KR100445912B1 - Method for screening of protein-protein interaction by double library two-hybrid system and the recombinant yeast strains used for the system - Google Patents

Abstract

The present invention develops an existing two-hybrid system using yeast, and relates to a double library two-hybrid system (DLTS) and recombinant yeast strains used therein.

According to the present invention, it is possible to provide a method for efficiently searching for binding pairs of library-level proteins simultaneously, rather than a single protein.

Description

Method for screening of protein-protein interaction by double library two-hybrid system and the recombinant yeast strains used for the system}

The present invention relates to a Double Library Two-Hybrid System and a recombinant yeast strain that can be used to investigate the mutual binding between proteins at the library level, and more specifically to the existing two- The development of a two-hybrid system relates to a search method by a double library two-hybrid system (DLTS) and to recombinant yeast strains used therein.

The basis of the exploration of life phenomena in vivo can be seen in how macromolecules such as DNA, RNA, and protein relate to each other, and among them, protein interactions are diverse and widely involved. Therefore, to gain a comprehensive understanding of the physiological phenomena that occur within an individual, a comprehensive understanding of the interactions between all proteins is required. In addition, in order to systematically identify the interrelationships between proteins using sequences that have recently been identified in the genome business, it is necessary not only to search for interactions between proteins but also to find unknown proteins that interact with specific proteins. do. Currently, the proteome method using MALDI / TOF or the two-hybrid system using genetic phenomena of yeast are commonly used for such research, but the interaction between proteins of individuals with many genes such as humans To understand the action as a whole, more efficient and comprehensive methods are needed.

Two-hybrid systems are used by many people to search for binding between known proteins or to find unknown proteins that bind to certain proteins (Fields S, Song O. 1989. A novel genetic system to detect protein-protein interactions.Nature 340: 245-6). This system attaches a bait to the DNA-binding domain of Gal4p (Gal4 protein), a protein that activates transcription by attaching DNA, to make a hybrid protein, and to a protein that binds to the protein. When hybrid proteins made with the transcriptional activation domain of Gal4p are made and expressed in yeast at the same time, the two hybrid proteins bind to each other if they have a property of binding between bait and prey, and they attach to functional Gal4 protein, that is, DNA. Transcription also becomes an active protein and acts as a transcriptional activator that mediates the expression of the gene following the Gal4p binding site. In this state, the reporter gene β-galactosidase followed by the Gal4p-binding site in yeast. (lac Z)IHIS3You can see whether the expression of these.

Many researchers around the world use two-hybrid systems for basic research on the mechanisms of different biological processes and for discovering useful genes or searching for bioactive substances. A typical example is the whole yeast genes put into bait and prey plasmids, respectively, to investigate the interaction between proteins (Uetz et al., 2000. A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae. Nature. 403: 623-7, McCraith et al., 2000. Genome-wide analysis of vaccinia virus protein-protein interactions.Proc Natl Acad Sci US A. 97: 4879-4884, Ito et al., 2000.Toward a protein-protein interaction map of the budding yeast: A comprehensive system to examine two-hybrid interactions in all possible combinations between the yeast proteins.Proc Natl Acad Sci US A. 97: 1143-1147, Ito et al., 2001.A comprehensive two-hybrid analysis to explore the yeast protein interactome.Proc Natl Acad Sci US A. 98: 4569-4574), and these methods and results are difficult to apply in the case of much larger genomes such as human genes. Because there are several types of false-positive and false-negative methods for identifying protein binding relationships using two-hybrids, the genome is large and complex. This is because the reliability of the results is low and the interpretation is very complicated.

At this point, it is necessary to overcome the limitations of the existing two-hybrid system and establish a search system for a large number of proteins rather than a single protein in order to identify the interactions between all the proteins in an individual. In other words, it is urgently needed to understand the whole life phenomenon using the base sequence known as the genome business.

Accordingly, the present inventors have made diligent research efforts to overcome the problems of the prior arts. As a result, the present invention has improved the disadvantages of the conventional two-hybrid system using yeast. When using the search method by two-hybrid system (DLTS), it is confirmed that it is possible to provide a method for efficiently searching for binding pairs of library-level proteins simultaneously, rather than a single protein, and to complete the present invention. It became.

Therefore, the main object of the present invention so as to get caught the GAL1 promoter between the coding regions of the HF7c a part of the URA3 gene of yeast nucleotide removal recombinant yeast strain and the 476 yeast URA3 and its upper part to the expression of the URA3 Gal4p To provide a yeast strain recombined gene to depend.

Another object of the present invention to provide a search method by the double library protein binding search method using the yeast strains.

1 is a schematic diagram illustrating the principles of library screening using a yeast two-hybrid system and library screening using the Dual Library Protein Binding Search Method (DLTS).

FIG. 2 is a diagram showing the principle of homologous recombination with DNA fragments designed to produce yeast strains essential for performing DLTS.

Figure 3 is a diagram showing the expression of URA3 in yeast strains and original strains of the present invention in which the gene is recombinant.

Figure 4 is a photograph showing the determination of the optimal concentration of 5-FOA that can selectively remove only fake positive reactions made randomly using the produced yeast strains.

Figure 5 is a simplified diagram illustrating the process of transforming both types of yeast into a library, removing false positive reactions, and then finding protein binding pairs.

In order to achieve the object of the present invention, the present invention comprises the steps of a) disrupting URA3 in the yeast strain to prepare a type strain, and then introducing plasmid in which URA3 is expressed by the GAL1 promoter; b) producing an α-type strain by directly modifying URA3 in the yeast strain to be expressed by the GAL1 promoter using a genetic recombination method; c) transforming the prepared a type strain and the α type strain into a bait library and a prey library, respectively; d) converting the transformed a type strain and the α type strain into 5-FOA, respectively. selectively removing the false positive reaction while incubating in a medium comprising fluoroorotic acid; e) and mating the two types of transformed yeast that survived the selection process to detect protein-protein binding pairs. The reason for using recombinant yeast as in a) and b) is to allow URA3 to be expressed dependent on Gal4p.

In the present invention, it is preferable to further include the step of recovering the yeast surviving after the step d), which is a positive reaction caused by the growth of mating diploids or by protein binding pairs. In order not to be affected by the previous removal process using 5-FOA, yeast that survived mating must be grown and recovered in a medium without 5-FOA.

In the present invention, the bait library may include a gene encoding a DNA binding domain and a bait protein of GAL4p, and the prey library may include a gene encoding a transcriptional activation domain and a prey protein of Gal4p.

In the present invention, the concentration of 5-FOA (5-fluoroorotic acid) is preferably 0.05 to 0.07% for PBN101 and 0.13 to 0.15% for PBN201, more preferably 0.06% for PBN101, and PBN201 In the case of 0.14%.

In the present invention, the detection of the protein-protein binding pair may use a HIS3 or lac Z reporter gene.

In order to achieve the object of the present invention, the invention is a yeast strain is characterized in that the URA3 in the HF7c yeast is destroyed by homologous recombination PBN101 (accession number: KCTC 10155BP) and 476 between the coding region and the upper region of the URA3 of the yeast By inserting the GAL1 promoter into the present invention provides a yeast strain PBN201 (Accession Number: KCTC 10156BP) characterized in that the expression of URA3 was engineered to be Gal4p dependent. The reason for using such a recombinant yeast is to allow for the elimination of false positive reactions to be URA3 expression is dependent on the Gal4p, using 5-FOA.

Hereinafter, the present invention will be described in detail.

The present invention has developed a method that eliminates the weaknesses of yeast two-hybrid systems, and the basic principle of searching for protein-protein binding utilizes that of two-hybrid systems. This method is based on the fact that the yeast transcriptional activator Gal4p's DNA binding and transcriptional activation functions are separable and independent of each other (Fields and Song, 1989. A novel genetic system to detect protein-protein interactions.Nature 340: 245-6). The yeast Gal4 protein consists of 881 amino acids, which have a DNA-binding domain at the amino acid terminal region (1-147 amino acids) and two transcriptional activation domains [activation domain I (149-196), activation domain II (768-). 881) at the middle and carboxy terminus of the protein. The Gal4p DNA binding domain causes Gal4p to enter the nucleus and attach to the upstream region of the gene regulated by Gal4p. And the transcriptional activation domain is responsible for contacting factors of transcriptional machinery required to initiate transcription. Based on these properties, the yeast's two-hybrid system is a way to investigate protein-protein binding in yeast. In other words, the protein to be studied (bait) and the protein that binds to the protein (prey) are attached to the DNA-binding domain and the transcriptional activation domain of Gal4p, respectively, to form hybrid proteins, which are expressed simultaneously in yeast, and the two proteins are mutually It binds to form a complex that regenerates the function of the normal Gal4 protein, thereby acting as a transcriptional activator that mediates the expression of the gene following the Gal4p-binding site (FIG. 1A). In this state, β-galactosidase ( lac Z ), a reporter gene, is introduced below the Gal4p-binding site in yeast cells, and β-galactosidase is expressed by the function of Gal4p, which is regenerated by two hybrid proteins, and X in the medium. It is easy to read by changing the color of colonies by decomposing -gal, and by attaching the HIS3 gene as a reporter, cultivating yeast in a histidine-free medium, only the yeast expressing HIS3 survives. Only yeast can be screened. Furthermore, by attaching a cDNA library to the prey using a two-hybrid system, it can also be used as a way to find out what protein is interacting with a particular bait (Fields and Sternglanz, 1994. The two-hybrid system: an assay for protein-protein interactions.Trends Genet. 10: 286-92). In addition, the two-hybrid system can be modified to investigate the association between RNA and protein or DNA and protein (SenGupta et al., 1996. A three-hybrid system). to detect RNA-protein interactions in vivo.Proc Natl Acad Sci US A. 93: 8496-501).

The false positive reaction in the two-hybrid system mentioned above is not due to the binding of bait and prey, but for other reasons (bait bound to the binding domain is regulated by transcription). Because of the nature of the factor, or when combined with other proteins in the yeast, etc., the result appears as if it is caused by bait and prey protein binding. Conversely, a false negative reaction is a phenomenon in which a protein from another individual does not have the proper function and properties in the yeast system, resulting in no protein binding. For example, in order for a two-hybrid system to work, hybrid proteins must move into the nucleus, but if full-length proteins are used, the intracellular migration of the proteins themselves. This is because movement to the nucleus may be inhibited by a localization signal. In the case of intact proteins, the binding domain may be covered by a regulatory domain and then exposed only in certain cases to participate in protein binding. In fact, when investigating protein-protein binding using a two-hybrid system, it is common to show a binding when only a part of gene is used but a fake negative response when using a whole gene.

The double library two-hybrid system (DLTS) has been added to the existing two-hybrid system to compensate for this. [Fig. 1 (B) ]. The biggest feature is bait, which improves the use of library DNA as well as specific genes. In the conventional two-hybrid system, such a search is impossible if the library DNA is put in a bait plasmid, so many false positive reactions result in the prey library search. DLTS has improved yeast strains and reporters to eliminate false positives.

The product of the URA3 gene, which is involved in the uracil biosynthesis process required for yeast growth, produces 5-fluorouracil, which is toxic when reacted with a substrate analog called 5-FOA (5-fluoroorotic acid). Yeast is killed. To use this property, URA3 was used as a reporter. If URA3 is made to be expressed only when a fake positive reaction occurs, then by adding 5-FOA, only yeast that is false positive can be selectively killed. Yeast strains commonly used in two-hybrid systems continue to express URA3 essentially, and to be able to do this, it must be regulated by Gal4p. That is, after destroying URA3 , it should be transformed with the URA3 reporter plasmid regulated by the GAL1 promoter, or the URA3 of the strain gene be placed under a promoter that can be regulated by Gal4p. In yeast, homologous recombination occurs with high efficiency and the genotype is well known for each strain. Thus, new strains can be obtained by designing DNA fragments capable of inducing such changes and putting them into yeast.

In the present invention, genetic strains were applied to yeast strains used in the existing system to produce strains capable of performing DLTS. By destroying the URA3 gene in HF7c (MATa: a type), which was used in a two-hybrid system, this strain was always able to live in the presence of 5-FOA. In this yeast strain, a plasmid (pGUA) containing the URA3 gene under the control of the GAL1 promoter was made and transformed at the same time with a plasmid containing a gene combining a Gal4p DNA binding domain and a kind of transcriptional activator to live in the presence of 5-FOA. can not do it. The HF7c is a strain provided as a component of the product "Clontech MATCHMAKER Two-hybrid system".

On the other hand, another strain 476 ( MATα : α type) is also so that wild type URA3 in yeast can be regulated by Gal4p in the same manner as other reporter genes ( HIS3, lac Z ) without being expressed by its transcriptional regulation mechanism, by inserting the GAL1 promoter in the homologous portion of the URA3 URA3 also made with a reporter controlled by Gal4p, such as HIS3, or lac Z. Using these strains, the bait or prey libraries can be transformed into yeast, and then grown in an appropriate concentration of 5-FOA to remove only those that show false positives. . The strain 476 can be easily obtained from the Dept. of Genetics and Medicine and Howard Hughes Medical Institute, University of Washington, Box 357360, Seattle, Washington 98195-7360, USA, Seattle, USA. .

Fake positive yeasts are removed and the remaining a-type and α-type yeasts are mixed and mated, then grown in histidine-deficient media, which are diploid and interact with each other. You can choose a pair of yeasts. A binding pair that participates in protein-protein binding by investigating the expression of another reporter, lac Z in the selected yeasts, and examining the sequences by separating bait and prey plasmids from the yeasts thus obtained. Identify.

As such, DLTS has solved the weaknesses of searching using library-level bait in the existing two-hybrid system. Therefore, it may be particularly useful for comprehensively and simultaneously examining interactions among proteins in cells. This is because the genome business has already revealed all the nucleotide sequences of humans, so that only the partial nucleotide sequences of genes obtained from DLTS can provide information on the entire gene.

Hereinafter, the present invention will be described in more detail with reference to Examples. Since these examples are only for illustrating the present invention, the scope of the present invention is not to be construed as being limited by these examples.

<1> Preparation of Plasmid

DNA manipulations were performed by standard methods (Sambrook et al., 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY). PYES2 (Invitrogen) was treated with restriction enzymes Nde I and Apa I (Roche) to remove DNA fragments that would be used to destroy URA3 in HF7c . The intermediate 459 nucleotides of the URA3 gene were removed and re-ligated. This plasmid was amplified in Escherichia coli, and then treated with Nhe I and Nsi I (Roche) to separate only the ura3 portion and used for transformation of the yeast (FIG. 2A).

On the other hand, polymerase chain reaction (PCR) was performed to prepare DNA fragments to be used to manipulate Gal4p-dependently expressing URA3 , which is expressed in 476 yeasts, in the following two oligonucleotides. It was carried out using (Fig. 2 (B)).

Primer 1:

5-GGTATATATACGCATATGTGGTGTTGAAGAAACATG

AAATTGCCCAGTATGAATTCGAGCTCGTTTAAAC-3

Primer 2:

5-ACAGGACTAGGATGAGTAGCAGCACGTTCCTTATAT

GTAGCTTTCGACATCATTTTGAGATCCGGGTTTT-3

These primers were templated using pFA6a-kanMX6-PGAL1 (Longtine et al., 1998. Additional modules for versatile and economical PCR-based gene deletion and modification in Saccharomyces cerevisiae. Yeast 14: 953-961). It was used to amplify kan ^R and GAL1 promoters. 908bp cut out of pYES2 with Nde I and Nsi I for the isolation of URA3 gene, using pYES2 as a vector treated with restriction enzyme Hind III (Roche) for the production of plasmid pGUA to express URA3 in Gal4p-dependent manner The pieces were treated with Klenow enzymes and ligated to the above vectors to create pGU. Treatment of pASZ11 (Stotz et al., 1990. The ADE2 gene from Saccharomyces cerevisiae : sequence and new vectors. Gene 95: 91-98) with Bgl II and Klenau to use ADE2 as a selective marker in yeast of this plasmid The fragment obtained was ligated to the position treated with Nhe I and Bst1107 I of pGU. In addition, in order to change the selective marker of pGBT9 (Clontech) having a Gal4p DNA binding domain from TRP to ADE, the ADE2 fragment prepared above was ligated to the position treated with Pvu II and Ban II of pGBT9.

<2> Production of New Yeast Strains

HF7c [ MATa, ura3-52, his3-200, lys2-801, ade2-101, trp1-901, leu2-3,112, gal4-542, gal80-538, commonly used in two-hybrid systems cyhr2, LYS2 :: GAL1 _UAS -GAL1 _TATA -HIS3, URA3 :: GAL4 _{17mer (× 3)} -CyC1 _TATA -lac Z ] and 476 [ MATα, ura3-52, his3-200, lys2-801, ade1, ade2, trp1-901, leu2-3,112, gal4, gal80, URA3 :: GAL1-lac Z, lys2 :: GAL1-HIS3 ] (genotypes are all italic) The strain URA3 gene can be manipulated by Gal4p. Yeast cultures and transformation methods were performed by conventional reports (Cho et al., 1999. Analysis of the C-terminal region of Arabidopsis thaliana APETALA1 as a transcription activation domain.Plant Mol. Biol. 40: 419-429 ). In HF7c, URA3 center of the 459 nucleotides truncated mutant Δ ura3 the homologous recombination in the 0.1% 5-FOA (Diagnostic Chemical Limited) is raised in a minimal medium 3 to 4, which contains the yeast was introduced into the HF7c (homologous recombination) allows only the yeast whose normal URA3 is destroyed to grow. The yeasts obtained were squeezed again in uracil-deficient SD medium (SD-Ura) and observed growth to confirm that URA3 was destroyed. Into these strains with the expression plasmid (pGUA) of URA3 receive the control of GAL1 promoter pGBT9 and pY-AD (Cho et al. , 1999. Analysis of the C-terminal region of Arabidopsis thaliana APETALA1 as a transcription activation domain.Plant Mol. Biol. 40: 419-429; plasmids in which the DNA-binding portion and the transcriptional-activating portion of Gal4p are bound) were transformed, respectively, and Ura3p was also expressed depending on Gal4p. This new yeast strain will be referred to as PBN101 (Accession No .: KCTC 10155BP; Date of Consignment: January 5, 2002; Trustee: KCTC).

In the case of 476, more site-directed recombination was used. To put the GAL1 promoter in front of the URA3 coding part in the yeast, the DNA fragment containing the geneticin (G418) resistance gene (kan ^R ) and the GAL1 promoter between the promoter end and the URA3 coding part in the yeast Switched. First grow in YEPD medium for 2 days, then replicate and make replicas with YEPD (YEPD + G418 + FOA) containing the selective marker 200 μg / ml G418 (Calbiochem) and 0.08% 5-FOA (replica-plating). 2 days later, transfer to the same medium again to obtain only yeast with stable recombination. As before, expression of URA3 was stopped in SD-Ura, and expression of Gal4p-dependent URA3 was also confirmed using pGBT9 and pY-AD. This new yeast strain will be referred to as PBN201 (Accession No .: KCTC 10156; Accession: January 5, 2002; Trustee: Genetic Bank (KCTC)).

<3> Determination of 5-FOA Concentration to Eliminate False Positives

Establishes appropriate concentrations of 5-FOA to pre-remove yeast colonies before protein binding screening that exhibit false positive reactions that can easily occur when the library is applied to DLTS. In order to eliminate false positive reactions in the yeast caused by a library bound to the DNA-binding domain, several known transcriptional activation factors cloned into the bait vector pGBT9 were transferred to the PBN101 (HF7c engineered as above) strain. After transformation, when treated with a certain concentration of 5-FOA, yeast transformed with a bait vector grows well and finds a condition that can kill if it has a transcriptional activation factor. Known transcriptional activators include pGB-9, OsMADS1 CI and pGBT9-OsMADS1 CI and pGBT9-OsMADS1 CII (Lim et al., 2000. Two rice MADS domain proteins). interact with OsMADS1.Plant Mol Biol. 44: 513-27), and pGBT9 itself was also transformed. In all cases, pGUAs were co-transformed together. These were grown in SD -Ade, -Trp medium containing various concentrations of 5-FOA in the range of 0 to 0.1% for 3-5 days and observed growth. Among the various concentrations of 5-FOA, only the yeast transformed with pGBT9 can be lived, and the concentration of 5-FOA that kills all the other transcriptional regulators transformed.

On the other hand, protein binding pairs that have been previously reported were used to eliminate sham positive reactions when the library to which the transcriptional activation domains were bound alone was transformed (Kim et al., 2000. Protein-Protein Interaction among hnRNPs Shuttling between Nucleus and Cytoplasm.J. Mol. Biol. 298: 395-405). PGBT9-hnRNP C1 and pGAD424-hnRNP C1, which encode strong pairs of proteins BD-hnRNP C1 and AD-hnRNP C1, and pGBT9-hnRNP A1, which encode weak pairs BD-hnRNP A1 and AD-hnRNP A1 And pGAD424-hnRNP A1, and non-binding pGBT9 and pGAD424, respectively, were co-transformed into PBN201 (engineered 476) strains, followed by various concentrations of 5-FOA within the range of 0-0.2%. Grown in SD -Ade, -Leu medium for 3-5 days. As in the case of PBN101, live only when pGBT9 and pGAD424 (Clontech) are transformed together, and find the concentration of 5-FOA that dies when the remaining strong or weak bonds are present.

<4> Removal of fake positive reaction

Prior to screening full-scale protein binding at the library-to-library level, DLTS was performed using a variety of previously identified protein pairs. PGBT9-hnRNP A1, pGBT9-hnRNP E2, pGBT9-hnRNP C1, 8.3 μg each of PBN101 strain with 50 μg of pGUA plasmid, and a transcriptional activator capable of producing a false positive reaction [C of activating protein APETALA1 of Arabidopsis thaliana Terminal region (193-256 amino acids), Cho et al., 1999. Analysis of the C-terminal region of Arabidopsis thaliana APETALA1 as a transcription activation domain. Plant Mol. Biol. 40: 419-429] and 25 µg of pY-AP1 (193-256), which form a hybrid protein in which the DNA binding site of Gal4p is bound, were transformed. The PBN201 strain contained 8.3 μg of pGAD424-hnRNP A1, pGAD424-hnRNP E2, pGAD424-hnRNP C1, and 25 μg of plasmid (data not shown) that had previously shown false positives in two-hybrid screening. Transformation was carried out by mixing, with 50 μg of pGBT9 in which the selective marker was substituted with ADE2. Transformed PBN101 was grown in 14 SD -Ade, -Trp and transformed PBN201 in 14 SD -Ade, -Leu medium for 60 hours, then divided into two parts, one part was 5-FOA, and was sham positive. The reaction is removed, and the other part is grown for another 12 hours and then mated directly without removing the fake positive reaction. PBN101 transformants were transferred to the same medium containing 0.06% 5-FOA, and PBN201 transformants were transferred to the same medium containing 0.14% 5-FOA and incubated for 2 days. Colonies that survived this process were transferred back to the same medium lacking 5-FOA for recovery for 3 days.

<5> Searching for protein binding pairs by mating

The two transformed strains after removal of the fake positive reactions, which were carried out as described above, were incubated at 30 ° C. for 4-5 hours in an appropriate amount of liquid YEPD, respectively. The two cultures are mixed together, then aspirated with a Millipore HA filter (Millipore), and the yeast-containing filter is incubated at 30 ° C. on YEPD medium for mating. After 6 hours, wash the filters with water to let the yeast run out, and grow them in SD -Leu, -Trp, -His medium to obtain colonies that grow only in the presence of protein binding, and use these colonies to perform β-galactosidase assay. The expression of lac Z is also confirmed (Lim et al., 2000. Two rice MADS domain proteins interact with OsMADS1. Plant Mol Biol. 44: 513-27). The protein binding pairs obtained above were isolated from colonies showing protein binding by separating the DNA binding domain side and the transcriptional activation domain side plasmids, and transforming them separately or together into HF7c, and performing β-galactosidase filter assay with colonies obtained. Confirm that it is present. On the other hand, the part that eliminates the process of removing the false positive reaction by 5-FOA proceeds in the same manner and compare the results.

<6> Binding Gene Identification by Restriction Enzyme Treatment

Finally, the plasmids identified as protein binding pairs are treated with restriction enzymes Pvu II and Stu I. To determine the identity of the gene encoded by the plasmid from the truncated pattern after the enzyme treatment, pvu II and Stu I was treated. The treated DNA was electrophoresed using 0.8% agarose gel and the patterns were compared.

Experiments based on the materials and methods described above led to the following conclusions and confirmation experiments.

<7> Production of yeast strains required for DLTS and system optimization

HF7c and 476, which are used in the existing two-hybrid system, are expressed in Gal4p-dependent manner.HIS3Wowlac ZHas a reporter gene but is essential for eliminating false positivesURA3Genes are expressed by their regulatory mechanisms. Many fake positive reactions that can occur when performing DLTS using the libraryURA3In order to eliminate the expression using, the expression of this gene is usually stopped and needs to be manipulated so that it occurs only in Gal4p. For this purpose, in the case of HF7cura3(Δura3) Transform the yeast with a coating film and grow in medium containing 5-FOA Δura3Only the yeast substituted with the surviving [Fig. 2 (A)], in the case of 476 induce site-directed recombination in the existingURA3The Gal4p was successfully manipulated to regulate its expression (Fig. 2 (B)). Δ into HF7cura3After transforming, four candidate colonies growing in SC medium containing 0.1% 5-FOA were obtained and all of them were found not to grow in SD-Ura, and the best growth candidates were selected among them. When pGBT9 and pY-AD were transformed together with pGUA in this strain, pGBT9 transgenic yeastURA3Expression does not occur in yeast transformed with pY-AD, depending on the Gal4p DNA binding domain and the transcriptional activation domain.URA3It was found that this is expressed. 476 In the case of yeastURA3Kan, an optional marker, with 50 nucleotides at both ends of the gene's coding start and upstream^RDNA fragments with and GAL1 promoters were synthesized by polymerase chain reaction (PCR). Therefore, in this case, fundamentally manifestURA3By inserting the GAL1 promoter precisely between the coding start site and the upper part of, the expression of Ura3p can be changed Gal4p-dependently, and unlike in the case of HF7c, pGUA does not need to be transformed together to induce Ura3p. There is this. 476 The yeasts were transformed into 476 yeasts and incubated for 2 days in YEPD to obtain numerous colonies, and 50 of them were transferred to YEPDs containing both G418 and 5-FOA to pick out yeasts with the DNA fragments in place. I got enough colonies. Some of these were transferred back to the same medium to obtain stable 14 colonies, but when cultured in SD-Ura as in the case of HF7c, it was confirmed that only 3 colonies did not grow. Also, with pGBT9 When pY-AD was transformed into these strains to detect Gal4p-dependent Ura3p expression, only two colonies were suitable, and among them, strains with stronger Ura3p induction were selected. The new strains thus obtained were named PBN101 and PBN201, respectively, and the characteristics of the strains identified above are shown in FIG. 3. That is, in PBN101 and PBN201ura33 does not occur in the medium containing 5-FOA, but grows well [Fig. 3 (A)], SD-Ura is not growing [Fig. Only ifURA3It can be seen that the Gal4p dependent aspect expressing the results of the proliferation in the medium without the uracil [FIG. 3 (C)] and the medium containing 5-FOA [FIG. 3 (D)].

If these strains were transformed with only bait and prey libraries and added 5-FOA, yeasts expressing Ura3p by sham positive reactions could not be grown, thus effectively eliminating false positive libraries. have. However, if too high concentrations of 5-FOA are added, the yeast strain itself does not grow well, so it is essential to treat 5-FOA at an appropriate concentration that can distinguish and eliminate yeasts that do false positives. to be. In other words, the concentration of 5-FOA that efficiently eliminates false positives while minimizing the reduction of the diversity of transformed libraries should be determined and the detail process should be optimized. To do this, after transforming the two strains with plasmids that can cause a positive reaction, these yeasts must find a concentration of 5-FOA that kills the yeast that does not cause a positive reaction. The bait library with DNA binding domains can cause a false positive reaction in yeast because a fragment with some transcriptional activator is bound behind the DNA binding domain. The DNA of the transcriptional activator was transformed with pGUA. The strong transcription activator Gal4p transcriptional activity domain, the previously reported CI of OsMADS1, rather weak CII domain, and DNA binding domains were tested. Since the PBN101 strain itself does not grow well at 5-FOA concentrations higher than 0.1%, the above transformed yeasts were cultured in medium containing 5-FOA at 0.02%, 0.04%, 0,06%, 0.08%, and 0.1% concentrations. It grew and observed the extent of its growth. In the case where only DNA binding domain was transformed in 0.06% of FOA (BD in FIG. 4A), it grew well but the other strong or weak transcriptional activators were transformed (in (A) of FIG. 4). BD-AD, BD-CI, and BD-CII) did not grow, so it was determined that the appropriate concentration of 5-FOA was 0.06% [Fig. 4 (A)].

On the other hand, since PBN201 has a library associated with a transcriptional active domain, a false positive reaction is likely to occur due to a mechanism different from that of a bait library. Therefore, we attempted to find a concentration of 5-FOA that transformed existing reported protein binding pairs so that a positive reaction could occur and then only those yeasts could be selectively removed. PGBT9 and pGAD424 with no positive reaction, pGBT9-hnRNP A1 and pGAD424-hnRNP A1 with slightly weaker binding, pGBT9-hnRNP E2 and pGAD424-hnRNP E2 with moderate binding, and pGBT9-hnRNP C1 with strong binding Each was transformed with DNAs of pGAD424-hnRNP C1. Since PBN201 strain itself grew well even at higher concentrations of 5-FOA than in the case of PBN101, it was observed to grow transgenic yeast in medium made with a difference of 0.02% in the range of 0.1% to 0.2%. Among them, only yeast transformed with vectors (BD + AD in FIG. 4B) in a medium containing 0.14% 5-FOA can be grown, and pGBT9-hnRNP A1 and pGAD424-hnRNP A1 are transformed together. Since yeasts (A1 + A1 in FIG. 4B) did not grow, the optimal concentration of 5-FOA was determined to be 0.14% [FIG. 4 (B)].

In performing DLTS, if each strain was transformed with a bait and prey library and mated surviving yeasts after removing the false positive reaction at the 5-FOA concentration determined above, Finally, the bait and prey libraries are in one yeast, and the protein-binding pairs between them produce a reporter, resulting in a new positive reaction. At this time, the yeast survived before mating so that the growth of the mating diploids or the real positives caused by protein binding pairs are not affected by the previous removal process using 5-FOA. Should be grown and recovered in a medium without 5-FOA.

<8> DLTS using several protein binding pairs

Before identifying binding pairs between proteins at the library level in earnest, DLTS was performed with known protein binding pairs (FIG. 5). At this time, DLTS was performed in a state in which a large amount of DNA that had already been shown to be false positive was mixed in order to confirm that only a positive reaction by a protein binding pair could be effectively detected in the presence of a false positive protein. The final results were compared with that of the same experimental group that eliminated the false positive reaction. That is, in PBN101, 50% of DNA to be transformed with hnRNP A1, hnRNP E2, and hnRNP C1 bound to the DNA binding domain is used, and the DNA of pY-AP1 (193-256), a previously reported transcriptional active domain, is remaining 50%. And transformed together. PBN201 contains 50% of hnRNP A1, hnRNP E2, and hnRNP C1 bound to the transcriptional active domain, and mixes the remaining DNA of the prey phony positive reaction in the conventional two-hybrid screening. Transformation. The yeasts were transferred to a medium containing 0.06% and 0.14% 5-FOA to remove the false positive reaction, and some experimental groups skipped this procedure. After removal of fake positive reactions, the yeasts that had undergone sufficient recovery were mated with each other and grown in SD -Leu, -Trp, -His, resulting in 148 yeast colonies expressing HIS3 , of which 102 colonies It has been shown to express another reporter, β-galactosidase. Of these, 20 colonies were randomly selected and the plasmids were taken out to confirm whether the positive reaction was fake or represented by protein binding pairs. Only 9 candidate colonies were due to protein binding pairs (Table 1).

In order to identify the identity of protein binding pairs from the nine colonies, cleavage patterns by restriction enzymes were used. Protein binding pairs are expected to be hnRNP A1-hnRNP A1, hnRNP E2-hnRNP E2, or hnRNP C1-hnRNP C1, so that the bait and prey plasmids from the colonies containing the protein binding pair are After isolation, the protein binding pairs were identified by comparing them with Pvu II and Stu I, restriction enzymes that can distinguish them from each other. As a result, it was found that most of the binding between hnRNP C1-hnRNP C1 shows the strongest binding. On the other hand, about 900 HIS3 expressing colonies were obtained from the experimental group that omitted the false positive reaction process, and most of them expressed β-galactosidase. However, when 20 colonies were randomly selected and tested for false positives, 19 were false positives, and the process of removing fake positives from DLTS can perform more efficient library-to-library binding search. It has been shown to be an essential process to ensure it is (Table 1). Although there was 55% of false positives in the experimental group that removed the false positives using 5-FOA, a large amount of artificially positive (50% of the total DNA) of pseudo-positive DNA was added during transformation. Considering this, it can be said to be an effective removal, and if necessary, the removal process may be repeatedly performed to improve the result.

As described above, the present invention can provide a system for efficiently searching for binding pairs of library-level proteins simultaneously instead of a single protein.

Claims (9)

a) preparing a type strain that disrupts URA3 in the yeast strain, and then introducing a plasmid in which URA3 is expressed by the GAL1 promoter; b) producing an α-type strain by directly modifying URA3 in the yeast strain to be expressed by the GAL1 promoter using a genetic recombination method; c) transforming the prepared a type strain and the α type strain into a bait library and a prey library, respectively; d) selectively removing the fake positive reaction while culturing the transformed a type strain and α type strain in a medium containing 5-FOA (5-fluoroorotic acid), respectively; And, e) mating the two types of transformed yeast that survived the selection process to detect protein-protein binding pairs. The method of claim 1, further comprising the step of recovering the yeast remaining after the step d) of claim 1. The method of claim 1, wherein the a type strain is HF7c URA3 a double library protein binding, characterized in that a yeast strain PBN101 that is destroyed by homologous recombination in yeast search method. The method of claim 1, wherein the α-type strain is a yeast strain PBN201 which is engineered to insert the GAL1 promoter between the coding region of the URA3 of 476 yeast and its upper region so that expression of URA3 is Gal4p-dependent Library protein binding search method. The method of claim 1, wherein the bait library comprises a gene encoding a DNA binding domain and a bait protein of GAL4p, and the prey library includes a gene encoding a transcriptional activation domain and a prey protein. Way. The method of claim 1, wherein the concentration of 5-FOA (5-fluoroorotic acid) is 0.05 to 0.07% for PBN101 and 0.13 to 0.15% for PBN201. The method of claim 1, wherein the detection of the protein-protein binding pairs uses a HIS3 or lac Z reporter gene. Yeast strain PBN101 (Accession No .: KCTC 10155BP) characterized by the destruction of URA3 in HF7c yeast by homologous recombination. Of between 476 yeast URA3 coding region and the upper region of the URA3 expression so as to interrupt into the GAL1 promoter, the operation such that the dependent Gal4p characterized PBN201 yeast strain (accession number: KCTC 10156BP).