Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/oxford-university-press/yeast-forward-and-reverse-n-hybrid-systems-S0b39rJVqI
References (2)
-
E. Golemis, V. Khazak (1997)
Alternative yeast two-hybrid systems. The interaction trap and interaction mating.Methods in molecular biology, 63
-
P. Bartel, S. Fields (1997)
The yeast two-hybrid system
- Publisher
- Oxford University Press
- Copyright
- © 1999 Oxford University Press
- ISSN
- 0305-1048
- eISSN
- 1362-4962
- DOI
- 10.1093/nar/27.4.919
- Publisher site
- See Article on Publisher Site
Abstract
1999 Oxford University Press Nucleic Acids Research, 1999, Vol. 27, No. 4 919–929 SURVEY AND SUMMARY Marc Vidal* and Pierre Legrain MGH Cancer Center, Charlestown, MA 02129, USA and Laboratoire du Métabolisme des ARNs, CNRS URA 1300, Institut Pasteur and Hybrigenics, 75724 Paris Cedex 15, France Received July 27, 1998; Revised and Accepted November 4, 1998 ABSTRACT recent co-immunoprecipitation experiments using antibodies raised against one of its subunits gave rise to a 55 subunit complex Since its original description almost 10 years ago, the required for accurate transcriptional regulation activity (4). yeast two-hybrid system has been used extensively to Similarly, nuclear pre-mRNA splicing requires a highly dynamic identify protein–protein interactions from many different and organized molecular machine composed of five small nuclear organisms. Simultaneously, a number of ‘variations on RNAs and >50 polypeptides (reviewed in 5,6). Second, the a theme’ based on the original concept have been transmission of regulatory signals, originally described as described. In one set of variations, systems were successive catalytic activities required for the amplification of weak developed to detect other macromolecular interactions: inputs into cellular responses, now appears to be mediated by protein DNA–protein (one-hybrid), RNA–protein (RNA-based complexes structurally constrained around scaffolding proteins three-hybrid) and small molecule–protein interactions (7–9). Finally, the widely accepted concept of transient enzyme– (ligand-based three-hybrid). These different versions substrate interactions is being challenged by the identification of are collectively referred to here as ‘n-hybrid systems’. enzymes which stably associate with their substrates. For In another set of variations, the original configuration example, the cyclin-dependent kinases (CDK) required for cell of the two-hybrid fusion proteins was modified to expand cycle regulation have been found stably associated with some of the range of possible protein–protein interactions that their cognate substrates (10–13). Thus, the identification of could be analyzed. For example, systems were devel- protein–protein interactions remains fascinating and very helpful oped to detect trimeric interactions, ligand–receptor in understanding biological phenomena. interactions or interactions that require particular The identification of potential protein–protein interactions post-translational modifications. Finally, the original leads to hypotheses that need to be tested in the relevant biological concept was turned upside down and ‘reverse n-hybrid systems. For such functional analyses, the isolation of mutant systems’ were developed to identify mutations, peptides proteins specifically altered in their ability to interact with a or small molecules that dissociate macromolecular potential partner (‘interaction-defective alleles’) can be critical. interactions. These reagents can be used to validate, In these experiments, the interaction-defective mutant proteins in the relevant biological systems, the potential inter- are compared with their wild-type counterpart in a functional actions identified with the ‘forward n-hybrid systems’. assay. Interaction-defective mutant proteins are predicted to The powerful genetic selections of the forward and exhibit functional defects if their corresponding wild-type reverse n-hybrid systems are proving useful in pro- versions indeed function in relevant protein–protein interactions. teomic projects aimed at generating macromolecular This strategy is exemplified in the characterization of the interaction maps. association between the adenovirus E1A oncoprotein and the retinoblastoma gene product (pRB) (14). To demonstrate the INTRODUCTION physiological relevance of this potential interaction, interaction- defective alleles of E1A were generated and shown to affect E1A’s Protein–protein interactions are critical to most biological ability to transform cells (15). In more recent developments of this processes, extending from the formation of cellular macromolecular strategy, compensatory mutations that restore the interaction were structures and enzymatic complexes to the regulation of signal isolated in the potential interaction partner. These compensatory transduction pathways. This statement is becoming somewhat changes are expected to restore the function mediated by the superfluous since in the last 20 years large numbers of stable interaction. For example, in the case of the yeast splicing factor interactions were uncovered and shown to be biologically composed of Prp9p, Prp11p and Prp21p, it had been demonstrated relevant, in practically every field of molecular biology (1,2). Nevertheless, this statement is still worth mentioning in light of that prp9 and prp11 thermosensitive mutations disrupt the recently published observations that implicate stable protein–protein interaction with Prp21p. Subsequently, prp21-91, an extragenic interactions in unexpected mechanisms. First, many enzymatic suppressor mutation of the prp9 thermosensitive phenotype, was activities are mediated by complexes much larger than originally shown to restore Prp9–Prp21 interaction (16,17). anticipated. For example, although the original purification Prior to the development of the yeast two-hybrid system, both procedure of RNA polymerase II led to a 12 subunit complex (3), the identification of physical protein–protein interactions and *To whom correspondence should be addressed. Tel: +1 617 724 3404; Fax: +1 617 726 7808; Email: [email protected] Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 920 Nucleic Acids Research, 1999, Vol. 27, No. 4 their subsequent functional characterization traditionally relied particular mutant (23,24). The two-hybrid system represents the upon time and labor intensive biochemical approaches. For ultimate example of such a strategy since it was designed from the example, the molecular cloning of genes encoding interacting outset for the detection of essentially any protein–protein interaction, proteins identified biochemically is often difficult, as is the independent of the function of the corresponding proteins. identification of interaction-defective alleles. Moreover, while The basic concept of the two-hybrid system emerged from the protein–protein interactions are important, many biological analysis of transcription factors such as the archetypal yeast processes rely upon other macromolecular interactions such as Gal4p. These transcription factors increase the rate of transcription DNA–protein and RNA–protein interactions. To understand how of their target genes by binding to upstream activating DNA cis-acting DNA sites are involved in transcription or replication sequences (UAS) and thus ‘activating’ RNA polymerase II at the control and to address how particular RNA molecules are involved corresponding promoters. It was demonstrated that the DNA in splicing, translation or development, one powerful approach is to binding and the activating functions are located in physically identify the protein(s) that stably binds these nucleic acids. separable domains of Gal4p (25). These two domains are referred As will be described in this review, the yeast one-, two- and to as the DNA-binding domain (DB) and the activation domain three- (or ‘n’-) hybrid systems in a forward or reverse configuration (AD), respectively. In the most extreme version of such provide genetic solutions to bypass the problems inherent in structure–function experiments, a hybrid protein, consisting of biochemical approaches, for both the identification of potential the bacterial LexA DB fused to the Gal4p AD, was shown to interactions and their characterization. activate, in yeast cells, the transcription of a bacterial reporter gene containing the LexA operator site in its promoter (26). In addition, the fusion between DB and AD was shown to exhibit an HISTORICAL PERSPECTIVE unexpected level of structural flexibility. For example, 1% of random bacterial DNA sequences were found to be capable of The yeast two-hybrid system presents three major advantages encoding a functional AD when fused to the Gal4p DB (DB–AD) over alternative assays for gene identification. First, since it is (27). This observation suggested, rather surprisingly, that the based on a powerful genetic selection scheme performed with a structural constraints of transcription factors for correctly activating convenient microorganism, it allows very high numbers of RNA polymerase II upon DNA binding can be extremely loose. potential coding sequences to be assayed in a relatively simple Furthermore, it was shown that functional DB–AD fusions need not experiment. Second, it relies on an assay performed in vivo and be covalent. For example, functional dissection of VP16, the thus it is not limited by the artificial conditions of in vitro assays. trans-activator of herpes simplex virus, predicted that AD domains Finally, since it is based on a physical binding assay, a wide variety can be recruited by protein–protein interactions (28). This aspect of protein–protein interactions can be detected and characterized was further exemplified by using Gal80p, a repressor protein which following one single commonly used protocol. In this section, we interacts with Gal4p and lacks a naturally occurring AD: an artificial attempt to give some historical perspective on these three aspects. Gal80p–AD fusion could restore, in trans, the ability of a Gal4p Geneticists have long exploited the advantages of growth mutant lacking a functional AD to activate transcription (29). selection to identify relatively rare events. In these experiments, Although the observations described above led to a clearer idea a few growing colonies detected among very high numbers of of ‘how transcription factors work’, the resulting potential non-growing cells are indicative of an interesting and rather rare practical applications remained unclear for a while, until a event, such as a mutation in a particular gene or the loss of a seminal concept was developed that extended the possibilities of plasmid. In all cases, the genotype and/or the growth medium are the system beyond the study of transcription factors. It was manipulated to obtain a set of conditions under which the starting demonstrated that protein–protein interactions unrelated to ‘wild-type’ strain is not able to grow. For example, yeast transcription factors can reconstitute a functional transcription auxotrophic mutants affected in their ability to synthesize factor by bringing DB and AD into close physical proximity (30). pyrimidine can be selected on the basis of their resistance to In this setting, the ‘architectural blueprint’ for the ‘reconstitution’ ureidosuccinic acid since wild-type strains are sensitive to this of a functional transcription factor can be summarized as follows: drug (18). Since mutations occurring at frequencies as low as –10 DB–X/AD–Y, where X and Y could be essentially any proteins 10 can be recovered (19), similar yeast selections applied to from any organism. This rather unexpected twist opened the the identification of mammalian cDNAs can be very powerful –6 –7 doors to a plethora of applications. For example, soon thereafter, (the frequency of relatively rare cDNAs is 10 –10 in most this system involving two hybrid proteins was shown to be useful non-normalized libraries). to identify, in complex AD–Y libraries, genes encoding proteins Almost exactly 20 years ago, it was demonstrated that functional that potentially interact with DB–X (often referred to as the ‘bait’ complementation gene cloning experiments could be performed in in these experiments) (31–33). yeast mutants with the goal of cloning yeast genes (20). Soon The reporter gene used in these pioneer experiments was the thereafter, it was also shown that human (or other species) cDNAs could also be identified by functional complementation of relevant bacterial lacZ gene. Subsequently, growth selection markers such yeast mutations (21,22). Since then many human genes have been as LEU2 and HIS3 were introduced to allow powerful growth cloned this way and, in most cases, the yeast complementation selections to be used for the detection of protein–protein cloning succeeded where biochemical attempts had not. The reasons interactions (34–36). Finally, responding to the need for functional for this success were probably related to the fact that, unlike in validation of newly identified protein–protein interactions by biochemical experiments, the physiological conditions need not be means of selecting for interaction-defective alleles, the original adjusted in each case. Although powerful, complementation cloning system was turned upside-down by introducing counterselectable is limited by the availability of the relevant yeast mutants. However, markers (37). The expression of such markers can be lethal under in some settings, functional assays were designed to clone and/or certain conditions (reverse two-hybrid system), such that yeast characterize human genes in yeast without the need for any growth selection can be applied to select mutations, proteins, Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 921 Nucleic Acids Research, 1999, Vol. 27, No. 4 921 Nucleic Acids Research, 1994, Vol. 22, No. 1 peptides or small molecules that dissociate particular interactions. A two-hybrid screen performed with the yeast Snf1p protein as The next two sections describe in greater detail the different versions bait demonstrated that the method could indeed lead to the of currently available forward and reverse two-hybrid systems. identification of interacting proteins with physiological relevance (38). Among the potential interactors identified, one was encoded by SIP1, a gene that exhibits genetic interactions with SNF1. For THE FORWARD TWO-HYBRID SYSTEM example, overexpression of SIP1 suppresses the defects caused by reduced Snf1p kinase activity. The two-hybrid system has rapidly become an attractive method Simultaneously, several groups demonstrated that protein–protein because it allows the genetic selection of genes encoding potential interactions from other species than yeast, such as mammals, interacting proteins without the need for protein purification. could also be identified in the yeast two-hybrid system (32,33). However, such a genetic screen only provides an indirect in vivo In these experiments, several dozen candidates were selected assay for proximity between two proteins. With this in mind, we from pools of 10 transformants. However, only a few exhibited describe the early versions of the two-hybrid system with bait-dependent reporter gene expression. Hence, these experiments emphasis on the strategies for the evaluation of screens and their helped define the limits of the lacZ-based colony color assay. limitations. We then detail critical parameters to control for in a Consequently, to address the question of the large numbers of two-hybrid screen and the most recent versions of the system that transformants that need to be screened to adequately survey the address these issues. We refer to Table 1 for a summary of the complexity of mammalian cDNA libraries, genetic selections pioneer and most widely used versions of the system. based on growth assays were designed to replace the lacZ-based screening strategy. This was achieved simultaneously by several Table 1. Description of the two different versions of the two-hybrid system groups which developed novel yeast strains containing selectable reporter genes. In most cases, after an initial selection of positive clones, a secondary screen was developed with a second and independent reporter gene such as lacZ to allow higher specificity (34–36). These different systems allowed the two-hybrid strategy to be used widely to analyze proteins from many model organisms and led to a large number of publications describing the successful use of the method to identify potential interacting proteins (e.g. 39). However, limitations of the system soon became evident but were not always reported in the literature. These limitations included very large numbers of clones with no biological relevance (‘false positives’) or the lack of recovery of expected interactions (‘false negatives’). Since the screen relies upon the transcriptional activation of reporter genes, any mutational events leading to an increase in the rate of transcription might be misinterpreted as the signature of a DB–X/AD–Y interaction. Thus, as for any genetic selection, it is crucial to develop the proper criteria to evaluate both the specificity and the sensitivity of the assay. Unfortunately, these parameters are not always considered and, consequently, some experiments have been some- what over-interpreted, leading to an abundance of false positives. Among the most important criteria used to sort through the (1) Gal4 amino acids 1–147. putative positive clones is the verification that the AD–Y (2) Gal4 amino acids 768–881. (3) VP16 amino acids 410–490. plasmids were selected because they indeed encode fusion (4) LexA amino acids 1–211. proteins and not fusions to irrelevant small peptides. In this (5) LexA amino acids 1–202. regard, it should be emphasized that in non-directional AD–Y (6) Ma and Ptashne (27). libraries, five out of six fusions lead to hybrid proteins involving polypeptides that do not correspond to those naturally expressed in the organism of interest. These fusions can result from the In the first two-hybrid screening experiment described, Sir4p, cloning of RNA-coding, non-coding, antisense or out-of-frame a yeast protein involved in transcriptional silencing, was the bait DNA sequences and usually give rise to short peptides fused to and the AD–Y library was generated using partially restricted yeast genomic DNA (31). The reporter gene consisted of lacZ AD (‘out-of-frame’ peptides). However , it has been suggested under the control of a Gal4p-responsive promoter, allowing a that in some cases out-of-frame fusions might encode bona fide screen for positive colonies on the basis of a colorimetric assay. fusion proteins through a frame shifting event (45). Among >200 000 transformants screened, two Sir4p-interacting Finally, it should be kept in mind that a two-hybrid screen does AD–Y fusion proteins were identified and they both corresponded not necessarily select for direct interactions. For example, to Sir4p itself. Thus, a potential homodimerization involved in peculiar DB–X fusions such as DB–lamin are capable of allowing Sir4p function was uncovered, although no novel protein had activation by the AD–Y fusions independent of a direct contact been identified. Nevertheless, this result was still very encouraging between X and Y. These ‘non-touching’ pairs often mediate their because it demonstrated that interacting fusion proteins could be effect in a promoter context-dependent manner (40). In addition, identified from a large collection of AD–Y clones. indirect interactions have been reported where an endogenous Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 922 Nucleic Acids Research, 1999, Vol. 27, No. 4 yeast protein serves as a bridge. For example the HIV-encoded library and to the size of the genome. For organisms that are protein Rev was first shown to interact with the yeast nuclear encoded by compact genomes, i.e. with small intergenic sequences protein Rip1, a member of the nucleoporin proteins (41). It has now and few introns, screening a genomic library instead of a cDNA been demonstrated that this interaction likely occurs via yeast library for two-hybrid experiments is advantageous. Indeed, such Crm1p, a protein involved in nuclear export (42). Nevertheless, the an exhaustive screen was recently published in which a highly selection of such indirect protein–protein interactions can still be complex yeast genomic library was screened with various yeast considered as an indication of a potential functional link between proteins as baits (45). these two proteins. In summary, the two central issues to be considered are the Most recently, modified two-hybrid strategies have been specificity and the sensitivity of the screen. Usually, higher designed to increase the specificity in order to limit the above specificity is desirable because a large number of false positives problems of false positives. Several strategies were chosen. First, can dangerously obscure the biologically relevant interactors. the expression level of the two hybrid proteins was reduced by However, it should be emphasized that increasing the specificity using yeast centromeric vectors, which are maintained at low usually leads to a reduction in sensitivity and thus a greater number of false negatives. copies in the cell, and truncated versions of promoters such ADH1 As a final comment, it is important to reiterate that the (32,37). Second, multiple reporter genes were designed for which magnitude of the two-hybrid read-out cannot be correlated with the corresponding promoters are unrelated, containing very the biological significance of the interaction nor with the affinity different TATA boxes and initiation sites. Since many false of the interaction between the two wild-type proteins in the relevant positives of the two-hybrid method are promoter context-dependent, organism. This is because, in addition to the affinity, many other the specificity of the assay is increased by scoring the phenotypes parameters can influence the two-hybrid read-out. These include the conferred by three different reporter genes in the same cell expression, stability, nuclear localization and three-dimensional (37,40,43). Third, mating strategies were introduced which allow structure of a fusion protein and the fact that discrete domains might screening with many different baits under reproducible conditions interact more strongly than the corresponding full-length protein. (44,45). The quality of a two-hybrid screen can also be analyzed in terms of the number of expected interacting proteins that were not THE REVERSE TWO-HYBRID SYSTEM recovered. These are often referred to as ‘false negatives’. In many cases, the reasons for the lack of detection in the two-hybrid Potential protein–protein interactions identified by the two-hybrid system are unknown, but several possibilities can be imagined. system merely represent hypotheses that need to be tested back in For example, the folding and/or the stability of a DB–X or AD–Y the relevant biological systems. Approaches conventionally used fusion protein could affect its transcriptional activation properties. include co-immunoprecipitation of endogenous proteins, co- Alternatively, particular fusion proteins might be toxic and affect immunolocalization or gradient sedimentation. However, the the viability of the corresponding transformed cells. These most direct approach genetically correlates the potential physical restrictions are exemplified by the fact that even though, in interaction with a biological parameter: the physical interaction principle, the reconstitution of an interaction in the context of the is dissociated and the consequences are analyzed in a functional two-hybrid system should be bidirectional, i.e. the DB–Y/AD–X assay. Logically, one would expect that if the newly detected combination would be expected to score as well as DB–X/AD–Y, interaction is critical for a function of interest, the dissociation of it is actually often not the case. In addition, the gene encoding an the interaction would impair that function. interacting protein may simply not be represented in the library Conceptually, protein–protein interactions can be dissociated due a low complexity or a bias in the representation of that clone. by the use of cis-acting mutations in one partner (referred to here For this reason, the source of DNA for the library is a key as interaction-defective alleles) or trans-acting molecules such as parameter for the assessment of the data. In most screens dissociating proteins, peptides or small molecules. For example, described above, cDNAs were derived from random- or interaction-defective alleles can be compared with their wild-type oligo(dT)-primed RNAs. It must be kept in mind that, in contrast to counterparts for their ability to functionally complement a genomic libraries, the relative representation of each cDNA closely knockout in the corresponding gene or for their ability to function reflects the endogenous expression level of the corresponding gene. in an expression assay in the relevant cells. Alternatively, the Thus, interesting interacting proteins might be under-represented, corresponding proteins can be expressed and purified and if their RNA is expressed at relatively low levels. One solution subsequently compared with their wild-type counterpart in an in here is the use of normalized AD–Y libraries (M.Brasch and vitro biochemical assay. However, since they usually score as M.Vidal, in preparation). The process of normalizing cDNA libraries recessive mutations, the use of interaction-defective alleles is consists of reducing the representation of highly expressed cDNAs compromised in cases where the wild-type protein cannot be (46). In addition, the choice of a random-primed versus an removed from the assay. This would be the case in model oligo(dT)-primed cDNA library considerably modifies the nature organisms when no knockout is available for a particular gene of of the screen. Discrete protein domains are more likely to be interest or in biochemical assays when the corresponding screened with random-primed libraries, while clones encoding wild-type protein cannot be immunodepleted from the tested nearly full-length proteins are enriched when oligo(dT)-primed fractions. In these cases, dominant trans-acting peptides or small libraries are used. However, in high complexity libraries, the molecules that specifically affect the ability of a particular protein occasional internal priming events of oligo(dT) primers to poly(A) pair to interact could be used. sequences can still lead to optimal coverage of N-terminal Until recently, this genetic strategy to validate potential domain-encoding sequences (M.Brasch, personal communication). interactions had not been used widely due to the technical In contrast, the complexity of genomic libraries is directly difficulties of identifying informative interaction-defective alleles or correlated to the number of independent clones that compose the specific dissociating molecules. The main challenge for interaction- Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 923 Nucleic Acids Research, 1999, Vol. 27, No. 4 923 Nucleic Acids Research, 1994, Vol. 22, No. 1 Figure 1. Forward and reverse n-hybrid systems. Four classes of macromolecular interactions can be detected using the original concept of reconstituting functional transcription factors. They include the following interactions: DNA–protein (one-hybrid system) (lines 1 and 2), protein–protein (two-hybrid) (lines 3 and 4), RNA–protein (RNA-based three-hybrid) and small molecule–protein (drug-based three-hybrid) (line 5). Potential interactions can be identified on the basis of the transcriptional activation of a ‘gene required for growth’ which confers a selective advantage. See the representation of two growing yeast patches (lines 1, 3 and 5) in the forward ‘n’-hybrid configuration. It has also been demonstrated for the one- and two-hybrid systems that reagents that dissociate interactions can be identified by reversing the system. In this ‘reverse’ n-hybrid configuration, the reconstituted transcription factor activates the expression of a ‘toxic gene’ and thus it is the dissociation of the interaction that provides a selective advantage under those conditions. See the representation of two growing yeast patches (lines 2 and 4). X, DNA, protein, RNA or small molecule; Y, potential interacting protein; DB, DNA-binding domain; AD, activation domain (details in text). defective alleles is the creation of subtle mutations that disrupt the proteins; (iii) review the different aspects of the reverse two-hybrid interaction without grossly affecting the protein, i.e. in the context that have already been validated. of a full-length, stable and correctly folded protein. Another Different yeast toxic markers that can be used for negative problem is that for many novel interactors, no structural selection include URA3 and CYH2. Traditionally, the URA3 information is available and interaction-defective alleles cannot marker has been used most extensively because it allows both be rationally designed. To circumvent these two problems, large negative and positive growth selection, on medium containing libraries of mutant alleles have to be generated randomly for each 5-fluoroorotic acid (5-FOA) or on medium lacking uracil, interaction partner with the goal of finding a few that exhibit the respectively. The two-hybrid inducible SPAL::URA3 reporter desired properties. Similarly, one might expect that complex gene was designed by combining in the same promoter a strong libraries need to be screened to find a few specific dissociating cis-acting repressing sequence to maintain very low basal levels peptides or small molecules. of expression and Gal4p-binding sites to allow Gal4p inducibility The problem of screening very large libraries, of alleles or of (47). A related reporter gene, based on an identical promoter molecules, should be overcome by using a genetic selection in conformation, is also available with LexA-binding sites (48). In which it is the dissociation of the interaction that provides a cells containing these reporter genes, wild-type interactions selective advantage. This situation is provided in the context of the confer a 5-FOA-sensitive phenotype. An alternative system uses reverse two-hybrid system. In this ‘upside-down’ version of the the CYH2 marker downstream of the GAL1 promoter and in this two-hybrid system, the wild-type DB–X/AD–Y interaction can be case, wild-type interactions confer a cycloheximide-sensitive toxic or lethal for the yeast cells because a toxic marker is used as phenotype (49). Finally, in another strategy, the two-hybrid a reporter gene (negative selection) (37; Fig. 1). In this setting, interaction activates expression of the Tet repressor, which in turn DB–X/AD–Y dissociation confers a selective growth advantage represses expression of the positive selectable marker HIS3 (50). which can conveniently identify both interaction-defective alleles Under these conditions, wild-type interactions confer a histidine and dissociating peptides or small molecules. auxotrophic phenotype. Although many presumptions can be Below we will: (i) describe the different toxic reporter genes made on the respective advantages of these different selections, that can be used to provide negative selection; (ii) compare the more experimental data will be needed for a thorough comparison. multiple genetic strategies used to identify interaction-defective Several genetic strategies have been implemented to select alleles in the context of stable full-length and correctly folded interaction-defective alleles in the context of a stable full-length Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 924 Nucleic Acids Research, 1999, Vol. 27, No. 4 and properly folded protein. The first set of strategies is based on AD–TGFβRIC using relatively simple and automatable plate the selection for conditional mutations that retain some level of assays (37,48,60). Although the interaction between FKBP12 activity. For example, mutations that affect the interaction only and the TGF-β receptor IC has not been completely validated in weakly can be selected out from nonsense alleles (51). Additionally, vivo, it had been previously shown that FK506 affects its thermosensitive mutations can also be selected that affect the two-hybrid read-out (61). Last but not least, it was demonstrated interaction at the restrictive temperature while retaining wild-type in one case that small molecules can indeed be selected de novo activity at the permissive temperature (37,52). Finally, interaction- from complex libraries using the reverse two-hybrid system (62). defective alleleles can be found that affect the interaction with one In this experiment, a novel compound, WAY141520, was particular partner but not with another (51,53,54). In this identified in a collection of ~ 150 000 molecules as a dissociator particular case, simultaneous selections can even be performed in of the interaction between two β3 and α1B subunits of N-type the same yeast cells if the first partner is fused to a DB that calcium channels. This compound was shown to inhibit N-type activates a reverse two-hybrid reporter gene while the second calcium channel activity in a specific manner (62). partner is fused to another DB that activates a forward two-hybrid reporter gene (55). Although conditional alleles can be very ALTERNATIVE n-HYBRID SYSTEMS informative in some cases, it might be more convenient in other cases to use an exogenously added C-terminal marker protein to Although protein–protein interactions form the basis of many eliminate nonsense alleles. In this second set of strategies, the biological processes, other macromolecular interactions such as wild-type protein to be mutagenized is fused at its C-terminal end DNA–protein and RNA–protein interactions are also critical. The to an easily scorable protein. So far, both β-galactosidase (50) and original two-hybrid DB–X/AD–Y configuration was modified to green fluorescent protein (Y.Jacob, H.Endoh and M.Vidal, in accommodate the detection of such interactions. preparation) have been used. Here, interaction-defective alleles The cloning of genes encoding DNA- or RNA-binding proteins are selected on the basis of negative selection combined with a has traditionally relied upon biochemical approaches. For β-galactosidase or fluorescence assay directly on the yeast example, the purification of a DNA-binding activity using DNA colonies. In addition, fusions with a small 18 amino acid domain affinity chromatography can be followed by the cloning of a of the transcription factor E2F1, which is necessary and sufficient to cDNA using a probe deduced from the sequence of the purified interact with pRB, have also been used. Here, interaction-defective protein (63). Another approach involves the screening of an alleles are selected that retain the binding activity to a referential expression library with a radiolabeled DNA or RNA probe DB–pRB fusion protein (H.Endoh and M.Vidal, in preparation). corresponding to the sequence of interest (64). Phage display can In the past 2 years, several aspects of the reverse two-hybrid also be used to identify RNA-binding proteins in vitro (65). system for interaction-defective alleles have been validated. First, Although these methods have been successful in isolated cases, it was shown that negative selection can be titrated to accommodate they are limited by the constraints of in vitro conditions. In a large range of protein–protein interaction affinities (47). addition, they are limited by the fact that each interaction tested Second, well-characterized interaction-defective alleles that are requires a careful adjustment of the conditions. As argued for the found in human diseases were shown to rescue the negative identification and characterization of protein–protein interactions, growth phenotype of the reverse two-hybrid (47,50). Third, de the use of yeast selection can bypass these limitations. novo selected interaction-defective alleles were described The one-hybrid system used to study DNA-binding proteins is (47,50,53) and shown to be deficient for interaction in an in vitro an extension, by simplification, of the two-hybrid concept (66; binding assay (47,50). Finally, some of the interaction-defective Fig. 1). In this configuration, the DB–X hybrid is eliminated and alleles selected using the reverse two-hybrid system were shown the the DNA Gal4p- or LexA-binding sites are replaced by a specific to be functionally defective in several different biologically DNA sequence identified as an important binding site in the relevant relevant assays. Human papillomavirus 16 E1 interaction-defective biological system. The DNA-binding protein corresponding to this mutant proteins were shown to be defective in a replication assay site can be identified and/or characterized as a fusion to AD, in transfected cells (56). Similarly, yeast ste5 interaction-defective which extends the use of the system to proteins that are not alleles were shown to fail to complement the phenotypes of a ste5 necessarily transcriptional activators, such as proteins involved in null mutation (53). transcriptional repression or DNA replication. Several versions of Finally, it should be added that such functional assays using the one-hybrid system have been published (referenced in 66). interaction-defective alleles to validate potential interactions They usually differ from each other by the choice of the selectable need to be substantiated by the proper controls. For example, in marker, such as HIS3 or lacZ, and whether the marker is on a a few published cases, compensatory mutations were selected in plasmid or integrated within the genome. The different steps of the interacting partner by using a forward two-hybrid strategy and the protocols are as follows. The DNA-binding site is identified, subsequently tested in a functional assay. This strategy implies located as precisely as possible and cloned into the promoter that for relevant interactions, the restoration of the interaction can driving the reporter gene(s). This construct is introduced into restore the function, at least partially (17,55,57,58). yeast cells and tested for basal transcriptional activity since it is Several aspects of the selection for trans-acting dissociating possible that an endogenously expressed yeast protein binds to the molecules have also been validated. On the one hand, a short site or a neighboring sequence and activates the reporter gene. adenovirus E1A peptide was shown to rescue the 5-FOA-sensitive The subsequent steps are identical to two-hybrid selections and phenotype in yeast cells expressing the DB–pRB/AD–E2F1 similar AD fusion libraries can be used (Fig. 1). An important interaction (47). It is well known that, upon adenovirus infection, control for the potential interactors selected from a one-hybrid E1A mediates the dissociation of pRB from E2F1 (59). On the experiment consists of verifying that the AD–Y protein loses its other hand, the small molecule FK506 was shown to rescue the ability to activate the reporter gene when a mutant DNA sequence, 5-FOA-sensitive phenotype conferred by DB–FKBP12/ known to be affected for its binding in the biological assay, replaces Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 925 Nucleic Acids Research, 1999, Vol. 27, No. 4 925 Nucleic Acids Research, 1994, Vol. 22, No. 1 the wild-type binding site used in the selection. So far the one-hybrid reactions. However, as for the RNA-based three-hybrid system, a system has been used successfully to identify a few proteins, the yeast genetic strategy has been developed that relies upon the yeast origin of replication complex Orc6p protein (67), the presence of a third hybrid molecule bridging DB–X and AD–Y mammalian olfactory neuronal transcription factor Olf-1 (68), a (Fig. 1). In this case, the third hybrid organic molecule is synthesized mammalian silencer that restricts sodium channel expression in in vitro and consists of a well-known ligand (referred to here as neurons (69) and a metal response binding factor (70). ligand) that binds to a known small molecule-binding protein The RNA-based three-hybrid system used to study RNA-binding (referred to here as LBP) and is covalently attached to a small proteins is a more complex extension of the two-hybrid concept molecule of interest (referred to here as X by analogy with the (Fig. 1). In this configuration, both DB–X and AD–Y hybrid two-hybrid configuration) (75). Such hybrid small molecules were proteins are expressed and a third hybrid, an RNA molecule, shown to be useful for the inducible homodimerization of membrane bridges them (71). The RNA hybrid molecule consists of a fusion receptors or activity of transcription factors (76). In the three-hybrid between a known RNA sequence (referred to here as RNA) which setting, LBP is fused to DB and reconstitution of the transcription binds to a known RNA-binding protein (referred to here as RBP) factor relies upon two interactions indicated as follows: DB–LBP/ ligand–X/AD–Y. Novel proteins Y that bind the small molecule X and a novel RNA sequence to be characterized (referred to here can be identified and/or characterized as fusions to AD. In one as X by analogy with the two-hybrid configuration). In this setting version of the small molecule-based three-hybrid system (75), RBP is fused to DB and reconstitution of the transcription factor DB–LBP corresponds to the LexA DB fused to the rat glucocorti- relies upon two interactions, as indicated in the following ‘blueprint’: coid receptor (GR) and the hybrid molecule is dexamethasone (Dex) DB–RBP/RNA–X/AD–Y. Novel RNA-binding proteins Y can be covalently attached to a small molecule X. In other words, DB–GR, identified and potentially characterized as fusions to AD. In one Dex and AD are fixed while X, a small molecule, and Y, a small version of the RNA-based three-hybrid (72), DB–RBP consists molecule-binding protein, vary among experiments. It should be of the LexA DB fused to the coat protein of bacteriophage MS2 noted that Dex has previously been shown to penetrate yeast cells and the hybrid RNA molecule consists of two MS2-binding sites and activate GR in the nucleus (77). The system was validated by (MS2RNA) linked to the RNA sequence X (72). In other words, reconstituting the well-characterized interaction between FK506 and DB–MS2, MS2RNA and AD are fixed while X, an RNA sequence, one of its binding proteins using micromolar concentrations of the and Y, an RNA-binding protein, vary among experiments. The Dex–FK506 hybrid molecule. In addition, small molecule-based system was validated by reconstituting the well-characterized three-hybrid selection was successful in identifying FKBP-encoding interaction between the RNA iron response element (IRE) and a cDNAs from a Jurkat cell cDNA library. In these cloning protein to which it binds tightly (IRP1). Subsequently, RNA–protein –7 experiments, the dependence upon the presence of the Dex–X interactions exhibiting a range of affinities between 10 and –11 molecule was tested by incubating the potential positives on medium 10 M were shown to be detected in the assay (71). Most containing an excess of X and the relevant positives corresponded importantly, RNA-based three-hybrid selections have proven to to those for which the read-out was eliminated. be successful in identifying novel and biologically relevant As for protein–protein interactions, there is a need for interacting proteins. In these experiments, the RNA sequence of interaction-defective alleles corresponding to DNA–, RNA– and interest is identified in a biological system and the corresponding small molecule–protein interactions as well. Conceptually, the DNA sequence is cloned into a vector from which the proper reasoning developed above on the negative selections possible hybrid RNA molecule can be expressed. This construct is with the reverse two-hybrid system should be applicable. We introduced into yeast cells, along with DB–MS2, and tested for refer to these strategies as reverse n-hybrid systems. Just as it is basal transcriptional activity since it is possible that the RNA–X possible to impose positive selections for AD–Y cDNAs that molecule directly or indirectly activates reporter gene transcription mediate different n-hybrid interactions, it should also be possible in the absence of any exogeneous AD–Y hybrid protein. The to select for interaction-defective alleles of AD–Y using the subsequent steps are identical to the two-hybrid selections and, negative selections described above. This concept has been again, identical AD cDNA libraries can be used. An important demonstrated in a reverse one-hybrid system configuration (47; control for the potential interactor selected from RNA-based Fig. 1). After reconstitution of p53 binding to its DNA recognition three-hybrid experiments is verification that MS2RNA–X is sequence incorporated within the promoter expressing the URA3 required for the read-out. Thus proteins that bind DB–RBP reporter gene, interaction-defective mutant alleles were selected directly or that are bridged with DB–RBP through an endogenous from a randomly generated library of mutant alleles. Remarkably, yeast RNA molecule or protein can be eliminated. In addition, many alleles recovered from this selection corresponded to when available, a functionally defective mutant version of X should mutations found in patients (78). It is also conceivable that be tested to verify specificity of the potential RNA-binding protein. compensatory mutations in the corresponding DNA or RNA So far, the RNA-based three-hybrid system has been used binding sites, or derivatives of the small molecules, could be successfully to select a Caenorhabditis elegans fem-3 mRNA recovered that restore the interaction. binding factor which was shown genetically to be required for sexual fate in the hermaphrodite germline (73). In addition, a histone mRNA-binding protein, called SLBP, was identified using the INTRINSIC LIMITATIONS AND VARIATIONS OF THE system and shown to be required for processing of this mRNA (74). NUCLEAR TWO-HYBRID Small organic molecules can be useful in elucidating biological mechanisms if they specifically affect functions of interest. Thus, it The conventional two-hybrid strategy suffers intrinsic limitations is often important to identify the protein targets of interesting small because of its reliance upon transcriptional activation. A major molecules. Traditionally, the identification of small molecule- limitation is that neither the bait protein nor the potential interacting binding proteins has relied upon biochemical methods including in protein should be able to activate transcription on their own. This can vitro affinity chromatography and radiolabeled ligand binding be a problem with transcriptional activators which naturally contain Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 926 Nucleic Acids Research, 1999, Vol. 27, No. 4 Figure 2. Variations on the two-hybrid theme. The conventional description of the two-hybrid system (line 1) has been extended to different configurations of reconstituted transcription factors (lines 2–4). Furthermore, it was demonstrated that molecules other than transcription factors can be reconstituted using two hybrid proteins as long as two interacting moieties can bring them into close proximity (lines 5 and 6). X, protein of interest; Y, potential interacting protein; DB, DNA-binding domain; AD, activation domain; PolIII, RNA polymerase III; τ138, activator of PolIII; Z, third molecule that modifies or stabilizes X–Y interaction; Nub , mutant form of N-terminal domain of ubiquitin; Cub, C-terminal domain of ubiquitin; Rep, reporter protein such as DHFR; MA, membrane-anchoring domain of SRC; GEF, GTPase exchange factor. The names and the references for the different systems are described in the text. domains that activate the reporter genes when fused to the DB. In was fused to the activation domain (AD–X) and a DB–Y library addition, many proteins other than transcription factors are also was screened. In the initial selection, many positive clones were found to activate transcription when artificially fused to DB (27). selected (5000 for 2 × 10 transformants), most of them Both classes of proteins are referred to here as ‘self-activators’. representing DB–Y that can self-activate. These self-activators When dealing with self-activators, several approaches can be could be eliminated by selecting for loss of the AD–X bait considered. First, the expression level of the DB–X fusion can be plasmid using CYH2 (above; 35) and subsequently testing the decreased by using centromeric vectors and/or weaker promoters ability of the DB–Y hybrid protein to activate HIS3 expression. (above). Second, when using HIS3 as a reporter gene, the Among the initial 5000 positive colonies, 46 expressed AD–X 3-aminotriazole (3AT) concentration can be increased to elevate bait-dependent interactors and among these a few have been the growth threshold of the host strain. The drug 3AT acts as a shown to be functionally relevant (80). competitive inhibitor of the HIS3-encoded enzyme (35). Under In some extreme cases, including the study of protein these conditions, it is expected that, even though DB–X activates complexes involved in RNA polymerase II-mediated transcription, HIS3 transcription to some extent, the DB–X/AD–Y interaction one might imagine that, in addition to a self-activating bait, the leads to more HIS3 expression to overcome the growth threshold expected interacting protein(s) could also act as a self-activator. imposed by 3AT (79). However, these two strategies are not In this case, the swapped two-hybrid configuration described always successful, especially for strong self-activators. above would eliminate the desired interacting protein. To bypass Finally, a swapped two-hybrid system can be used to deal with these problems, an RNA polymerase III (PolIII)-based two-hybrid self-activator baits (80; Fig. 2). In this approach, the bait protein system was designed (81; Fig. 2). Here the endogenous SNR6 Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 927 Nucleic Acids Research, 1999, Vol. 27, No. 4 927 Nucleic Acids Research, 1994, Vol. 22, No. 1 gene, which encodes the essential U6 snRNA, is replaced by a N-terminal domain (Nub ) was found to be unable to mediate this mutated allele that contains Gal4p-binding sites in its 3′ functional interaction and, under these conditions, the two regulatory region to constitute the reporter gene of the assay. The domains fail to reconstitute active ubiquitin. The ‘ubiquitin-based bait protein X is fused to the Gal4p DB (DB–X) and the potential split protein sensor’ (USPS) takes advantage of these observations interacting proteins Y are fused to the PolIII activating protein to detect exogenous protein–protein interactions (86). In this τ138 (τ138–Y). In this particular configuration, AD–X/τ138–Y configuration, the protein of interest is fused to Cub (Cub–X) and m m interaction reconstitutes a functional PolIII-activating transcription the potential interacting proteins are fused to Nub (Nub –Y). factor. This two-hybrid version was shown to be capable of Thus the Cub–X/Nub –Y interaction reconstitutes a functional detecting a well-characterized interaction (81). However, it ubiquitin which can target specific reporter proteins. In the remains to be demonstrated that a screening procedure will be original description of the system, the reporter protein used was possible in such a PolIII-driven transcriptional system. human dihydrofolate reductase (DHFR) expressed as a fusion to Another potential limitation of the two-hybrid involves the Cub–X and cleavage of that fusion was used as an indication for failure of yeast cells to undergo various post-translational X/Y interaction. This two-hybrid version was proven capable of modifications required for particular interactions in higher reconstituting the well-characterized homodimerization of the organisms, in particular, glycoproteins or lipoproteins for which yeast Gcn4p leucine zipper domain. However, this strategy has the non-peptidyl part of the molecule is suspected to be involved not yet been adapted for the selection of novel interacting proteins. in the interaction. To our knowledge, it has not been reported that It is possible to imagine a reporter protein that would be functionally such modifications naturally found in cytoplasmic proteins can inactive when fused to Cub–X. Under these conditions, Cub–X/ indeed be reproduced in yeast to generate active two-hybrid Nub –Y interaction would provide a selective advantage by fusions. However, other modifications have been reported. In one releasing the reporter protein from its covalent link to Cub–X (86). case, it has been shown that a tyrosine phosphorylation-dependent A recent modification to bypass the reconstitution of a interaction could be detected when the corresponding kinase was transcription factor takes advantage of a cell proliferation co-expressed in yeast cells (82; Fig. 2). Several plasmids have signaling pathway (87; Fig. 2). The ‘Sos recruitment system’ been designed to allow conditional expression of such a ‘third’ (SRS) uses as a signal recruitment to the yeast plasma membrane trans-acting partner. In addition to mediating post-translational of the human guanyl nucleotide exchange factor (GEF) hSos, modifications of one of the two interacting proteins, this third which can functionally substitute for the essential yeast Ras-GEF protein can directly contribute to the formation of a trimeric Cdc25p. In a yeast strain containing a thermosensitive allele of complex with greater stability (83). These different variations that CDC25, growth at non-permissive temperature occurs only if involve third partners as native proteins, in the absence of any hSos can be recruited efficiently to the membrane and this fused domains, are referred to as ‘tribrid systems’. requires a myristylation signal, such as the one contained in the Finally, it has been shown that the two-hybrid system can be Src protein kinase. Thus, in the SRS system the bait protein of used as a tool to select for peptides that can interact with a protein interest is fused to the GEF domain of hSos (GEF–X) and the of interest (84). This strategy has been applied with a combinatorial interacting proteins are fused to the Src membrane-anchoring library of constrained 20mer peptides displayed by the active site domain (MA) (MA–Y). Under these conditions, GEF–X/MA–Y loop of bacterial thioredoxin (85). If this protein–peptide interaction allows recruitment of hSos to the membrane and two-hybrid system turns out to be applicable for the general rescue of the yeast cdc25 mutation. The SRS system has already selection of ligands that affect the activity of proteins, it might been used successfully in a screen for c-Jun interacting proteins open an avenue toward the design of powerful research tools for (87). In that experiment, expression of the MA–Y hybrids was in vivo functional studies in organisms for which genetic controlled by a galactose-inducible promoter, thus allowing the strategies are not readily available. convenient elimination of chromosome-encoded cdc25-2 suppressors. However, one of the limitations of the system is that Cdc25p function could be rescued by overexpression of mammalian TWO-HYBRID SYSTEMS BASED ON DIFFERENT Ras proteins and other MA–Y fusion proteins, independent of any CELLULAR PROCESSES interaction with c-Jun. This class of proteins could be a serious limitation of this otherwise efficient alternative two-hybrid One of the main limitations of the two-hybrid system is that both strategy. A possible solution to the problem is based on the the DB–X and the AD–Y fusions need to be transported to and observation that overexpression of GTPase activating protein properly folded in the nucleus. Thus, most current two-hybrid (GAP) can suppress the bypass of Cdc25 function by Ras (88). vectors encode DB and AD with a nuclear localization signal to However, it remains to be shown whether this improvement target the fusion proteins to the nucleus. However, proper significantly reduces the number of false positives obtained in localization can represent a major difficulty, especially when screening procedures. dealing with membrane-anchored proteins. Variations of the system that have been designed for those proteins that are not active in the yeast nucleus are described in this section. TWO-HYBRID SYSTEMS IN PROTEOMICS One such system relies on the properties of the ubiquitin protein (86; Fig. 2). It is based on the observation that covalent addition The recent release of the complete genomic sequence of several of ubiquitin polypeptides can recruit particular proteases for the organisms has introduced the need for large-scale projects that specific degradation of target proteins. When ubiquitin is address the function of the predicted proteins. The generation of experimentally cleaved into C-terminal (Cub) and N-terminal protein interaction maps corresponding to complete genomes, or (Nub) domains, neither domain can mediate ubiquitin function, sets of expressed genes in particular tissues, would represent a while simultaneous expression of both domains reconstitutes reasonable strategy to add valuable predictive functional information active ubiquitin in trans. However, a particular allele of the to crude sequence alignments (89,90). As a genetic system, the Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 928 Nucleic Acids Research, 1999, Vol. 27, No. 4 two-hybrid assay has been proposed as the optimal method for this enormously challenging task. Indeed, recent publications suggest that the method should allow such protein interaction maps to be generated in a reasonable period of time. A first attempt to analyze protein–protein interactions in a systematic manner was proposed for Drosophila cell cycle regulators (91). It had been previously demonstrated that the plasmids encoding the DB–X and AD–Y fusions can be introduced into the same yeast cell by a mating procedure (44). In this method, one plasmid is introduced into haploid cells of one mating type and the other plasmid is introduced into cells of the opposite mating type. Since yeast mating is very efficient, numerous combinations of DB–X/AD–Y can be assayed simultaneously in diploid cells. Using a similar mating technique, a two-dimensional matrix was generated for cyclin-dependent kinases in which potentially interacting proteins were detected in diploid cells. This approach is nevertheless restricted by several considerations. First, only known or predicted proteins could be tested and this is an obvious key problem in genome-wide projects. Second, the matrix Figure 3. Forward and reverse two-hybrid systems in proteomics. Proteomes approach necessitates the use of a single set of growth conditions, predicted by large-scale genome sequence analysis and cDNA sequencing lead which precludes the possibility of using an adapted selective to the choice of proteins of interest (DB–X baits). Forward two-hybrid pressure for every specific interaction. For example, the 3AT selections with DB–X baits lead to potential interactors. These interactions can concentration cannot be adjusted to account for weak self-activation be validated using interaction-defective alleles and/or specific dissociators in of certain DB–X baits. Third, the use of full-length proteins for both functional assays. The combination of such functional assays and two-hybrid screens could lead to comprehensive protein interaction maps. Adapted from the DB–X bait fusions and the AD–Y interacting proteins might Walhout et al. (97). prevent the identification of several interactions due to various intrinsic problems such as toxicity, folding and degradation. used here to generate the large number of combinations required Another systematic approach was carried out for a small viral to test every single pair of predicted proteins (95). genome (92). In this experiment, a large number of combinations It is conceivable that in the near future, similar strategies will between the proteins encoded by the Escherichia coli bacteriophage be applied to protein interaction maps for higher eukaryotes T7 were tested and 25 interactions were identified. The approach (96,97). Many interactions will be reported and it will be critical consisted of mating yeast cells containing libraries of randomly to define parameters that allow a careful comparison of results generated fragments of T7 phage DNA cloned into the DB and within and between organisms. Two-hybrid screens will have to AD fusion-encoding plasmids. Non-self-activating bait colonies 5 7 be performed under comparable standardized procedures. The were mated, 10 at a time, with 10 –10 yeast colonies expressing combination of advanced forward and reverse n-hybrid strategies, in the AD–Y library, leading to hundreds of positive combinations their present forms as well as using future improvements, might defining 19 interactions. Subsequently, more specific screens continue to deliver functional information on a more rational with defined baits led to the identification of six additional basis (Fig. 3). interactions. This analysis of a 55 protein-encoding genome led to the discovery of many novel interactions between different ACKNOWLEDGEMENTS proteins as well as between different domains of the same protein, thus defining intramolecular interaction domains. The potential We thank our colleagues and friends Mike Brasch, Rong Jiang, interactions identified have yet to be correlated with biological Linda Ko, Stan Fields and Marian Walhout for careful reading of significance. However, this study demonstrated that it is possible the manuscript. We apologize to the authors of the papers not cited to systematically and efficiently identify many of the possible because of space limitations. The work in our respective connections between proteins encoded by a given genome. laboratories is funded by NHGRI grant 1 R01 HG01715-01 A1 Recently, an efficient mating strategy coupled with the to M.V. and by EU grant Biotech 95-0009 to P.L. generation of a large genomic library was used to perform exhaustive screens of the yeast Saccharomyces cerevisiae genome REFERENCES with proteins involved in RNA splicing (45). By using this strategy it was possible to classify the potential interacting 1 Lane,D.P. and Crawford,L.V. (1979) Nature, 278, 261–263. proteins in sets of various heuristic values. The most likely 2 Linzer,D.I. and Levine,A.J. (1979) Cell, 17, 43–52. 3 Sawadogo,M. and Sentenac,A. (1990) Annu. Rev. Biochem., 59, 711–754. candidates for interaction were subsequently used as baits and, 4 Koleske,A.J. and Young,R.A. (1994) Nature, 368, 466–469. within a reasonable period of time, potential interaction networks 5 Kramer,A. (1996) Annu. Rev. Biochem., 65, 367–409. were built ‘around’ proteins of well-known functions. Subsequently, 6 Guthrie,C. (1995) Harvey Lect., 90, 59–80. dedicated functional assays were applied to confirm the suspected 7 Choi,K.Y., Satterberg,B., Lyons,D.M. and Elion,E.A. (1994) Cell, 78, function of the various interacting proteins (93,94). Finally, a 499–512. 8 Printen,J.A. and Sprague,G.F. (1994) Genetics, 138, 609–619. directed approach has been suggested to generate a yeast protein 9 Tsunoda,S., Sierralta,J., Sun,Y., Bodner,R., Suzuki,E., Becker,A., interaction map by systematically cloning the complete set of Socolich,M. and Zuker,C.S. (1997) Nature, 388, 243–249. predicted ORFs into both the DB and AD vectors of the 10 Dowdy,S.F., Hinds,P.W., Louie,K., Reed,S.I., Arnold,A. and two-hybrid system (95). Similar mating strategies could also be Weinberg,R.A. (1993) Cell, 73, 499–511. Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018 929 Nucleic Acids Research, 1999, Vol. 27, No. 4 929 Nucleic Acids Research, 1994, Vol. 22, No. 1 11 Ewen,M.E., Sluss,H.K., Sherr,C.J., Matsushime,H., Kato,J. and 58 White,M.A., Nicolette,C., Minden,A., Polverino,A., Van Aelst,L., Karin,M. and Wigler,M. (1995) Cell, 80, 533–541. Livingston,D.M. (1993) Cell, 73, 487–497. 59 Fattaey,A.R., Harlow,E. and Helin,K. (1993) Mol. Cell. Biol., 13, 7267–7277. 12 Zhu,L., Harlow,E. and Dynlacht,B.D. (1995) Genes Dev., 9, 1740–1752. 60 Vidal,M. and Endoh,H. (1998) Trends Biotechnol., submitted for publication. 13 Krek,W., Xu,G. and Livingston,D.M. (1995) Cell, 83, 1149–1158. 61 Wang,T., Donahoe,P.K. and Zervos,A.S. (1994) Science, 265, 674–676. 14 Whyte,P., Buchkovich,K.J., Horowitz,J.M., Friend,S.H., Raybuck,M., 62 Young,K., Lin,S., Sun,L., Lee,E., Modi,M., Hellings,S., Husbands,M., Weinberg,R.A. and Harlow,E. (1988) Nature, 334, 124–129. Ozenberger,B. and Franco,R. (1998) Nature Biotechnol., 16, 946–950. 15 Whyte,P., Williamson,N.M. and Harlow,E. (1989) Cell, 56, 67–75. 63 Kadonaga,J.T. and Tjian,R. (1986) Proc. Natl Acad. Sci. USA, 83, 16 Legrain,P. and Chapon,C. (1993) Science, 262, 108–110. 5889–5893. 17 Legrain,P., Chapon,C. and Galisson,F. (1993) Genes Dev., 7, 1390–1399. 64 Singh,H., LeBowitz,J.H., Baldwin,A.S.,Jr and Sharp,P.A. (1988) Cell, 52, 18 Bach,M.L. and Lacroute,F. (1972) Mol. Gen. Genet., 115, 126–130. 415–423. 19 Silar,P., Butler,G. and Thiele,D.J. (1991) Mol. Cell. Biol., 11, 1232–1238. 65 Laird-Offringa,I.A. and Belasco,J.G. (1995) Proc. Natl Acad. Sci. USA, 92, 20 Nasmyth,K.A. and Reed,S.I. (1980) Proc. Natl Acad. Sci. USA, 77, 11859–11863. 2119–2123. 66 Chong,J.A. and Mandel,G. (1997) In Bartel,P.L. and Fields,S. (eds), 21 Lee,M.G. and Nurse,P. (1987) Nature, 327, 31–35. The Yeast Two-Hybrid System. Oxford University Press, New York, NY, 22 Elledge,S.J. and Spottswood,M.R. (1991) EMBO J., 10, 2653–2659. pp. 289–297. 23 Klein,R.D., Gu,Q., Goddard,A. and Rosenthal,A. (1996) Proc. Natl Acad. 67 Li,J.J. and Herskowitz,I. (1993) Science, 262, 1870–1874. Sci. USA, 93, 7108–7113. 68 Wang,M.M. and Reed,R.R. (1993) Nature, 364, 121–126. 24 Ishioka,C., Frebourg,T., Yan,Y.-X., Vidal,M., Friend,S.H., Schmidt,S. and 69 Chong,J.A., Tapia-Ramirez,J., Kim,S., Toledo-Aral,J.J., Zheng,Y., Iggo,R. (1993) Nature Genet., 5, 124–129. Boutros,M.C., Altshuller,Y.M., Frohman,M.A., Kraner,S.D. and Mandel,G. 25 Keegan,L., Gill,G. and Ptashne,M. (1986) Science, 231, 699–704. (1995) Cell, 80, 949–957. 26 Brent,R. and Ptashne,M. (1985) Cell, 43, 729–736. 70 Inouye,C., Remondelli,P., Karin,M. and Elledge,S. (1994) DNA Cell Biol., 27 Ma,J. and Ptashne,M. (1987) Cell, 51, 113–119. 13, 731–742. 28 Triezenberg,S.J., Kingsbury,R.C. and McKnight,S.L. (1988) Genes Dev., 2, 71 Zhang,B., Kraemer,B., Sengupta,D., Fields,S. and Wickens,M. (1997) 718–729. In Bartel,P.L. and Fields,S. (eds), The Yeast Two-Hybrid System. 29 Ma,J. and Ptashne,M. (1988) Cell, 55, 443–446. Oxford University Press, New York, NY, pp. 298–315. 30 Fields,S. and Song,O. (1989) Nature, 340, 245–246. 72 SenGupta,D.J., Zhang,B., Kraemer,B., Pochart,P., Fields,S. and 31 Chien,C.-T., Bartel,P.L., Sternglanz,R. and Fields,S. (1991) Proc. Natl Wickens,M. (1996) Proc. Natl Acad. Sci. USA, 93, 8496–8501. Acad. Sci. USA, 88, 9578–9582. 73 Zhang,B., Gallegos,M., Puoti,A., Durkin,E., Fields,S., Kimble,J. and 32 Chevray,P. and Nathans,D. (1992) Proc. Natl Acad. Sci. USA, 89, 5789–5793. Wickens M.P. (1998) Nature, 390, 477–484. 33 Dalton,S. and Treisman,R. (1992) Cell, 68, 597–612. 74 Wang,Z.F., Whitfield,M.L., Ingledue,T.C., Dominski,Z. and Marzluff,W.F. 34 Gyuris,J., Golemis,E., Chertkov,H. and Brent,R. (1993) Cell, 75, 791–803. (1996) Genes Dev., 10, 3028–3040. 35 Durfee,T., Becherer,K., Chen,P.-L., Yeh,S.-H., Yang,Y., Kilburn,A.E., 75 Licitra,E.J. and Liu,J.O. (1996) Proc. Natl Acad. Sci. USA, 93, 12817–12821. Lee,W.-H. and Elledge,S.J. (1993) Genes Dev., 7, 555–569. 76 Crabtree,G.R. and Schreiber,S.L. (1996) Trends Biochem. Sci., 21, 418–422. 36 Vojtek,A.B., Hollenberg,S. and Cooper,J. (1993) Cell, 74, 205–214. 77 Godowski,P.J., Picard,D. and Yamamoto,K.R. (1988) Science, 241, 812–816. 37 Vidal,M. (1997) In Bartel,P.L. and Fields,S. (eds), The Yeast Two-Hybrid 78 Brachmann,R., Vidal,M. and Boeke,J. (1996) Proc. Natl Acad. Sci. USA, , System. Oxford University Press, New York, NY, pp. 109–147. 93, 4091–4095. 38 Yang,X., Hubbard,E.J.A. and Carlson,M. (1992) Science, 257, 680–682. 79 Yasugi,T., Benson,J.D., Sakai,H., Vidal,M. and Howley,P.M. (1997) 39 Golemis,E.A. and Khazak,V. (1997) Methods Mol. Biol., 63, 197–218. J. Virol., 71, 891–899. 40 Bartel,P., Chien,C.T., Sternglanz,R. and Fields,S. (1993) BioTechniques, 80 Du,W., Vidal,M., Xie,J.-E. and Dyson,N. (1996) Genes Dev., 10, 14, 920–924. 1206–1218. 41 Stutz,F., Neville,M. and Rosbash,M. (1995) Cell, 82, 495–506. 81 Marsolier,M.C., Prioleau,M.N. and Sentenac,A. (1997) J. Mol. Biol., 268, 42 Neville,M., Stutz,F., Lee,L., Davis,L.I. and Rosbash,M. (1997) Curr. Biol., 243–249. 7, 767–775. 82 Osborne,M.A., Dalton,S. and Kochan,J.P. (1995) BioTechnology, 13, 1474. 43 James,P., Halladay,J. and Craig,E.A. (1996) Genetics, 144, 1425–1436. 83 Tirode,F., Malaguti,C., Romero,F., Attar,R., Camonis,J. and Egly,J.M. 44 Bendixen,C., Gangloff,S. and Rothstein,R. (1994) Nucleic Acids Res., 22, (1997) J. Biol. Chem., 272, 22995–22999. 1778–1779. 84 Yang,M., Wu,Z. and Fields,S. (1995) Nucleic Acids Res., 23, 1152–1156. 45 Fromont-Racine,M., Rain,J.C. and Legrain,P. (1997) Nature Genet., 16, 85 Colas,P., Cohen,B., Jessen,T., Grishina,I., McCoy,J. and Brent,R. (1996) 277–282. Nature, 380, 548–550. 46 Soares,M.B., Bonaldo,M.F., Jelene,P., Su,L., Lawton,L. and Efstratiadis,A. 86 Johnsson,N. and Varshavsky,A. (1994) Proc. Natl Acad. Sci. USA, 91, (1994) Proc. Natl Acad. Sci. USA, 91, 9228–9232. 10340–10344. 47 Vidal,M., Brachmann,R., Fattaey,A., Harlow,E. and Boeke,J.D. (1996) 87 Aronheim,A., Zandi,E., Hennemann,H., Elledge,S.J. and Karin,M. (1997) Proc. Natl Acad. Sci. USA, 93, 10315–10320. Mol. Cell. Biol., 17, 3094–3102. 48 Huang,J. and Schreiber,S.L. (1997) Proc. Natl Acad. Sci. USA, 94, 88 Aronheim,A. (1997) Nucleic Acids Res., 25, 3373–3374. 13396–13401. 89 Lander,E.S. (1996) Science, 274, 536–539. 49 Leanna,C.A. and Hannink,M. (1996) Nucleic Acids Res., 24, 3341–3347. 90 Fields,S. (1997) Nature Genet., 15, 325–327. 50 Shih,H.M., Goldman,P.S., DeMaggio,A.J., Hollenberg,S.M., Goodman,R.H. 91 Finley,R.L.,Jr and Brent,R. (1994) Proc. Natl Acad. Sci. USA, 91, and Hoekstra,M.F. (1996) Proc. Natl Acad. Sci. USA, 93, 13896–13901. 12980–12984. 51 Vidal,M., Braun,P., Chen,E., Boeke,J.D. and Harlow,E. (1996) Proc. Natl 92 Bartel,P.L., Roecklein,J.A., SenGupta,D. and Fields,S. (1996) Nature Acad. Sci. USA, 93, 10321–10326. Genet., 12, 72–77. 52 Cayrol,C., Cabrolier,G. and Ducommun,B. (1997) Nucleic Acids Res., 25, 93 Siomi,M.C., Fromont,M., Rain,J.-C., Wan,L., Wang,F., Legrain,P. and 3743–3744. Dreyfuss,D. (1998) Mol. Cell. Biol., 18, 4141–4148. 53 Inouye,C., Dhillon,N., Durfee,T., Zambryski,P.C. and Thorner,J. (1997) 94 Rain,J.-C., Rafi,Z., Rhani,Z., Legrain,P. and Krämer,A. (1998) RNA, 4, Genetics, 147, 479–492. 551–565. 54 Jiang,R. and Carlson,M. (1996) Genes Dev., 10, 3105–3115. 95 Hudson,J.R.,Jr, Dawson,E.P., Rushing,K.L., Jackson,C.H., Lockshon,D., 55 White,M.A. (1996) Proc. Natl Acad. Sci. USA, 93, 10001–10003. Conover,D., Lanciault,C., Harris,J.R., Simmons,S.J., Rothstein,R. and 56 Yasugi,T., Vidal,M., Sakai,H., Howley,P.M. and Benson,D. (1997) J. Virol., Fields,S. (1997) Genome Res., 7, 1169–1173. 71, 5942–5951. 96 Regalado,A. (1998) Technol. Rev., 101, 51–53. 57 Amati,B., Brooks,M.W., Levy,N., Littlewood,T.D., Evan,G.I. and Land,H. 97 Walhout,M., Endoh,H., Thierry-Mieg,N., Wong,W. and Vidal,M. (1998) (1993) Cell, 72, 233–245. Am. J. Hum. Genet., 63, 955–961. Downloaded from https://academic.oup.com/nar/article-abstract/27/4/919/2902483 by Ed 'DeepDyve' Gillespie user on 06 February 2018
Journal
Nucleic Acids Research – Oxford University Press
Published: Feb 1, 1999