Interaction trap systems for analysis of protein networks

Mammalian protein expression systems are the best choice for the production of eukaryotic proteins, especially when correct folding and post-translational modification is required. They produce eukaryotic recombinant proteins in the most natural state, with native tertiary structure, physiochemical characteristics and bioactivities. They have been successfully applied in the biopharmaceutical production of cytokines, monoclonal antibodies, growth factors and so on. Email:marketing@medicilon.com.cn web:www.medicilon.com

Disclosed are improved methods for detecting protein-protein interactions. These methods involve either a determination of whether three or more proteins are capable of interacting in a trimeric or higher order complex or whether two or more mammalian proteins interact, the latter method utilizing yeast mating to bring candidate proteins into contact with one another.

In general, in one aspect, the invention involves a method for determining whether at least three proteins are capable of interacting with each other. This method involves: (a) providing a first host cell which contains (i) a reporter gene operably linked to a protein binding site; (ii) a first fusion gene which expresses a first fusion protein, the first fusion protein including a first protein covalently bonded to a binding moiety which is capable of specifically binding to the protein binding site; and (iii) a second fusion gene which expresses a second fusion protein, the second fusion protein including a second protein covalently bonded to a gene activating moiety; (b) measuring reporter gene expression in the first host cell, an increase in expression indicating an interaction between the first and second proteins; (c) providing a second host cell which contains (i) the reporter gene; (ii) the first fusion gene; and (iii) a third fusion gene which expresses a third fusion protein, the third fusion protein including a third protein covalently bonded to a gene activating moiety; (d) measuring reporter gene expression in the second host cell, an increase in expression indicating an interaction between the first and third proteins; (e) constructing a fourth fusion gene which expresses a fourth fusion protein, the fourth fusion protein including the protein binding site covalently bonded to a second protein shown in step (b) to be capable of interacting with the first protein; (f) introducing into a third host cell (i) the reporter gene; (ii) the third fusion gene; and (iii) the fourth fusion gene; and (g) measuring reporter gene expression in the third host cell, an increase in expression indicating an interaction between the second and third proteins.

In a second aspect, the invention involves a method for determining whether a first mammalian protein is capable of interacting with a second mammalian protein. This method involves: (a) providing a first yeast cell which contains (i) a reporter gene operably linked to a protein binding site; and (ii) a first fusion gene which expresses a first fusion protein, the first fusion protein including the first mammalian protein covalently bonded to a binding moiety which is capable of specifically binding to the protein binding site; (b) providing a second yeast cell which contains a second fusion gene which expresses a second fusion protein, the second fusion protein including the second mammalian protein covalently bonded to a gene activating moiety; (c) mating the first yeast cell with the second yeast cell; and (d) measuring reporter gene expression, an increase in expression indicating an interaction between the first and the second mammalian proteins.

In preferred embodiments of both of the above methods, the gene activating domain is a weak gene activating domain; the host cells contain a second reporter gene, the second reporter gene being different than the first reporter gene (for example, a LEU2 gene and a lacZ gene); the host cell is a yeast cell; the protein binding site is a LexA binding site and the binding moiety comprises a LexA DNA binding domain; the reporter gene is assayed by a color reaction; the reporter gene is assayed by cell viability.

In a third aspect, the invention features a set of DNA molecules, each molecule encoding a candidate interacting protein fused to a DNA binding domain to which it is not naturally bonded.

In a fourth aspect, the invention features a set of DNA molecules, each molecule encoding a candidate interacting protein fused to a weak gene activating domain to which it is not naturally bonded.

In preferred embodiments of both the third and fourth aspects, each of the candidate interacting proteins is involved in signal transduction; each of the candidate interacting proteins is a cytokine; each of the candidate interacting proteins is involved in DNA replication; each of the candidate interacting proteins is involved in a function occurring in the cell nucleus; and each of the candidate interacting proteins is involved in intermediary metabolism.

In a fifth aspect, the invention features a set of eukaryotic cells (preferably, yeast cells), each cell containing one of the set of DNA molecules of the third and fourth aspects.

In a sixth aspect, the invention features a pair of haploid yeast cells, the first cell of the pair containing (a) a reporter gene operably linked to a protein binding site; and (b) a first fusion gene which expresses a first fusion protein, the first fusion protein including a first protein covalently bonded to a binding moiety which is capable of specifically binding to the protein binding site; and the second cell of the pair containing a second fusion gene which expresses a second fusion protein, the second fusion protein including a second protein covalently bonded to a gene activating moiety. In preferred embodiments, the second cell further contains a second reporter gene, said second reporter gene being different than the first reporter gene.

As used herein, by "reporter gene" is meant a gene whose expression may be assayed; such genes include, without limitation, lacZ, amino acid biosynthetic genes, e.g. the yeast LEU2, HIS3, LYS2, TRP1, or URA3 genes, nucleic acid biosynthetic genes, the mammalian chloramphenicol transacetylase (CAT) gene or GUS gene, or any surface antigen gene for which specific antibodies are available. Reporter genes may encode any enzyme that provides a phenotypic marker, for example, a protein that is necessary for cell growth or a toxic protein leading to cell death, or one encoding a protein detectable by color assay or one whose expression leads to an absence of color. Particularly preferred reporter genes are those encoding fluorescent markers, such as the GFP gene (i.e., Green Fluorescent Protein gene). Reporter genes may facilitate either a selection or a screen for reporter gene expression, and quantitative differences in reporter gene expression may be measured as an indication of interaction affinities.

By "operably linked" is meant that a gene and a regulatory sequence(s) are connected in such a way as to permit gene expression when the appropriate molecules (e.g., transcriptional activator proteins or proteins which include transcriptional activation domains) are bound to the regulatory sequence(s).

By "covalently bonded" is meant that two domains are joined by covalent bonds, directly or indirectly. That is, the "covalently bonded" proteins or protein moieties may be immediately contiguous or may be separated by stretches of one or more amino acids within the same fusion protein.

By "protein" is meant a sequence of amino acids, constituting all or a part of a naturally-occurring polypeptide or peptide, or constituting a non-naturally-occurring polypeptide or peptide.

By a "binding moiety" is meant a stretch of amino acids which is capable of directing specific polypeptide binding to a particular DNA sequence (i.e., a "protein binding site").

By "weak gene activating moiety" is meant a stretch of amino acids which is capable of weakly inducing the expression of a gene to whose control region it is bound. As used herein, "weakly" is meant below the level of activation effected by GAL4 activation region II (Ma and Ptashne, Cell 48:847, 1987) and is preferably at or below the level of activation effected by the B112 activation domain of Ma and Ptashne. One preferred weak gene activating moiety is the B42 domain of Ma and Ptashne (supra). Levels of activation may be measured using any downstream reporter gene system and comparing, in parallel assays, the level of expression stimulated by the GAL4 region II-polypeptide with the level of expression stimulated by the polypeptide to be tested.

By "set" is meant 5 or more (as used herein) DNA molecules or eukaryotic cells. Preferably, such a set includes at least 10 different DNA molecules or eukaryotic cells, more preferably 25 different DNA molecules or eukaryotic cells, and most preferably 100's or even 1000's of different DNA molecules or eukaryotic cells.

The interaction trap systems described herein provide advantages over more conventional methods for isolating interacting proteins or genes encoding interacting proteins. For example, applicants' systems provide rapid and inexpensive methods having very general utility for identifying and purifying genes encoding a wide range of useful proteins based on the protein's physical interaction with a polypeptide of known diagnostic or therapeutic usefulness. This general utility derives in part from the fact that the components of the systems can be readily modified to facilitate detection of protein interactions of widely varying affinity. Inducible promoters used to express the interacting proteins further increase the scope of candidate interactors which may be detected since even proteins whose chronic expression is toxic to the host cell may be isolated simply by inducing a short burst of the protein's expression and testing for its ability to interact and stimulate expression of a reporter gene.

If desired, detection of interacting proteins may be accomplished through the use of weak gene activation domain tags. This approach avoids restrictions on the pool of available candidate interacting proteins which may be associated with stronger activation domains (such as GAL4 or VP16); although the mechanism is unclear, such a restriction apparently results from low to moderate levels of host cell toxicity mediated by the strong activation domain.

In addition, certain of the claimed methods facilitate the ready identification of higher order protein interactions, for example, protein interactions involving three or more polypeptides. Certain other claimed methods exploit a yeast mating assay to rapidly screen for interactions among extremely large numbers of mammalian proteins. According to this method, an uncharacterized protein is introduced by mating into an extensive panel of yeast strains, each carrying a different putative interacting protein, and interactions are identified, for example, through visual screening. Identification of protein interactions provides important knowledge regarding the protein's function and also provides immediate access to mammalian proteins of interest.

Other features and advantages of the invention will be apparent from the following detailed description thereof, and from the claims. These examples are designed to illustrate, not limit, the invention.