CLASSIFICATION:
~viii) ATTORN~Y/AGENT INFORMATION:
(A) NAME: Cimbala, Michelle A.
(B) REGISTRATION NUMBER: 33,851 (C) REFERENCE/DOCRET NUMBER: 0609.3630004 (ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (202) 833-7533 (B) TELEFAX: (202) 833-8716 (2) INFORMATION FOR SEQ ID NO:l:
(i) SEQVENCE CHARACTERISTICS:
(A) LENGTH: 11 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

(2) INFORMATION FOR SEQ ID NO:2:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 10 base pairs (B) TYPE: nucleic acid (c) STRANDEDNESS: double .
.

2 ~

(D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

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

(2) INFORMATION FOR SEQ ID NO:3:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 10 ba~e pair#
(8) TYPE: nucleic ac~d (C) STRaNDEDNESS- double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

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

(2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1419 ba~e pair~
(B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:
AGGAGGAACA AGAAGATGAG GAAGA~ATCG ATGTTGTTTC TGTGGA~AAG AGGCAGGCTC 60 CTGGCAaAAG GTCAGAGTCT GGATCACCTT CTGCTGGAGG CCACAGGAaA CCTCCTCACA 120 TGAG~CAGAT CAGCAACAAC CGAAAATGCA CCAGCCCCAG GTCCTCGGAC ACCGAGGAGA 300 ATGTCAAGAG GCGAACACAC AACGTCTTGG AGCGCCAGAG GAGGAACGAG CTAaAACGGA 360 GCTTTTTTGC CCTGCGTGAC CAGATCCCGG AGTTGGAAaA CAATGA~AAG GCCCCCAAGG 420 TAGTTATCCT TAAAAAAGCC ACAGCATACA TCCTGTCCGT CCAAGCAGAG GAGCA~AAGC 480 TCATTTCTGA AGAGGACTTG TTGCGGA~AC GACGAGAACA GTTGAaACAC A~ACTTGAAC 540 AGCTACGGAA CTCTTGTGCG TAAGGAAAAG TAAGGAaaAC GATTCCTTCT AACAGAaATG 600 .

- 2~818~7 TA~ATGTAAA TAACTTTAAT AAAACGTTTA TAGCAGTTAC ACAGAATTTC AATCCTAGTA 900 AGTTTTCCTC TGTTGAAATG GGTCTGGGGG CCTTAAGGTC TTTA~GTTCT TGGAGGTTCT 1200 ~2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH:.186 amino acid~
~B) TYPE: amino acid (D) TOPOLOGY: both (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:
Glu Glu Gln Glu Asp Glu Glu Glu Ile Asp Val Val Ser Val Glu Lys Arg Gln Ala Pro Gly Lys Arg Ser Glu Ser Gly Ser Pro Ser Ala Gly Gly His Ser Lys Pro Pro His Ser Pro Leu Val Leu Lys Arg Cys His Val Ser Thr His Gln His Asn Tyr Ala Ala Pro Pro Ser Thr Arg Lys Asp Tyr Pro Ala Ala Lys Arg Val Ly~ Lsu Asp Ser Val Arg Val Leu Arg Gln Ile Ser Asn Asn Arg Lys Cys Thr Ser Pro Arg Ser Ser Asp Thr Glu Glu Asn Val Lys Arg Arg Thr His Asn Val Leu Glu Arg Gln Arg Arg Asn Glu Leu Lys Arg Ser Phe Phe Ala Leu Arg Asp Gln Ile Pro Glu Leu Glu Asn Asn Glu Lys Ala Pro Lys Val Val Ile Leu Lys `

2~18~7 Lys Ala Thr Ala Tyr Ile Leu Ser Val Gln Ala Glu Glu Gln Lys Leu Ile Ser Glu Glu Asp Leu Leu Arg Ly~ Ar3 Arg Glu Gln Leu Lys His Lys Leu Glu Gln Leu Arg Asn Ser Cys Ala (2) INFORMATION FOR SEQ ID NO:6:
(i) SEQUENCE CHARACTERISTICS:
~A) LENGTH: 101 base pairs (B) TYPE: nucleic acid tc) STRaNDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA

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

(2) INFORMATION FOR SEQ ID NO:7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 15 ba~e pair~
(B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA -(xi) SEOUENCE DESCRIPTION: SEQ ID NO:7:

- ~ :
, ~ .

Claims (15)

1. A method for identifying and classifying a protein partner wherein said method comprises:
(a) transformation of a host cell with a genetic construct capable of expressing a fusion protein, wherein said fusion protein contains a DNA
binding domain and a dimerization domain complementary to itself, and wherein said fusion protein forms a homodimer which confers a detectable phenotype upon said host cell;
(b) transformation of said host cell of part (a) with a genetic construct capable of expressing said protein partner;
(c) culturing said host cell of part (b) under conditions which express said fusion protein and said protein partner, said protein partner beingexpressed at levels equivalent to or greater than said fusion protein;
(d) determining whether the phenotype conferred by said fusion protein of part (a) is present in said host cell of part (c); and (e) classifying said protein partner on the basis of the presence or absence of said phenotype.

2. A method of identifying and classifying a compound as an inhibitor of a protein partner, wherein said method comprises:
(a) transformation of a bacterial host cell with a genetic construct capable of expressing a fusion protein, wherein said fusion protein contains a DNA binding domain and a dimerization domain complementary to itself, and wherein said fusion protein forms a homodimer which confers a detectable phenotype upon said host cell;
(b) transformation of said host cell of part (a) with a genetic construct capable of expressing said protein partner;
(c) culturing said host cell of part (b) in the presence of said compound and under conditions which express said fusion protein and said protein partner, said protein partner being expressed at levels equivalent to orgreater than said fusion protein;
(d) determining the ability of said compound to prevent protein-partner-induced interference of the phenotype conferred by said fusion protein of part (a); and (e) classifying said compound as an inhibitor of protein partner formation on the basis of the presence or absence of said phenotype.

3. The method of any one of claims 1 or 2, wherein said phenotype conferred by said fusion protein in homodimer form is the repression of expression of an assayable marker gene.

4. The method of claim 3, wherein said assayable marker is under the transcriptional control of the bacteriophage .lambda. PL promoter.

5. The method of claim 4, wherein said assayable marker is the lacZ
gene.

6. The method of any one of claims 1 or 2, wherein said DNA
binding domain of said fusion protein is the DNA binding domain of bacteriophage .lambda. cl repressor protein.

7. The method of claim 6, wherein said DNA binding domain of said cl repressor protein is the N-terminal 112 amino acids of said repressor protein.

8. The method of any one of claims 1 or 2, wherein said dimerization domain is a bHLH domain.

9. The method of claim 8, wherein said bHLH domain is from myc.

10. The method of claim 9, wherein said myc is c-myc.

11. The method of claim 10, wherein said bHLH domain is amino acids 255-410 of c-myc.

12. The method of any one of claims 1 or 2, wherein said dimerization domain is a bZIP domain.

13. The method of any one of claims 1 or 2, wherein said dimerization domain is a zinc finger domain.

14. A method for identifying and classifying a protein partner wherein said method comprises:
(a) transformation of a bacterial host cell with a genetic construct capable of expressing a first fusion protein and a second fusion protein, wherein said first fusion protein contains a DNA binding domain and a first dimerization domain, and wherein said second fusion protein contains said DNA binding domain and a second dimerization domain complementary to said first dimerization domain, and wherein said first fusion protein and said second fusion protein form a DNA binding domain homodimer which confers a detectable phenotype upon said host cell;
(b) transformation of said host cell of part (a) with a genetic construct capable of expressing said protein partner;
(c) culturing said host cell of part (b) under conditions which express said first fusion protein, said second fusion protein, and said protein partner, said protein partner being expressed at levels equivalent to or greaterthan either said first fusion protein or said second fusion protein;
(d) determining whether the phenotype conferred by said DNA
binding domain homodimer of part (a) is present in said host cell of part (c);
and (e) classifying said protein partner on the basis of the presence or absence of said phenotype.

15. A method of identifying and classifying a compound as an inhibitor of a protein partner, wherein said method comprises:
(a) transformation of a bacterial host cell with a genetic construct capable of expressing a first fusion protein and a second fusion protein, wherein said first fusion protein contains a DNA binding domain and a first dimerization domain, and wherein said second fusion protein contains said DNA binding domain and a second dimerization domain complementary to said first dimerization domain, and wherein said first fusion protein and said second fusion protein form a DNA binding domain homodimer which confers a detectable phenotype upon said host cell;
(b) transformation of said host cell of part (a) with a genetic construct capable of expressing said protein partner;
(c) culturing said host cell of part (b) in the presence of said compound and under conditions which express said first fusion protein, said second fusion protein, and said protein partner, said protein partner being expressed at levels equivalent to or greater than either said first fusion protein or said second fusion protein;
(d) determining the ability of said compound to prevent protein-partner-induced interference of the phenotype conferred by said DNA binding domain homodimer of part (a); and (e) classifying said compound as an inhibitor of protein partner formation on the basis of the presence or absence of said phenotype.