Genome-wide construction of Schizosaccharomyces pombe heterozygous deletion mutants containing gene-specific barcodes by the methods of 4-round serial or block PCR, or total gene synthesis thereof

Abstract

A method comprising transforming Schizosaccharomyces pombe with a deletion cassette, constructed by four-round serial PCR, block PCR or total gene synthesis, containing a homologous recombination site is provided for preparing gene-targeted heterozygous deletion Schizosaccharomyces pombe. Also provided are gene-targeted hetero2ygous deletion Schizosaccharomyces pombe mutants prepared by the method, and a library of gene-targeted heterozygous deletion Schizosaccharomyces pombe mutants. Further, the library is useful in constructing a method and a kit for screening a drug's modes of action.

Claims

1. A method for constructing gene-targeted heterozygous Schizosaccharomyces pombe mutants, comprising: 1) constructing a deletion cassette for gene targeting by block PCR or gene synthesis; 2) introducing the deletion cassette into a Schizosaccharomyces pombe strain; and 3) identifying gene-targeted Schizosaccharomyces pombe mutants in which a target gene is replaced by a selectable marker gene of the deletion cassette, wherein the deletion cassette comprises a selectable marker gene; a pair of gene-specific barcode sequences, positioned respectively at both flanks of the selectable marker gene, and a pair of homologous recombination regions, assigned respectively to both sides of the barcode sequences positioned upstream and downstream of the selectable marker gene, wherein the homologous recombination regions for block PCR range from 350 bp to 450 bp, and those for gene synthesis ranges from 150 bp to 250 bp, and wherein the barcode sequences are selected from the group consisting of the base sequences of SEQ ID NOs: 18 to 9993, wherein for the block PCR, the deletion cassette is constructed by fusing 5 and 3 DNA fragments for the homologous recombination regions separately synthesized by PCR to 5 and 3 ends of a DNA fragment comprising the selectable marker gene and gene-specific barcode sequence through block PCR, and wherein for the gene synthesis, the deletion cassette is constructed by designing three DNA fragments of a) the 5 homologous recombination region bearing the 5 barcode, b) the selectable marker, and c) the 3 homologous recombination region bearing the 3 barcode, and connecting these three DNA fragments using 5- and 3-link oligos.

2. The method according to claim 1, wherein each of the barcode sequences ranges from 20 bp to 30 bp.

3. The method according to claim 1, wherein the barcode sequences are selected from the group consisting of the sequences of SEQ ID NOs: 18 to 9993 and two of the barcode sequences are respectively assigned upstream and downstream of the selectable marker gene.

4. The method according to claim 1, wherein the selectable marker gene is selected from the group consisting of a kanamycin resistance gene, a neomycin resistance gene, a hygromycin resistance gene, a tetracycline resistance gene, auxotrophic markers, histidinol dehydrogenase (hisD), guanine phosphoribosyltransferase (gpt), hypoxanthine phosphoribosyltransferase (hprt), ura3, ble and sacB.

5. The method according to claim 4, wherein the selectable marker gene is a kanamycin resistance gene.

6. The method of claim 1, wherein the deletion cassette further comprises a pair of universal primers for amplifying the barcode sequences positioned at the 5 side of the selectable marker gene and a pair of universal primers for amplifying the barcode sequences positioned at the 3 side thereof.

7. The method of claim 1, wherein the target gene is selected from the group consisting of erg1, rps901, rpa2, smb1, rps2402, rp11801, cct1, rpp13702 and pmm1.

8. A library of gene-targeted heterozygous fission yeast mutants prepared by the method according to any one of claims 1, 2-4, 5, 6, and 7.

9. A method of screening a drug's mode of action, comprising: 1) treating the library of gene-targeted heterozygous fission yeast mutants according to claim 8 with a chemical; 2) culturing the chemically treated fission yeast mutants; 3) isolating genomic DNA from the cultured mutants; and 4) measuring and analyzing growth rates of the mutants with the isolated genomic DNA to detect strains whose relative growth is inhibited.

10. A screening kit for a drug's mode of action, comprising the library of gene-targeted heterozygous fission yeast mutants according to claim 8.

11. The method of screening a drug's mode of action according to claim 9, wherein the growth rate of the mutant is measured by applying the isolated genomic DNA to microarray gene chip or to quantitative real-time PCR.

Description

DESCRIPTION OF DRAWINGS

(1) FIG. 1 is a schematic view showing a structure of a typical deletion cassette for gene targeting,

(2) FIG. 2 shows a process of constructing a Kan.sup.r-barcode module for use in a deletion cassette,

(3) FIG. 3 shows a process of constructing a deletion cassette through four-round serial PCR,

(4) FIG. 4 shows a process of constructing a deletion cassette through block PCR,

(5) FIG. 5 shows a process of constructing a deletion cassette through gene synthesis,

(6) FIG. 6 shows the identification of a desired mutant strain by colony PCR,

(7) FIGS. 7 to 55 show base sequences of gene-specific barcodes to be inserted into deletion cassettes for the construction of gene-targeted mutant strains

(8) FIG. 56 schematically shows a process of screening a drug's modes of action using GeneChip on the basis of haploinsufficiency,

(9) FIG. 57 is a process of preparing a library of gene-targeted mutants for screening a drug's modes of action by use of GeneChip,

(10) FIG. 58 shows an order-made GeneChip from Affymetrix,

(11) FIG. 59 shows the measurement of the antifungal agent terbinafine for IC.sub.50 applied to the screening of a drug's modes of action on Genechip, and

(12) FIG. 60 shows the reconfirmation of the Genechip-based terbinafine's modes of action on plates.

BEST MODE

(13) In the present invention, four-round serial PCR, block PCR or gene synthesis is employed to amplify deletion cassette modules with which transformed gene-targeted mutants are then constructed. In turn, the mutants are used to construct a library of DNA of which advantage is taken to search for a drug's modes of action.

(14) In an embodiment of the present invention, provided is a method for constructing gene-targeted heterozygous deletion mutants of fission yeast using a deletion cassette which comprises (a) a selectable marker, (2) a pair of gene-specific barcode base sequences, positioned respectively at both flanks of the selectable marker gene, for a microarray, and (3) a pair of at least 150 bp homologous recombination regions, respectively assigned to sides of the barcode base sequences at positions upstream and downstream of the selectable marker gene. A detailed description will be given of each component, below.

(15) Selectable Marker

(16) A selectable marker is used to discriminate successfully transformed yeasts from unsuccessful ones. Examples of the selectable marker useful in the present invention include auxotrophic markers, neomycin resistance (neo.sup.R), kanamycin resistance (kan.sup.R), hygromycin resistance (hyg), histidinol dehydrogenase (hisD), guanine phosphoribosyltransferase (gpt), tetracycline (tet.sup.R), hypoxanthine phosphoribosyltransferase (hprt), ura3, ble and sacB, with preference for kanMX, which is derived from kanamycin resistance gene. The deletion cassette comprising a selectable marker may be integrated into the genome by homologous recombination.

(17) An autographic marker that requires a specific amino acid or nucleic acid for survival, such as ura4, leu1, his3, is currently used for targeted gene integration and disruption in fission yeasts. However, a strain with one or more auxotrophic mutations is requisite for the use of such a marker. Also, a gene conversion may occur in the auxotroph due to the introduction of DNA. Further, auxototrophic mutation itself; together with other mutations, may result in phenotypes which are difficult even to guess, such as somosensitivity, nitrogen-starvation regulated cellular differentiation, sporulation, pseudohyphae, etc. In addition, many auxotrophic mutations may make it difficult for the yeast to grow in a typical medium.

(18) In order to overcome these limitations of auxotrophic markers, the Philippsen group developed the KanMX module as a dominant drug resistance marker. This marker was used for the deletion project of Saccharomyces cerevisiae. Resistant to Geneticin (G418), the kanMX module was designed to serve as a selectable marker. Over auxotrophic markers, the G418 marker has the advantage of giving a high rate of deletion even upon the use of PCR-DNA fragments. Now, additional dominant drug resistance markers have been developed, including those resistant to hygromycin B, nourseothricin, and bialaphos/phosphinothricin.

(19) Used in a preferred embodiment of the present invention is the selection maker KanMX4 which is 810 bp in length and contains Kan transposon Tn903 ORF from E. coli. With aminoglycoside phosphotransferase activity, KanMX4 is reported to inhibit kanamycin or its derivative G418 (Lang-Hinrichs et al., Current Genetics. 18:511-6, Oka et al., J. Mol. Biol. 147:217-26). This kanMX module, one of the dominant drug resistance markers, contains the known kan.sup.r open reading-frame (ORF) of the E. coli transposon Tn903 fused to transcriptional and translational control sequences of the TEF gene of the filamentous fungus Ashbya gossypii. This heterogonous module permits efficient selection of transformants resistant against geneticin (G418).

(20) A high false rate of integration to the target region may generally occur if PCR products are used, or if a marker has low sequence homology to the target region. In contrast, the KanMX module (G418 marker) serving as a heterologous selectable marker guarantees a high rate of deletion thanks to its dominant resistance even if PCR-DNA fragments are used.

(21) Barcode Sequences for Microarrays

(22) In order to efficiently measure growth rates of a library of (6,000) gene-targeted fission yeast mutants, a practical and systematic strategy is necessary for the growth measurement through a gene chip technique. In this regard, barcode sequences for microarrays are introduced into the deletion cassettes (FIG. 3). These barcode sequences make it easy to detect the strains whose growth is inhibited by a drug in such a manner that a chromosome pool isolated from 6,000 strains is used as a template for PCR amplification of barcodes regions, followed by analysis on a gene chip. The barcode sequences must be specifically designated to each strain, thus requiring a population two or more times greater than that of the genomic 1 DNAs of the strains (each one assigned upstream and downstream of genomic DNA, one genomic DNA flanked by at least two barcode sequences). As long as it satisfies this condition, there is no limitation imparted to the length of the barcode sequences.

(23) In a preferred embodiment of the present invention, the KanMX4 of the deletion cassette is flanked by two copies of a 20-mer gene-specific barcode. The theoretically possible number of 20-mer barcodes is 4.sup.20 because the four bases G, A, T and C are used in the barcode sequences. Barcode sequences used in the present invention are those shown in FIGS. 7 to 55.

(24) Homologous Recombination Regions

(25) A success rate in constructing gene-targeted strains is dependent on properties of strains because it is determined by homologous recombination efficiency which differs from one strain to another. However, the homologous recombination efficiency in the same strain is highly proportional to the length of genomic homologous regions (Michael et al., Gene. 158:113-17; Palaniyandi et al., Nucl Acid Res. 3:2799-2800). Thus, the DNA fragments corresponding to genomic homologous regions must have their lengths optimized in order to construct gene-targeted strains with efficiency.

(26) According to an embodiment of the present invention, this homologous recombination region is 150 bp or greater in length and is positioned at each of 5 and 3 sides in the deletion cassette. Preferably, the homologous recombination region ranges in length from 150 bp to 450 bp. A more preferable embodiment of the present invention provides a deletion cassette in which the homologous recombination region with a length of from 150 to 450 bp is positioned at each of 5 and 3 sides.

(27) TABLE-US-00002 TABLE 2 Construction of Deletion Homologous Success Cassettes Yeasts Length(bp) rate(%) Note PCR with Serial PCR Budding 45~60 81 gene- Single Budding 38~50 10 specific Oligo Oligo Single Fission 40 1~3 Oligo 4-Round Fission 80 50 Present Serial PCR invention Block PCR Fission 350~450 80 Gene Synthesis Fission 150 80~98

(28) Construction of Deletion Cassette for Gene Targeting

(29) The deletion cassette of the present invention plays an essential role in targeting about 5000 yeast genes in such a manner that a marker gene of the deletion cassette transformed into the yeast is substituted for a target gene. As a prerequisite to the construction of gene-targeted, heterozygous fission yeast mutants, the deletion cassette preferably has a gene structure in which the selectable marker is flanked by the gene-specific barcodes which are in turn respectively inserted to chromosome-homologous recombination regions of suitable lengths (FIG. 1).

(30) In a preferred embodiment of the present invention, the KanMX4 module is designed to express the kanamycin resistant gene of Tn903 in the fission yeast, having the structure in which the resistance gene is preceded by the 381 bp TEF promoter sequence of Ashbya gossypii just before the initiation codon ATG thereof and followed by the 242 bp TEF terminator of Ashbya gossypii just after the stop codon thereof. When used, the promoter or terminator of the budding yeast (Saccharomyces cerevisiae) or the fission yeast itself also acts as an additional chromosome-homologous region, resulting in a false homologous recombination. For this reason, the promoter used in the deletion cassette is preferably from a heterogeneous yeast. The KanMX4 module thus constructed becomes 1,433 bp in length. After the deletion cassette is introduced into the fission yeast through transformation, the marker gene undergoes homologous recombination with a chromosomal target gene, so that the marker gene is inserted into the chromosome while the target gene is deleted.

(31) In an example of the present invention, a primary PCR was performed in the presence of a pair of 70 bp-long 5 and 3 barcode primers, each containing a 20 bp barcode, with the kanamycin resistance gene (Kan.sup.R) serving as a template, to afford a Kan.sup.R-barcode module (FIG. 2) which was in turn used as a template for secondary PCR with a pair of 6080 bp primers, each consisting of a 4060 bp chromosome-homologous recombination sequence and a 20 bp-long 5 or 3 terminal region of the Kan.sup.R-barcode module, to produce a deletion cassette, useful as a component of the present invention.

(32) Four-Round Serial PCR, Block PCR and Gene Synthesis for Construction of Deletion Cassette

(33) The construction of the deletion cassette for gene targeting in accordance with the present invention is accomplished by (1) four-round serial PCR; (2) block PCR; or (3) gene synthesis, preferably as follows.

(34) (1) The term four-round serial PCR, as used herein, is intended to refer to PCR which is performed four times in series. A primary PCR is performed with a pair of about 70 bp-long 5 and 3 barcode primers to afford a barcode module which is in turn used as a template in three successive PCRs with a pair of 50-mer primers for the first two and with a pair of 40-mer primers for the final one, so as to give a gene-specific barcode and a 80 bp-long gene-homologous recombination region.

(35) (2) The term block PCR, as used herein, is intended to refer to a block PCR which is performed with 5 and 3 DNA fragments for homologous recombination on the chromosome, which are separately produced to a length of 350500-bp in length by PCR, in admixture with the produced Kan.sup.r-barcode module, to afford a deletion cassette.

(36) (3) The term gene synthesis is quite different from PCR and enjoys the following advantages: first, the gene synthesis method can be conducted by one round of phosphorylation and ligation and two rounds of PCR while the four-round PCR is accomplished by four rounds of PCR and four rounds of purification. Thanks to the simplicity thereof, the gene synthesis achieved the genome-wide construction of heterozygous deletion mutants at a success rate of 90% or higher whereas a success rate of 80% was obtained with the block PCR (Table 2); secondly, the gene synthesis can be conducted in series on 96-well plates, thus improving workability. For example, one experienced worker can construct ten strains a week through conventional PCR. In contrast, gene synthesis enables a worker to construct 100 or more strains a week, increasing the efficiency at least 10 times. Finally, only two rounds of PCR are needed whereas the conventional PCR is conducted four times in series, which decreases the possibility of point mutations by two or more times. For instance, 20 strains randomly selected from each group of the mutants constructed according to the methods were subjected to PCR for the amplification of an approximately 2.1-kb target gene, followed by DNA sequencing to find mutations at a rate of 0.2 bases/1 kb, which is two or more times lower than 0.5 bases/1 kb in the conventional PCR.

(37) FIG. 1 is a gene map illustrating a deletion cassette constructed by gene synthesis according to the present invention. The methods (1) and (2) are common in that deletion cassettes are constructed on the basis of PCR, but the method (3) is different in that the construction of deletion cassettes is based on the ligation of synthetic oligonucleotides, but not on PCR. Particularly, the method (3) permitted successful construction of about 200 heterozygous gene-targeted mutants which are difficult to make by conventional PCR methods, showing a success rate as high as 97% from a total 4,988 genes of the fission yeast.

(38) In accordance with an aspect thereof, the present invention provides a high-throughput screening system of in vivo drug's modes of action in which a library of the heterozygous deletion mutants of fission yeast is prepared with the deletion cassettes constructed as described above and is applied to order-made gene chips. In a preferred example of the present invention, the performance of the system was demonstrated by use of the antifungal agent terbinafine.

(39) Heterozygous Fission Yeast

(40) Among the fission yeast group are Schizosaccharomyces octosporus and Schizosaccharomyces pombe. Any fission yeast may be used in the present invention. The mutant strains transformed with the deletion cassettes are heterozygous deletion mutants in which the gene-specific sequences of the deletion cassettes are exchanged with the chromosomal target genes of the fission yeast by homologous recombination to delete the target genes from the chromosome of the fission yeast.

(41) In a preferable embodiment, the fission yeast may be Schizosaccharomyces pombe, but is not limited thereto. The fission yeast Schizosaccharomyces pombe, useful in a preferable embodiment of the present invention, was first isolated from African millet beer and is widely used as a model eukaryote for cell and molecular studies as well as in the production of foods such as beer, breads, etc. S. pombe is particularly useful in investigating the cell cycle of cells which proliferate through equal division. Introns are distributed among 43% of S. pombe genes, showing the feature of higher eukaryotic organisms. Found to have a compact genomic structure with about 5,000 ORFs (open reading frames), the fission yeast is in addition used as a material suitable for gene analysis. Further, S. pombe is recognized for making broad contributions to the life sciences, particularly with applications involved with the development of new drugs.

(42) In accordance with a further aspect thereof, the present invention pertains to a gene-targeted, heterozygous deletion mutant of fission yeast which is prepared with a gene-target deletion cassette comprising a selectable marker gene, barcode gene sequences for microarray analysis, and homologous recombination regions.

(43) In accordance with still a further aspect thereof; the present invention pertains to a library of gene-targeted heterozygous fission yeast mutants prepared by the method of the present invention.

(44) In accordance with still another aspect thereof the present invention pertains to a method for screening a drug's modes of action by chemically treating a library of heterozygous fission yeast mutants, culturing the chemically treated library, isolating total genomic DNA from the culture, and applying the isolated genomic DNA to microarray GeneChip or to real-time PCR.

(45) As a tool for identifying a drug's modes of action, a library of heterozygous fission yeast mutants is provided. To identify a drug's mode of action is essential to the improvement of preexisting drugs in medicinal efficiency and the development of a new drug free of or with few side effects. The method for screening a drug's mode of action in accordance with the present invention comprises the steps of:

(46) 1) treating a library of heterozygous fission yeast mutants with a chemical;

(47) 2) culturing the library of the chemically treated, heterozygous fission yeast mutants;

(48) 3) isolating genomic DNA from the cultured library; and

(49) 4) measuring and analyzing growth rates of the mutants with the isolated genomic DNA applied on microarray gene chips to detect strains which are relatively inhibited from growing.

(50) As long as it inhibits the growth of the mutants, any chemical may be used in the present invention. In a preferred example of the present invention the antifungal agent terbinafine, which targets the gene erg1, was used to examine whether the screening system of a drug's modes of action worked well or not.

(51) As long as it is known in the art, any technique may be used to isolate genomic DNA from the fission yeast. In a preferred example of the present invention, the fungal/bacterial DNA kit (Zymo Research, catalog #D6005) was employed to obtain genomic DNA.

(52) The term gene chip or GeneChip, as used herein, refers to a tool for analyzing the expression of genetic information, which is typically composed of an arrayed series of hundreds to tens of thousands of microscopic spots of DNA oligonucleotides on a solid support of about 1 cm.sup.2. The DNA oligonucleotides of known, specific genetic information are arranged on the surface of the chip and are used as probes to hybridize a target under high-stringency conditions. Probe-target hybridization is usually detected and quantified by fluorescence-based detection of fluorophore-labeled targets to determine a relative abundance of nucleic acid sequences in the target. These gene chips are not only useful for gene studies, but also can detect gene-related diseases quickly. Further, these new analysis systems find a wide spectrum of applications in, for example, selecting or designing optimal customized drugs.

(53) As long as it recognizes the gene-specific barcode sequence inserted to the deletion cassette, any gene chip may be used in the present invention. A preferable embodiment of the present invention employed deletion cassettes with 20-mer barcodes inserted thereinto, and an order-made GeneChip with a serial number of KRIBBSP1-a520429 from Affymetrix to detect the barcodes.

(54) In an alternative embodiment of the present invention, real-time PCR, instead of GeneChip, was performed with inter-chelating dye or flourophore-labeled probe to screen a drug's modes of action. A detailed description will be given of detection with probes, below.

(55) TaqMan probes and cycling probes have been developed for use as fluorescent probes in real-time PCR. The former is a technology utilizing the 5.fwdarw.3 exonuclease activity of the Taq DNA polymerase for measuring the amount of target sequences in the samples while the cycling probe technology is a method utilizing a combination of chimera probe, composed of RNA and DNA, and RNase H. In this method, one end of the probe is labeled with a fluorescent substance and the other end is labeled with a quencher, which quenches the fluorescence emitted from the fluorescent substance. When this probe forms a hybrid with the complementary sequence of amplified product, RNase H specifically cuts the RNA region of this probe, resulting in emissions of strong fluorescence. Below, a detailed description will be given of the TaqMan probe technology for screening a drug's modes of action.

(56) In a preferable embodiment of the present invention, real-time PCR is conducted with a Taqman probe in addition to common primers. A Taqman probe is an oligonucleotide which is labeled with a reporter fluorophore at the 5 end and with a quencher at the 3 end. During PCR, the probe anneals specifically between the forward and reverse primers to an internal region of the PCR product at the annealing step, with the fluorescence quenched by the quencher. While the polymerase then carries out the extension of the primers and replicates the template to which the TaqMar probe is bound, the 5 exonuclease activity of the polymerase cleaves the probe, releasing the reporter molecule away from the close vicinity of the quencher. As a result, the fluorescence intensity of the reporter dye increases and can be detected to determine the expression of the target gene. In the present invention, a Taqman probe may be constructed to have a barcode sequence as the oligonucleotide sequence. In this case, real-time PCR with each Taqman probe gives quantitative data on the real time expression of corresponding barcodes, on the basis of which significantly increased or decreased regions can be regarded as a drug's modes of action.

(57) In accordance with still another aspect thereof, the present invention pertains to a novel drug-screening kit, comprising the library of heterozygous fission yeast mutants.

(58) The drug-screening kit of the present invention is used in the method for screening a drug's modes of action. In greater detail, the screening kit is utilized in screening novel drugs as well as a drug's modes of action by 1) culturing the library of heterozygous fission yeast mutants of the screening kit in media containing respective drug candidates and 2) comparatively analyzing the growth of the fission yeast mutants to detect growth-inhibited fission yeast mutants.

MODE FOR INVENTION

(59) A better understanding of the present invention may be obtained through the following examples which are set forth to illustrate, but are not to be construed as limiting the present invention.

Example 1

Construction of Deletion Cassette by Four-Round Serial PCR or Block PCR

(60) Deletion cassettes can be constructed in two different PCR techniques: four-round serial PCR and block PCR. These two PCR-based methods are common in preparing a Kan.sup.r-barcode module in advance. The base sequence of KanMX4 is well known in the art and is not described specifically. Primary PCR was performed with a pair of 70-mer 5- and 3-barcode primers, each containing a gene-specific barcode while KanMX4 served as a template to prepare a Kan.sup.r-barcode module (FIG. 2). After electrophoresis on agarose gel stained with ethidium bromide (EtBR), the Kan.sup.r-barcode module thus prepared was extracted from the gel by excising under a UV lamp.

(61) The four-round serial PCR for constructing deletion cassettes is similar to the conventional one used for constructing gene-targeted budding yeast or fission yeast mutants, but different thereto in that 80-bp homologous recombination region is added to either ends of the deletion cassette (Giaever et al., Nature Genetics 14:450-56). Dr. Giaever et al. succeeded in targeting 90% or more of the total 6000 genes of the budding yeast by providing a 45-bp recombination region in one PCR, with Kan.sup.r-barcode module serving as a template. As for the fission yeast, however, the 45-bp homologous recombination region is too short to effectively induce homologous recombination on the genome. Thus, the method used for the construction of gene-targeted fission yeast mutants was modified such that PCR was performed three times in series with a pair of 50-mer primers for first two rounds and with a pair of 40-mer primers for the last round, to give a 80-bp homologous recombination region together with a gene-specific barcode (FIG. 3).

(62) This four-round serial PCR was successfully applied only to a quarter of the total 4,988 genes of the fission yeast, with a success rate of 50% or higher in the construction of mutant strains, but failed for the remaining three fourths. Thus, the homologous recombination region was extended using block PCR (FIG. 4). In the block PCR, 5 and 3 DNA fragments for chromosome-homologous recombination, each 350500 bp long, are separately synthesized by PCR and then subjected, together with the purified Kan.sup.r-barcode module, to block PCR to prepare a deletion cassette in which the 5 and 3 350500 bp DNA fragments are fused to 5 and 3 ends of the Kan.sup.r-barcode module, respectively. As a result, the chromosome-homologous recombination regions in the deletion cassette are extended to 350500 bp from the 80 bp of the four-round serial PCR, remarkably increasing the construction efficiency of deletion mutants to 90% from 50%.

Example 2

Construction of Deletion Cassette by Gene Synthesis

(63) Even the PCR technology described above failed to prepare gene-targeted mutants for as many as 4% of the total genes. The reason is because the high contents of both adenine (A) and thymine (T) in the promoter and the terminator of the gene do not allow for primer sites of conventional PCR. In consideration of this problem, the gene synthesis method was employed. The present invention is the first to use gene synthesis in constructing deletion cassettes and preparing deletion fission yeast mutants.

(64) A deletion cassette was designed to consist of three fragments: 1) 5 chromosome-homologous region bearing 5 common liker sequence and 5 barcode, 2) kanamycin resistance gene (Kan.sup.r), and 3) 3 barcode and 3 chromosome-homologous region bearing 3 common liker sequence. To connect these three fragments into a one line, overlapping link oligo-sequences are arranged between the fragments to be connected with each other (FIG. 5). A length of 250 bp was given as the length of the chromosome-homologous region, which is the most important factor to determine the construction efficiency of mutants. The 5 chromosome-homologous region was identical to the chromosomal nucleotide sequence stretching in the direction toward the promoter from the initiation codon ATG by 250 bp while the 3 chromosome-homologous region was identical to the chromosomal nucleotide sequence stretching in the direction toward poly-A from the stop codon TGA, TAG or TAA, by 250 bp.

(65) The 5 and the 3 chromosome-homologous region, each 250 bp long, are different from one gene to another and the description of their nucleotide sequences is omitted because anyone skilled in the art can infer them.

(66) The link oligo-sequences in the fragments to be linked to each other are designed to overlap with each other, with no gaps between the fragments, so that a double-stranded deletion cassette DNA fragment, albeit having nicks thereon between the fragments, can be obtained only by overlapping the oligo-sequences. On the basis of the Santa Lucia Calculations, all of the oligo-sequences were designed to have a Tm value of 603 C. (Santa Lucia PNAS. 95:1460-5). Two relatively short oligo-sequences, 1216 bp long, were respectively arranged at the opposite terminal regions of the deletion cassette, producing blunt ends. Then, treatment with ligase sealed the nicks to give an intact deletion cassette DNA template.

Example 3

Transformation of Deletion Cassette and Identification of Gene-Targeted Mutants

(67) Using the lithium acetate method, the deletion cassette prepared through Examples 1 and 2 was transformed into diploid fission yeast SP286 strains (ade6-M210/ade6-M216, ura4-D18/ura4-D18, leu1-32/leu1-32) (Moreno et al., Methods Enzymol. 194:795-823). The transformed strains were spread over YES agar plates and cultured at 30 C. for 34 days to form colonies. Colony PCR was performed to examine whether desired genes were targeted. A small amount of cells were picked from edges of grown colonies on agar plates and suspended in deionized water. A portion of the suspension was subjected to PCR with a pair of primers. As shown in FIG. 6, colony PCR was performed with a pair of CP5 and CPN1 or CPN10 for the 5 side of the deletion cassette inserted into the chromosome and with a pair of CP3 and CPC1 or CPC3 for the 3 side of the deletion cassette. CP5 and CP3 were located 100200 bp upstream and downstream of corresponding chromosome homologous regions, respectively. CPN1 and CPC1 were located within the KanMX4 gene, with a distance of 200 bp from 5 and 3 ends thereof, respectively, while CPN10 and CPC3 were also located within the KanMX4 gene, with a distance of 300 bp from 5 and 3 ends thereof, respectively. Accordingly the colony PCR products thus obtained had a length of 300400 bp plus the length of the chromosome homologous region, that is, 80450 bp, amounting to 5001,000 bp in total. Reference may be made to Korean Patent No. 10-0475645 for reaction conditions of colony PCR.

(68) The base sequences of CP5 and CP3 which are different depending on the genes are not described in detail because they are readily apparent to those skilled in the art. Base sequences of CPN1, CPN10, CPC1 and CPC3 are given, along with their sequence identification numbers, in Table 3, below.

(69) TABLE-US-00003 TABLE3 SEQ Oligo Base ID Names Sequences NOS. CPN1 5-CGTCTGTGAGGGGAGCGTTT-3 1 CPN10 5-GATGTGAGAACTGTATCCTAGCAAG-3 2 CPC1 5-TGATTTTGATGACGAGCGTAAT-3 3 CPC3 5-GGCTGGCCTGTTGAACAAGTCTGGA-3 4

Example 4

Determination of Base Sequences of Gene-Specific Barcodes and their PCR Amplification

(70) Gene-specific barcodes must be needed in order to systemically identify strains using a microarray GeneChip as will be described in Example 5, below. In this regard, two gene-specific barcodes, each 20 bp, were given to 5 and 3 sides of the deletion cassette, respectively (FIG. 1). According to the gene database of the Sanger Institute, 4,988 genes are present in the fission yeast. Hence, as shown in FIGS. 7 to 55, a total of 9,976 barcodes were assigned to the 4,988 genes, with two per gene. The barcodes were named in such a manner that the systematic names of corresponding genes were followed by the extension _UP or _DN. For example, barcodes specific for the gene SPAC1002.09c were named SPAC1002.09c_UP for 5 up-tag, identified by SEQ ID NO. 18, and SPAC1002.09c_DN for 3 down-tag, identified by SEQ ID NO. 19 (FIG. 7). Barcodes were thus artificial DNA sequences which were not present on the chromosome of the fission yeast and were designed to have a Tm value of 601 C. according to a computer algorithm.

Example 5

Screening of A Drug's Mode of Action Using a Library of Gene-Targeted Diploid Fission Yeast Mutants

(71) A modification of the Giaever's method for screening drug's modes of action in the budding yeast on the basis of haploinsufficiency was applied to the fission yeast in accordance with the present invention (FIG. 56) (Giaever et al., Nature Genetics 14:450-56, Pierce et al., Nature Protocols 11:2958-2974, Pierce et al., Nat Methods: 601-3).

(72) The analysis of a drug's modes of action by use of a GeneChip comprises the following seven steps: 1) pooling of strain libraries, 2) activation, chemical treatment and sampling of a frozen strain pool, 3) isolation of chromosomal DNA and PCR amplification of barcode labels, 4) Order of GeneChip from Affymetrix, 5) hybridization, 6) staining, washing and colorimetry, and 7) result analysis.

(73) 1) Pooling of Strain Libraries

(74) A pool in which the mutants were equally mixed was prepared in the manner illustrated in FIG. 57. Because it was not easy to handle a library of 4,884 gene-targeted heterozygous mutants at the same time, they were divided into unit packages of 96 mutants. In detail, 96 mutants were cultured at 30 C. for 3 days on YES agar plates containing G418 at a concentration of 200 mg/ml, followed by incubation at 30 C. for 3 days on 96-well type plates. The strains cultured on the 51 plates were scraped, and suspended in 5.5 ml of respective 2 YES broths. After the addition of glycerol to the concentration of 30%, the 51 cultures were aliquoted at a volume of 1 ml and stored at 80 C. All of the 51 1 ml-aliquots in cryogenic storage were pooled together to afford a library of the mutants which was then aliquoted at a volume of 1 ml and stored at 80 C. A total of 510.sup.7 cells were used so that at least 10,000 cells were assigned to one mutant strain per one round of screening a drug's modes of action.

(75) 2) Activation, Chemical Treatment and Sampling of Frozen Strain Pool

(76) Prior to chemical treatment, the strain pool stored at 80 C. was activated. After being thawed, an aliquot of the strain pool was initially inoculated into a YES broth at O.D.sub.600=0.2 and grown at 25 C. to O.D.sub.600=2.0. These activated strains were diluted to a density of O.D.sub.600=0.2 and treated with a chemical at a suitable concentration, followed by sampling therefrom once every three sub-cultures over a total of 2030 subcultures (typically 4 hours are required for one sub-culture of the fission yeast). Since it was easy to determine the inhibitory activity of the chemical on the growth of cells when the strains must grow at a constant rate without the influence of nutrients, the sampled strains were diluted again at a density of O.D.sub.600=0.2 in fresh broth containing the chemical. A sample taken from chemical-free broth was used as a control. The samples were centrifuged to give cell pellets which were then stored at 80 C. Genomic DNA was isolated from all of the samples at the same time and subjected to GeneChip analysis.

(77) In order to determine the concentration of the chemical for the treatment, 50% inhibitory concentration (IC.sub.50) was first measured. Wild-type SP286 strain was grown to O.D.sub.600=2.0 in YES broth and inoculated with from 5- to 10-fold serial dilutions of the chemical to O.D.sub.600=0.1, followed by incubation at 30 C. for 24 hours. At each concentration of the chemical, the cultures were measured for final O.D.sub.600 values which were then plotted against Log [Chemical Concentration]. On the basis of these plots, IC.sub.50 values were automatically calculated using Prism Software (version 3.0, GraphPad Software Inc., San Diego, Calif.). Scientific experience made it possible to determine the concentrations of the chemical at around the appropriate IC.sub.50 values.

(78) 3) Isolation of chromosomal DNA and PCR Amplification of Barcode Label

(79) Using a Fungal/Bacterial DNA kit (Zymo Research, catalog # D6005), 2-4 g of genomic DNA was isolated from 200 mg of the cell sample. A 20-50-fold dilution of the isolated genomic DNA was mixed with a 200-fold dilution of Pico Green dye (Pico Green dsDNA Assay Kit (Invitrogen Inc. cat # P11495) at a volume ratio of 1:1, followed by quantification with NanoDrop (ND-1000, NanoDrop Inc.). 200 ng of each sample was used for labeling PCR.

(80) For use as probes in GeneChip, only the barcodes present in the isolated genomic DNAs were amplified by PCR using biotin-labeled primers. To this end, first, base sequences of common primers for the amplification of the 20-bp barcodes were determined. It was important to design the common primers which did not form non-specific bonds with the genome of the fission yeast. The base sequences determined through algorithm and serial experiments are given in Table 4, below. Biotin, serving as a label in the PCR amplification of the barcodes, was conjugated to the 5-ends of the primers which are directed towards the kanamycin resistance gene. The PCR products of the 5 and the 3 barcode were 70 bp and 73 bp in length. Solutions and conditions for PCR amplification are summarized in Table 5, below. After completion of PCR, PCR products were analyzed for length and amount by the electrophoresis of 5 l of the PCR reaction in 3% agarose gel and stored at 4 C. in a refrigerator until hybridization. About 30 l of the PCR product was used for each hybridization.

(81) TABLE-US-00004 TABLE4 SEQ Oligo Direc- Base ID Names Positions tions Sequences NOS. U1 5 Up- Sense 5-GCTCCCGCCTTACTTCGCAT-3 5 U2 tag Anti- 5-Biotin-CGGGGACGAGGCAAGC 6 sense TAA-3 D1 3 Down- Sense 5-Biotin-GCCGCCATCCAGTGTC 7 tag G-3 D2 Anti- 5-TTGCGTTGCGTAGGGGGG-3 8 sense

(82) TABLE-US-00005 TABLE 5 Sol'n Stock Final Conc./Amount Volumes (l) Water (H.sub.2O) 75-x PCR Buffer 10X 1X 10 MgCl.sub.2 20 mM 2.5 mM 10 dNTP 10 mM 0.2 mM 2 Primer Mix 50 uM 1 M 2 Taq 5 U/l 5 U 1 Polymerase Genomic ~200 ng x DNA Final Vol. 100 PCR 94 C., 2 min, 1 cycle - (94 C., 30 s; 58 C., 30 s; Conditions 72 C., 30 s) 29 cycles - 72 C., 3 min, 1 cycle

(83) 4) Order of GeneChip from Affymetrix

(84) A made-to-order GeneChip with serial number KRIBBSP1-a520429 from Affymetrix was used to recognize the 20-bp barcodes specifically inserted into the mutant strains (FIG. 58). On this DNA chip with a size of 0.80.8 cm were planted 100,000 probes which each have a feature of 1111 m. The probes consisted of 1) perfect match (PM) and mismatch (due to mismatch bases assigned to middle sites of barcodes) with 5 and 3 barcodes (each 10,000) in triplicate (60,000 in total); 2) 10,000 control probes for minimizing false positive signals induced by non-specific hybridization; 3) 20,000 spare probes for barcodes; and 4) 10,000 basic structure probes necessary for indicating the manufacturer Affymetrix and Oligo texts and compartmenting.

(85) 5) Hybridization

(86) Hybridization between the GeneChip from Affymetrix and the probes obtained by PCR in 3) was conducted as follows. As shown in Table 6, first, the hybridization solution comprising a biotin-labeled text oligo-mix and a blocking oligo-mix in addition to the biotin-labeled barcode probes was mixed in a 1.5 ml tube. The biotin-labeled text oligo-mix gave a standard to the optical recognition system upon analysis on the GeneChip. The blocking oligo-mix composed of 8 oligonucleotides served to block the common primers within the barcode and the gap between the primers and the barcode to expose only the 20-bp barcodes in single strands, thereby allowing the sense single strands to hybridize with the antisense probe oligonucleotides of the GeneChip. Each of the 8 oligonucleotides was used as a 300 pM stock. In the blocking oligo-mix, 5U-Block, K5U-Block, 5U-Block (rev comp) and K5U-Block (rev comp) were each present at a concentration of 37.5 pmole/l while 3U-Block, K3U-Block, 3U-Block (rev comp) and K3U-Block (rev comp) were each present at a concentration of 12.5 pmole/l. Immediately after being boiled at 98 C. for 5 min, the hybridization solution was quenched in ice water to expose the 20-bp sense barcode sequences in single strands. To the GeneChip which was pre-treated for 5 min with 140 l of hybridization buffer was injected 140 l of the hybridization solution, followed by hybridization at 42 C. for 16 hours.

(87) TABLE-US-00006 TABLE6 SEQ CompositionsandBase Vol ID Solutions Sequences (l) NOS. 5 Up-and 4,884Gene-specific 60 3 Down-tag UpandDownbarcodes (each 30) 2 Hybridization 200mMMES,2MNa+,40mM 75 Buffer EDTA0.02%Tween20 TextOligo 5-biotin- 0.5 9 (20fmole/l) GTCGTCAAGATGCTACCGTTCAGGA-3 Blocking 5U-Block 5-CGCTCCCGCCTTACTTCGCATTTAAA-3 12 10 Oligo- Mix K5U-Block 5-GGGGACGAGGCAAGCTAAGATATC-3 11 3U-Block 5-TTGCGTTGCGTAGGGGGGATTTTAAA-3 12 12 K3U-Block 5-CGCCATCCAGTGTCGAAAAGTATC-3 13 5U-Block 5-TTTAAATGCGAAGTAAGGCGGGAGCG-3 14 (revcomp) K5U-Block 5-GATATCTTAGCTTGCCTCGTCCCC-3 15 (revcomp) 3U-Block 5-TTTAAAATCCCCCCTACGCAACGCAA-3 16 (revcomp) K3U-Block 5-GATACTTTTCGACACTGGATGGCG-3 17 (revcomp) 50 Denhardt'sSol'n 1%Ficoll400,1%PVP,1%BSA 3 Totalvolume 150.5

(88) 6) Staining and Washing and Colorimetry

(89) After completion of the hybridization, the GeneChip was stained with phycoerythrin-conjugated streptavidin using a fluidics station (Affymetrix). Streptavidin was strongly bound to the biotin to give fluorescence, affording the quantification of the biotin-labeled barcodes hybridized with the probes on the GeneChip.

(90) The Fluidics station was used according to the protocol recommended by the manufacturer using staining solution (600 l: 20SSPE 180.57 l+50Denhart's solution 11.94 l+10% Tween20 0.597 l+phycoerythrin-conjugated streptavidin 1.019 l+deionized water 405.874 l) and a washing solution (Wash A: 20SSPE 300 ml, 10% Tween 1 ml, distilled water 699 ml in 1 liter; Wash B: 20SSPE 150 ml, 10% Tween 1 ml, distilled water 849 ml in 1 liter). Thereafter, the GeneChip was scanned with GeneChip Scanner 3000 G7 (Affymetrix) to detect fluorescence.

(91) TABLE-US-00007 TABLE 7 Conditions No. of Round Notes Wash A1 Recovery Mixes 0 Wash A1 Temperature ( C.) 25 Number of Wash A1 Cycle 2 Mixes per Wash A1 Cycle 4 Wash B Recovery Mixes 0 Wash B Temperature ( C.) 42 Number of Wash B Cycle 6 Mixes per Wash B Cycle 4 Stain Temperature ( C.) 25 First Stain Time (seconds) 0 (dummy step) Wash A2 Recovery Mixes 0 (dummy step) Wash A2 Temperature ( C.) 25 Number of Wash A2 Cycle 1 Mixes per Wash A2 Cycle 2 Second Stain Time (seconds) 600 Third Stain Time (seconds) 0 (dummy step) Wash A3 Recovery Mixes 0 (dummy step) Wash A3 Temperature ( C.) 25 Number of Wash A3 Cycles 6 Mixes per Wash A3 Cycle 4 Holding Temperature ( C.) 25

(92) 7) Result Analysis

(93) The binding between the PCR-amplified gene-specific sense barcodes and the antisense oligonucleotide probes planted on the GeneChip could be quantified by the fluorescent intensity detected. The fluorescence values measured were stored in files with the extension name cel. Strong fluorescent intensity means that cells with corresponding genes are predominantly growing. On the other hand, weak fluorescent intensity means the inhibition of cell growth. Based on the ANOVA/ANCOVA model, fluorescent data from the GeneChip were analyzed for the heterozygous gene-targeted mutants which were inhibited from growing.

(94) According to the ANCOVA model, the drug-induced inhibition of cell growth is expressed as the reduced passage number which can be obtained according to the following formula.
Reduced passage No. by drug in unit time (12 hrs)=Changed passage No. upon drug treatment in unit time (12 hrs)Changed passage No. in the absence of drug in unit time (12 hrs)

(95) Passage No. is calculated according to the following formula:
Passage No.=Log.sub.2(Fluorescent Intensity of Gene X at a predetermined time)Log.sub.2(Fluorescent Intensity of Gene X at the previous time)

Example 6

Screening of Terbinafine's Modes of Action Using GeneChip and Verification

(96) In order to screen terbinafine's modes of action with the GeneChip, first, the treatment concentrations of drug were determined by the measurements of IC.sub.50 obtained in the above-mentioned method. In this regard, relative rates of drug-induced inhibition against cell growth were obtained by treating the cells with terbinafine at doses of 0.1 nM, 1 nM, 5 nM, 10 nM, 50 nM, 100 nM, 500 nM and 1 nM. Analysis with Prism software determined IC.sub.50=40 nM (FIG. 11).

(97) After the treatment of a strain pool with 40 nM of terbinafine, as shown in Table 8, samples were taken every three passages (approximately 12 hours) till 3035 passages, GeneChip analysis was performed with 13 controls and 12 drug-treated samples.

(98) TABLE-US-00008 TABLE 8 Accumulated Chip Specimen Specimen Passages Passage Nos. Group Names (Round) Time(h) 1 Chip control: Strain mix in Glycerol 2 Cell activation Activation Stage 1 6.9 3 Activation Stage 2 6.7 4 Activation Stage 3 4.2 5 Drug control: Control Data Point 1 3.1 3.6 6 YES media Control Data Point 2 6.6 3.9 7 Control Data Point 3 9.4 3.6 8 Control Data Point 4 12.8 3.8 9 Control Data Point 5 16.7 3.7 10 Control Data Point 6 21.2 3.6 11 Control Data Point 7 25.6 3.5 12 Control Data Point 8 32.7 3.4 13 Control Data Point 9 35.9 3.8 14 Drug-Treated: Drug Data Point 1 3.1 3.6 15 YES + 40 nM Drug Data Point 2 6.3 4.3 16 Terbinafine Drug Data Point 3 8.6 4.3 17 Drug Data Point 4 10.9 5.6 18 Drug Data Point 5 13.7 5.2 19 Drug Data Point 6 17.1 4.7 20 Drug Data Point 7 20.9 4.0 21 Drug Data Point 8 23.8 4.0 22 Drug Data Point 9 26.7 4.5 23 Drug Data Point 10 29.5 4.1 24 Drug Data Point 11 32.3 4.3 25 Drug Data Point 12 34.9 4.4

(99) The gene candidates which were ranked as the top 10 (or top 20) by ANCOVA analysis, defined as cell growth over time after treatment with terbinafine, are summarized in Table 9, below. The erg1 gene, known as a terbinafine's mode of action, codes for squalene monooxygenase, which plays an important role in the biosynthesis of the membrane component ergosterol. The screening of terbinafine's modes of action with the conventional budding yeast also resulted in a first ranking for the erg1 gene. This coincidence demonstrates that the system of screening drug's modes of action with a library of the gene-targeted fission yeast mutants in accordance with the present invention works properly.

(100) Since the drug's modes of action detected with the GeneChip might include the false positive signals resulting experiment errors and false chip analysis, all the top 10 drug's modes of action, except non-specific ribosome-related genes, were cultured again on YES agar plates containing terbinafine. As shown in FIG. 60, the same results as in the chip were obtained for the genes erg1 and pmm1 whereas there was a difference from the chip result on the smb1 gene. Therefore, the genes erg1 and pmm1 within the red boxes were verified as being targets of terbinafine.

(101) Taken together, the data obtained in the examples demonstrate that the present invention is very useful in accurately searching for novel drug's modes of action as well as already known target proteins.

(102) TABLE-US-00009 TABLE 9 Systematic Target Nos. Gene Names Genes Description of Genes 1 SPBC713.12 erg1 squalene monooxygenase Erg1 (predicted); similar to S. cerevisiae YGR175C 2 SPCC622.06c dubious; similar to S. pombe SPCC622.03c; tandem duplication; ORF in compositionally biased region 3 SPAC24H6.07 rps901 40S ribosomal protein S9; similar to S. cerevisiae YPL081W and YBR189W; similar to S. pombe rps902 4 SPBP23A10.07 rpa2 DNA-directed RNA polymerase I complex subunit Rpa2; similar to S. cerevisiae YPR010C 5 SPAC26A3.08 smb1 small nuclear ribonucleoprotein (snRNP) (subunit B); complexed with Cdc5p (PMID 11884590); similar to S. cerevisiae YER029C 6 SPBC17G9.07 rps2402 40S ribosomal protein S24; similar to S. cerevisiae YER074W and YIL069C 7 SPAC1F7.13c rpl801 60S ribosomal protein L2A; similar to S. cerevisiae YFR031C-A and YIL018W 8 SPBC12D12.03 cct1 chaperonin-containing T-complex alpha subunit Cct1; similar to S. cerevisiae YDR212W 9 SPCC1223.05c rpl3702 60S ribosomal protein L37; similar to S. cerevisiae YLR185W and YDR500C 10 SPAC1556.07 pmm1 phosphomannomutase; similar to S. cerevisiae YFL045C

INDUSTRIAL APPLICABILITY

(103) As described hitherto, the present invention provides a deletion cassette with two 20-mer gene-specific barcodes and two 250-350-bp homologous regions at both ends thereof, which is very useful in constructing gene-targeted yeast mutants. A library of the heterozygous gene-targeted yeast mutants can be used to systematically screen a drug's modes of action on a genomic level, thus affording the effective development of new drugs.

(104) Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims.