JP2003516716A - Characterization of the yeast transcriptome - Google Patents

Abstract

(57) Abstract A yeast gene whose expression changes during the cell cycle is described. Such genes can be used to study and monitor the cell cycle of eukaryotic cells and to influence them. It can be used to obtain human homologs involved in cell cycle regulation. It can be used to identify antifungals and other drug classes. Arrays on solid supports can also be formed to examine the transcriptome of cells under various conditions.

Description

Detailed Description of the Invention

This application is a continuation-in-part application of co-pending patent application No. 09 / 012,031 filed January 22, 1998, the disclosure of which is incorporated herein by reference.
This invention received governmental support from CA57345 awarded by the National Institutes of Health. The government has certain rights in this invention.

[0002] The present invention relates to the analysis of genes expressed in the yeast genome. More specifically,
The present invention relates to the identification and use of previously unrecognized genes.

[0003] Phenotype of background organisms present invention, it is now obvious that it is determined by the genes expressed predominantly therein. These expressed genes can be represented by a "transcriptome" which indicates the identity and expression level of each expressed gene for a given cell population. Unlike the genome, which is an essentially static entity, the transcriptome can be regulated by external and internal factors. The transcriptome is thus a dynamic link between an organism's genome and its physical characteristics.

The transcriptome described above has not been analyzed in any eukaryotic or prokaryotic organisms, mainly due to technical limitations. However, some common features of gene expression patterns were analyzed 20 years ago by RNA-DNA hybridization measurements (Bishop et al., 1974; Hereford and Rosbash, 1977). Therefore, it was found that many organisms identified at least three transcript classes with high, moderate, or low expression levels, and the number of transcripts per cell was estimated (Lewin, 1980). Of course, these data gave little information about the specific genes belonging to each class. As new genes are discovered, data on expression levels of individual genes have been accumulated. However, the absolute amount of expression of a particular gene was measured and compared to other genes in cells of the same species,
Only in very few cases.

The description of any cellular transcriptome will provide new information useful in understanding many aspects of cell biology and biochemistry.

[0006] SUMMARY One object of the present invention the invention is to provide a method of identifying the isolated DNA molecules, and a candidate agent affects the cell cycle using such molecules. These and other objects of the invention can be accomplished by providing the art with one or more of the embodiments described below.

In one aspect of the invention, isolated DNA molecules are provided. This is SEQ ID NO: 67-81
It contains a coding sequence for a yeast gene selected from the group consisting of NORF genes containing a SAGE tag as shown in 1.

In another aspect of the invention, methods of using the NORF gene are provided. This method is for affecting the cell cycle of cells. This method provides at least 1 between any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M.
Administering to the cell an isolated DNA molecule containing the coding sequence of the NORF gene with 0% altered expression.

In yet another aspect of the invention, a method of screening antifungal drug candidates is provided. This method alters expression by at least 10% between contacting a yeast cell with a test substance and any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. Test substances that regulate the expression of the yeast gene, including the step of monitoring the expression of the NORF gene, are candidates for antifungal agents.

In yet another aspect of the invention, there is provided a method of identifying a human gene involved in cell cycle progression. This method comprises at least 14 contiguous nucleotides of the NORF gene whose expression changes by at least 10% between any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. Contacting the containing probe with human DNA. Human DNA sequences that hybridize to the probe are identified as candidate sequences for human genes involved in cell cycle progression.

The present invention provides a probe comprising at least 14 contiguous nucleotides of the NORF gene containing a SAGE tag as set forth in SEQ ID NOS: 67-811.

The present invention also provides a probe array on a solid support. At least one probe in the array comprises at least 14 contiguous nucleotides of the NORF gene containing a SAGE tag as shown in SEQ ID NOs: 67-811.

[0013] Yet another aspect of the invention is a method of identifying a candidate drug as a member of a class of drugs that has a characteristic effect on gene expression in yeast cells. Contact the yeast cell with the candidate agent. At least one whose expression is affected by the class of drug
Expression of the two NORF genes is monitored in yeast cells. A candidate drug is identified as a member of the drug class if a difference is detected in the expression of at least one NORF gene relative to the expression in the absence of the drug candidate.

These aspects of the invention and other aspects that will be apparent to those of skill in the art upon reading the detailed disclosure provided below, are of previously unrecognized genes, and global at the organism level. It provides the art with information on specific gene expression.

Table Legend Table 1, Highly Expressed Genes. The tag indicates a 10-bp SAGE tag adjacent to the NlaIII site. A gene is one or more genes that correspond to a particular tag (many genes that match a unique tag are from related families with an average identity of 93%)
Indicates. The locus and description mean the locus name and functional description of each ORF, respectively. Copy number / cell indicates the amount of each transcript contained in the SAGE library, estimated at 15,000 total transcripts and 60,633 confirmed transcripts per cell.

[0016]   Table 2, Expression of putative coding sequences. The top of the columns of the table is the same as in Table 1.

Table 3, Maximum amount of NORF gene expression. SAGE tag, locus, copy number / cell,
It is the same as in Table 1. Chromosomes and tag positions mean the positions of the chromosomes and their respective tags. The ORF size indicates the size of the ORF corresponding to the designated tag. In each case, the tag was located at or below the 250 bp 3'end of NORF.

Table 4, NORF gene expression. The SAGE tag and copy number / cell are the same as in Table 1.
The chromosome and tag position indicate the position of the chromosome and each tag.

Table 5, gene expression varies with different cell cycle phases. L indicates logarithmic phase; S indicates synthetic phase; G2 / M indicates mitotic phase; tag sequence indicates 10 bp SAGE tag adjacent to NlaIII site. By "L to S ratio" is meant the ratio of logarithmic phase expression to synthetic phase expression. "Ratio of S to G2 / M" means the ratio of expression in the synthetic phase to expression in the G2 / M phase. "G2 / M ratio to L" means
It means the ratio of G2 / M phase expression to log phase. # DIV / 01! Indicates a degree of increase in expression from 0; a value of 0 indicates a decrease in expression at 0; a value of 1 indicates no change; a value smaller than 1 indicates decrease in expression A value greater than 1 indicates increased expression.

Table 6, Intergenic open reading frames containing or flanking the observed SAGE tags. The copy number / cell indicates the amount of each mRNA transcript, as in Table 1. Positive expression levels indicate that the tag is on the + strand of the chromosome.
A negative expression level indicates that the tag is on the-strand.

[0021] present in a particular gene (NORF) yeast was not known up to the detailed description which, to express it is the discovery of the present invention. These genes, like other previously identified or previously deduced genes, can be used to study, monitor, and affect the phase of the cell cycle. The present invention confirms that genes are differentially expressed during the cell cycle. Differentially expressed genes can be used as markers of cell cycle phase. They can also be used to influence cell cycle phase changes. Furthermore, they can be used to screen drugs that affect the cell cycle by affecting the expression of genes. It is also assumed that human homologues of these eukaryotic genes are present and can be identified using yeast genes as probes or primers for identifying human homologues.

A new gene named NORF (not the previously assigned open reading frame) was found. They are uniquely identified by their SAGE tag. Furthermore, the complete nucleotide sequences of each are known and publicly available. In general, they have not been previously identified as genes because of their small size. However, they are now found to be expressed.

Differentially expressed yeast genes are expressed in different growth phases, especially log phase, S phase, and G2 / M.
It is a gene whose expression changes with a statistically significant difference (confidence level of 95% or more) in the period. Preferably the difference is at least 10%, 25%, 50% or 100%. In some cases, differentially expressed genes do not express detectable levels during one or more of the cell cycles, as determined by SAGE analysis. NORF numbers are assigned to genes found to have differential expression characteristics. 1, 2, 4, 5, 6, 17,
25, 27, TEF1 / TEF2, ENO2, ADH1, ADH2, PGK1, CUP1A / CUP1B, PYK1, YKL056C, Y
MR116C, YEL033W, YOR182C, YCR013C, ribonucleotide reductase 2 and 4,
And YJR085C. Differential expression can be detected by any means known in the art, such as hybridization to a particular probe or immunoassay.

An isolated DNA molecule according to the present invention may be smaller than a complete chromosome and lack a genome, or cDNA, ie an intron. The isolated DNA molecule comprises a yeast gene, ie, a SAGE tag as set forth in SEQ ID NOS: 67-811, eg, NORF.
It contains the coding sequences for yeast genes involved in cell cycle progression, such as genes. An isolated DNA molecule comprising a yeast gene or a coding sequence for a yeast gene containing a SAGE tag as set forth in SEQ ID NOs: 37-12,203 is also an isolated DNA molecule of the present invention. Isolation
The DNA molecule may also include a yeast gene, ie, the coding sequence for a yeast gene, which includes a SAGE tag as set forth in SEQ ID NOs: 37-12, 203 or 67-811.

Methods for obtaining DNA of known sequence may be used to obtain the isolated DNA molecule of the present invention. Preferably, they are isolated free of other cellular components such as proteins and lipids that are membrane components. They can be made by cells and isolated or synthesized using PCR or automated synthesizers. Methods of purifying and isolating DNA are routine and well known in the art.

Non-limiting methods of DNA delivery known in the art can be used to administer the yeast gene to the cells. These include liposomes, transfection, matching, transduction, transformation, viral infection, electroporation. Vectors with specific purposes and characteristics can be selected by those of skill in the art for known characteristics. Cells that can be used as gene recipients are yeast and other fungi, mammalian cells, including humans, and bacterial cells.

Antibacterial drugs can be identified using yeast cells as described herein. Expression of the differentially expressed NORF gene can be monitored by means known in the art. If the test substance alters the expression of such a differentially expressed gene, for example by increasing or decreasing its expression, it is a candidate drug which can influence the growth of the fungus and also an antibacterial agent. May be useful as The expression of one or more NORF genes may be monitored. For example, 2, 3, 4, 5, 10, 15
, 20, 30, 40, 50, 60, 75, 100, 150, 250, 300, 350, 400, 450, or 500
Expression of one or more NORF genes may be monitored in one or multiple assays.

These genes may be common between species, as differentially expressed genes may be involved in cell cycle progression. This differentially expressed NORF gene identified by the present invention is a homolog of humans and other mammals, DNA from these mammals, and a probe containing at least 10 consecutive nucleotides of the differentially expressed NORF gene. Can be used to identify by contacting This DNA may be genomic or cDNA as is known in the art. Methods for identifying genes homologous between different species are well known in the art. Briefly, the stringency of hybridization will be reduced due to imperfect sequence matching hybridization. This is especially true for Southern blots, Northern blots, colony hybridizations,
Or it can happen in case of PCR. Any hybridization method known in the art can be utilized. The DNA sequence that hybridizes to the probe is
Identified as the sequence of a candidate gene associated with cell cycle expression.

The probe according to the present invention comprises at least 10, preferably at least 12, 14, 16, 18, 20, or 25 of a particular NORF gene or other differentially expressed genes.
An isolated DNA molecule with a contiguous number of nucleotides. The probe may or may not be labeled. They can also be used, for example, as primers for PCR assays or for detection of gene expression by Southern or Northern blots or in situ hybridization. Preferably the probe is immobilized on a solid support.
The solid support can be any surface to which a probe can be attached. Suitable solid supports include, but are not limited to, glass or plastic slides, tissue culture plates, microtiter wells, tubes, or beads such as, but not limited to, latex, polystyrene, or glass beads. Such particles are included. The probe can be attached to the solid support by any method known in the art, including covalent or non-covalent binding, passive absorption, or binding to the probe and solid support, respectively. Includes the use of site pairs.

More preferably, the probes are present in an array so that multiple probes can hybridize to a biological sample simultaneously. The probe can be attached to the array or synthesized in situ on the array. Lo
ckhart et al., Nature Biotechnology, Volume 14, 1996, December "Expression monitoring by hybridization to high-density oligonucleotide arrays (Expression mo
nitoring by hybridization to high-density ologonucleotide arrays) ". An array will contain at least one NORF probe, but will contain 100, 500, or even 1,000 or more different probes in distinct locations. If desired, one or more NORFs present on this array are
It can be a nucleotide sequence from the NORF gene that is differentially expressed during the cell cycle.

The genes identified by the present invention are differentially expressed during the cell cycle and can also be used to obtain a gene expression profile characteristic of the yeast gene's response of yeast cells to a particular drug or class of drugs. it can. A particular class of drugs of interest for generating gene expression profiles include drugs that affect the cell cycle or other cellular processes such as chemotherapeutic agents. If desired, complex gene expression profiles can be generated by obtaining and using gene expression profiles characteristic of one or more drugs of a particular class. For example, microtubule toxic drugs such as vinblastine, taxol, vincristine, and taxotere can be used to obtain gene expression profiles characteristic of microtubule toxic drugs.

To obtain a gene expression profile characteristic of a particular drug or class of drugs, yeast cells are contacted with the particular drug or one of the classes of drug. Expression of at least one yeast gene is monitored either before or after contacting, or in the contacted cells and in another yeast cell that has not been contacted with the drug. The genes to be monitored are any yeast genes, including NORFS.
Preferably these genes are differentially expressed during the cell cycle. For example, the yeast gene has S as shown in Tables 3, 4, 5 and 6 (SEQ ID NOs: 67-12, 203).
It can be selected from genes containing AGE tags. NORF number, if desired 1, 2, 4
, 5, 6, 17, 25, or 27, TEF1 / TEF2, ENO2, ADH1, ADH2, PGK1, CUP1A / CUP
Genes such as 1B, PYK1, YKL056C, YMR116C, YEL033W, YOR182C, YCR013C, ribonucleotide reductases 2 and 4, and YJR085C are available to monitor changes in gene expression.

These genes, for example 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 40, 50, 6
Any number of expressions can be measured, such as 0, 75, 100, 150, 250, 500, 1000, 2000, 3000, 4000, 5000, or 5,500 genes. It is particularly convenient to monitor the expression of differentially expressed genes using nucleic acids immobilized on a solid support or on an array such as the gene arrays described above.

Many genes, particularly cell cycle genes, may be common between yeast and mammals, including humans. Thus, by using a gene expression profile characteristic of a drug or drug class, identifying a candidate drug as a member of a drug class with a known characteristic gene expression profile can determine the effect of that drug on human cells. Can be predicted. The candidate drug may be an agent already known in the art, or a compound previously unknown to possess any pharmaceutical activity. The candidate drug may be naturally-occurring or may be laboratory-designed. They can be isolated from microorganisms, animals, or plants and can be recombinantly produced or synthesized by chemical methods known in the art.

The effect of a candidate drug on the expression of at least one gene whose expression is affected by that class of drug is monitored. A gene expression profile obtained with a candidate drug that is similar to the gene expression profile for a particular drug or drug class identifies the candidate drug as a member of that class of drugs.

The effect of altering a particular substituent of a known drug, or the effect of altering a particular substituent of a candidate drug may be similarly tested. Such methods may alter, for example, the solubility or absorption of certain drugs to have unintended and potentially harmful effects on genes that are differentially expressed during the cell cycle. Used to determine if they tend to be.

The above disclosure generally describes the present invention. A more complete understanding can be obtained by reference to the following specific examples, which are described herein solely for purposes of illustration and are not intended to limit the scope of the invention.

[0038] Example Summary We refer to the set of genes in the present specification is expressed from the genome of the yeast called transcriptome (transcriptome), series analysis of gene expression (ser
ial analysis of gene expression: SAGE). Analysis of 60,633 transcripts revealed 4,665 genes with expression levels ranging from 0.3 to over 200 transcripts per cell. Of these genes, 1,981 have known functions, whereas 2,684 have not been previously characterized. The integration of location information with gene expression data allowed the creation of chromosomal expression maps, the identification of physical regions of transcriptional activity, and the identification of genes that were not predicted solely by sequence information. These studies provide insights into the global patterns of gene expression in yeast and demonstrate the potential for extensive genomic expression studies in eukaryotic cells.

Results Features and Principles of the SAGE Approach Several methods for high-throughput evaluation of gene expression have recently been described (Nguyen et al., 1995; Schena et al., 1995; Velculescu et al., 1995). We used SAGE (Sequential Analysis of Gene Expression) because SAGE can provide quantitative gene expression data without the need for a hybridization probe for each transcript in advance. The SAGE technique is based on two basic principles (Fig. 1). First, the short sequence tag (9-11 bp) contains sufficient information to uniquely identify a transcript, provided that it is derived from the defined position of the transcript. Secondly, many transcript tags are linked to one molecule in a chain and then sequenced to simultaneously reveal the identities of multiple tags. The expression pattern of transcripts in a population can be evaluated quantitatively by measuring the amount of individual tags and identifying the gene corresponding to each tag.

Whole Genome Expression In order to maximize the presentation of genes associated with normal growth and cell cycle progression,
The SAGE library was generated from three-stage yeast cells: log phase, resting S phase, and resting G2 / M phase yeast cells. Overall supports 60,633 total transcripts
SAGE tags were identified (20,184 from log phase, 20,034 from resting S phase,
And it contains 20,415 cells from resting G2 / M phase cells). Of these tags,
56,291 tags (93%) matched the yeast genome exactly, 88 tags matched the mitochondrial genome, and 91 tags matched the 2 micron plasmid.

The number of SAGE tags required to define the yeast transcriptome is
It relies on the level of confidence required to detect low abundance mRNA molecules. A previously derived estimate of 15,000 mRNA molecules per cell (Hereford and Rosbash, 1
977), with 20,000 tags, m in a single copy per cell.
It has been shown to have a 1.3-fold coverage for the presence of RNA molecules, which could provide a 72% probability of determining such transcripts (as determined by Monte Carlo simulations). Analysis of 20,184 tags obtained from log phase cells confirmed 3,298 unique genes. Having independently confirmed the mRNA copy number per cell, the present inventors have found that among several genes whose absolute mRNA levels are easily detected by quantitative hybridization experiments (Iyer and Struhl, 1996). The expression level of SUP44 / RPS4, one of which was compared with the expression level measured by SAGE. SUP44 / RPS44 is 7 by hybridization
5 +/- 10 copies / cell (Iyer and Struhl, 1996), which was SA
It is in good agreement with the 63 copies / cell of the GE data, suggesting that the estimate of 15,000 mRNA molecules per cell is fairly accurate. Analysis of the SAGE tags obtained from S-phase of resting cells and G2 / M-phase of resting cells showed that the expression level of this gene (52 to 55 copies / cell) was the same, which The expression level was similar to a large number of expressing cells. Since less than 1% of the genes are expressed at dramatically different levels in these three steps (see below), SAGE tags from all libraries were used to combine to analyze a wide range of patterns of gene expression. did.

The number of unique transcripts was analyzed by ˜60,000 when the tags confirmed by the increased amount were analyzed.
A plateau was reached with individual tags (Fig. 2). This suggests that additional gene production with the additional SAGE tag yielded almost no additional gene, which was expressed in a small amount equivalent to 60,000 transcripts per cell. This is in agreement with the fact that it showed a 4-fold excess for the gene. Similarly, Monte Carlo simulation
Analysis of the 60,000 tags would be identical to at least one tag producing 97% transcripts when its expression level was one copy per cell.

The 56,921 tags, which exactly matched the yeast genome, displayed 4,665 copies of different genes. This number reflects the RNA-DNA reassociation kinetics (Hereford and Rosbas
h, (1977)) is consistent with the estimate of 3,000 to 4,000 expressed genes. These expressed genes included 85% of the genes with characteristic functions (1,981 out of 2,340) and 76% of all genes predicted from yeast genomic analysis (4,665 out of 6,121). . These numbers represent a relatively complete sampling of the yeast transcriptome, given the limited number of physiological stages examined and a very large number of genes that were only deduced based on genome sequence analysis. Is consistent with

Transcript expression per gene was observed varying between 0.3 and 200 or more copy numbers per cell. Analysis of the distribution of gene expression levels revealed several developmental classes similar to those observed in previous studies using reassociation kinetics.
"Virtual Rot" of genes observed by SAGE (Fig. 3
In A), three major components of the transcriptome were identified with abundance in the third or higher amplification range. Lot curves derived from RNA-cDNA reassociation kinetics also contained three major components distributed in a similar range of abundance (Herefo
rd and Rosbash, 1977). Although the kinetics of reassociation of a particular class of RNA and cDNA can be affected by a large number of experimental variables, there are striking similarities between lots and virtual lot analysis. (Fig. 3B). Since lot analysis cannot detect all low abundance transcripts (Lewin, 1980), SAGE has both a high total number of expressed genes and a higher fraction of the transcriptome belonging to the low abundance transcript class. It is not a surprise to show.

Integration of Expression Information Using Genome Map SAGE expression data can generate a chromosomal expression map by integrating existing position information (FIG. 4). These maps were created using the yeast genome sequence and the ORF position coordinates obtained from the Stanford Yeast Genome Database. Although there are several genes that are physically close and are noted to have similar high levels of expression, it is clear that there are clusters that are particularly high or low on either chromosome. I didn't. Histone H3- and H4-like genes are known to have different co-regulatory promoters and are directly adjacent to chromosome 14 (Smith and Murray, 1983).
, With very similar expression levels (5 and 6 copies per cell, respectively). The distribution of transcripts on this chromosome showed that the total transcript was evenly spread, leading to total transcript levels roughly linearly related to chromosome size (r ² = 0.85, data not shown). It is suggested that it has become. However, the region within the 10 kb telomere does not seem to be transcribed uniformly and contains 3.2 tags per gene on average, compared to 12.4 tags per gene in the non-telomeric region. (Fig. 4). This is in agreement with the previously described observation for "telomere silencing" in yeast (Gottschling et al., 1990). Recent studies reported the effect of a telomere position 4 kb away from the telomeric end (Renauld et al., 1993).

Gene Expression Patterns Table 1 lists the 30 most highly expressed genes, all of which are expressed at mRNA copy numbers of 60 or more per cell. As expected, these genes
Most corresponded to well-characterized enzymes involved in energy metabolism and protein synthesis, which were expressed at similar levels in all three growth stages (Example in Figure 5). Some of these genes include ENO2 (McAlister and Holland, 1982
), PDC1 (Schmitt et al., 1983), PGK1 (Chambers et al., 1989), PYK1 (Nishizawa et al., 1989), and ADH1 (Denis et al., 1983), which are glucose-enriched growth conditions used in this study. It is known to be drastically induced by. In contrast, GAL1
/ GAL7 / GAL10 cluster (St John and Davis, 1979) and GAL3 (Bajwa et al., 1988)
It was observed that glucose-repressive genes such as were expressed at very low levels (0.3 or less copy number per cell).

As expected for the yeast strains used in this study, the factor genes (MFA1, MFA2) (M
ichaelis and Herskovitz, 1988), and the alpha factor receptor (STE2) (Burkho
It was observed that all of the match-type specific genes (such as Lder and Hartwell, 1985) were expressed at a significant level (range of 2-10 copy number per cell), whereas the match-type alpha-specific gene was expressed. Genes (MFα1, MFα2, STE3) (Hagen et al.,
1986; Kujan and Herskowitz, 1982; Singh et al., 1983) at very low levels (<0.
.3 copy number / cell).

Three of the genes expressed at high levels in Table 1 were not previously characterized. One contains an ORF with the predicted liposomal function, which was previously identified only by genomic sequence analysis. Analyzes of all SAGE data correspond to uncharacterized ORFs transcribed at detectable levels 2
It suggested that there were 684 such genes. Of these transcripts, the one with the highest abundance of 30 corresponds to at least 8 transcripts per cell.
It was observed 0 or more times (Table 2). Genes not with features expressed other the two altitude corresponding to ORF that was not predicted by the analysis of the yeast genome sequence (NORF = ORF not annotated (N onannotated ORF)). Analysis of the SAGE data suggested that there were at least 160 NORF genes transcribed at detectable levels. A maximum of 30 of these transcripts was observed at least 9 times (
Example of Table 3 and FIG. 5).

Interestingly, one of the NORF genes (NORF5) was expressed only in cells that were quiescent in S phase, and its abundance was maximally different (> 49-fold, FIG. 5) among the three steps analyzed. Correspond. When S-phase arrested cells were compared to other stages, the RNR2 and RNR4 transcripts
A more than double increase was confirmed (Fig. 5). Induction of these ribonucleoside reductase genes may be due to the hydroxyurea treatment used to arrest cells in S phase (Elledge and Davis, 1989). Similarly, a comparison of cells arrested in the G2 / M phase confirmed an increase in RBL2 and dynein light chain, both microtubule-associated proteins (Archer et al., 1995; Dick et al., 1996). If RNR transduction occurs, these increased levels may be associated with the nocodazole treatment used to arrest cells at the G2 / M phase. Although there are many relatively small differences between this stage (eg NORF1, Figure 5), surprisingly few significant differences appeared in the comparison of all three stages. That is, at the three different stages analyzed, the yield was 10
Only 29 transcripts differed more than fold (Tables 4 and 5).

A comprehensive analysis of the NORF gene was performed using SAGE data. The intergenic region of the yeast genome was defined as the region outside the annotated ORF, i.e., the 500 bp region downstream of the annotated ORF (the yeast genome sequence and annotated ORF table is
Obtained from SGD at http://genome-www.stanford.edu/Saccharomyces/). Based on sequence analysis, the entire 9524 putative ORFs of 25-99 amino acids reside in the intergenic region, and 510 of these ORFs contain the observed SAGE tag, or Or adjacent to it (Table 6). Among the 60,633 analyzed SAGE tags, there were 302 unique SAGE tags located in or adjacent to the intergenic ORF (100 bp upstream or 500 bp downstream of the ORF) (Table 6). Note that in some cases, one or more NORFs contain or are adjacent to the SAGE tag. These genome-specific matched tags were in the correct location and were expressed at transcript copy number levels of 0.3 or more per cell.

The expression level for each NORF shown in Table 6 corresponds to the copy number of mRNA transcript per cell. If the expression level is positive, the tag is on the + strand of the chromosome, if negative, the tag is on the − strand of the chromosome.

Discussion Analyzing the yeast transcriptome provides a unique perspective on RNA components that define cell survival. Comparing gene expression patterns from altered physiological stages can identify genes that are important in various processes.
Comparing transcriptomes from various physiological stages, expression can provide the minimal set of genes required for normal growth and only if and in response to specific environmental stimuli, or Another set can be provided that consists of genes that are expressed only when different steps occur. For example, recent studies have identified a minimal set of 250 genes required for prokaryotic survival (Mushegian and Koo
Nin, 1996). Examination of the yeast genome readily identified homologous genes for 196 of these, with over 90% being observed to be expressed by SAGE analysis. Detailed analysis of the transcriptome from other organisms, as well as detailed analysis of the yeast transcriptome, ultimately yields the minimal set of genes required for eukaryotic cell survival It seems that

Like other whole-genome analysis methods, SAGE analysis of the yeast transcriptome has some potential limitations. First, a small number of transcripts were predicted to lack the NlaIII site and therefore would not be detectable by our analysis. Second, our analysis was limited to transcripts found at least as frequently as 0.3 copies per cell. Transcripts expressed in only insignificant fractions of the cell cycle, or transcripts expressed in only one fraction of the cell population, would not be readily detected by our analysis. Finally, mRNA sequence data is practically not available for yeast, and therefore some SAGE tags cannot unambiguously match the corresponding gene. Overlapping genes or abnormally long
It seems that tags derived from genes with 3'untranslated regions are misassigned.
Increased availability of the 3'UTR sequence in yeast mRNA molecules would help resolve the ambiguity.

Despite these potential limitations, the analytical methods described herein
It is clear that it provides both a comprehensive and localized depiction of gene expression precisely defined at the nucleotide level. These data, as well as the sequences of the yeast genome itself, provide simple and basic information essential for the interpretation of many future experiments. The potential for EST sequencing and mRNA sequence information from various genomic projects will eventually enable the clarification of transcriptomes from various organisms, including humans. The data recorded here are achieved using a tag, i.e. a small number of automated sequencers, for which a reasonably complete representation of the human cell transcriptome is only about 10-20 times more than evaluated here. It suggests that only a sufficient number of tags within the range of possible implementations will be needed. Analysis of global expression patterns in highly eukaryotic cells is generally expected to be similar to that reported herein for S. cerevisiae. However, analysis of transcriptomes in different cells, as well as transcriptomes from different humans, will yield a wealth of information related to gene function in normal, developmental, and disease states.

Experimental Steps Culture of Yeast Cells The origin of the transcripts for all experiments was S. cerevisiae strain YPH499 (MATa ura3
-52 lys2-801 ade2-101 leu2-Δ1 his3-Δ200 trp1-Δ63) (Sikorski and Hicter
, 1989). Cells growing logarithmically were enriched in YPD (Rose et al., 1990) rich medium.
(YPD supplemented with 6 mM uracil, 4.8 mM adenine, and 24 mM tryptophan)
Obtained by growing yeast cells in the early log phase (3 × 10 ⁶ cells / ml) at 30 ° C. in medium. Nocodazole (15 μg / ml) for arrest in the G1 / S phase of the cell cycle
Was added to early log phase cells and the culture was incubated at 30 ° C. for an additional 100 minutes. Harvested cells were washed once with water before freezing at -70 ° C. Microscopic and flow cytometric analysis of the growth stage of the harvested cells (Basrai et al., 199).
Confirmed by 6).

SAGE Protocol The SAGE method was performed as described above (Velculescu et al., 19), except as noted below.
95; Kinzler et al., US Pat. Nos. 5,866,330 and 5,695,937). Poly A RNA
Except that including biotin -5'-T ₁₈ -3 'primers were converted to double stranded cDNA by BRL synthesis kit using the manufacturer's protocol. This cDNA was cloned into NlaIII (Anchoring, E
nzyme). The NlaIII site contains three randomly selected yeast chromosomes (1,
It was predicted that 95% of the yeast transcripts would be detectable with the NlaIII-based SAGE approach, as each was observed to generate 309 base pairs once in 5, 10). After capture of the 3'cDNA fragment on streptavidin-coated magnetic beads (Dynal), the bound cDNA was split into two pools and one of the following linkers containing the recognition site for BsmFI was added to each pool. I ligated.

BsmFI (tagging enzyme) is cleaved at a distance of 14 bp from the recognition site, and the NlaIII site is 1 bp BsmFI.
A 15 bp SAGE tag was opened with BsmFI to overlap the mFI site. The SAGE tag overhang was filled with Klenow and the tags from the two pools were combined and ligated together. This ligation product was diluted and subsequently used as a primer PCR was performed for 28 cycles for amplification. The PCR products were analyzed by polyacrylamide gel electrophoresis (PAGE) and the PCR products containing two tail-to-tail linked tags (ditag) were removed. The PCR product was then digested with NlaIII, the band containing the ditag was excised and self-ligated. After ligation, the chained products were separated by PAGE and the product between 500 bp and 2 kb was removed. These products were cloned into the SphI site of pZero (Invitrogen). Colonies were screened for inserts by PCR using the M13 forward and M13 reverse sequences located outside the cloning site as primers.

PCR products obtained from the selected clones were TaqFS Dye Primer kit (Perkin
Elmer) and sequenced using a 377 ABI automated sequencer (Perkin Elmer). Each successive sequencing reaction identified an average of 26 tags, giving a sequencing success rate of 90%, corresponding to an average of about 850 tags per sequencing gel.

SAGE Data Analysis The sequence file was analyzed by the means of the SAGE program group (Velculescu et al., 1995) and it was identified that it was attached to the enzyme site with appropriate spacing, and two intervening tags were identified. Extract and record them in the database. 68
, 691 tags were obtained, including 62,965 tags obtained from unique ditags and 5,726 tags obtained from repeated ditags. The latter was counted only once to estimate potential PCR bias for quantification, as described (Velculescu et al., 1995). Of the 62,965 tags, 2,332 tags corresponded to the linker sequence and were excluded from further analysis. Of the remaining tags, 4,342 could not be assigned, likely due to sequencing errors (either in the tags or in the yeast genomic sequence). If all of these are due to tag sequencing errors, it is
Corresponding to a sequencing error rate of about 0.7% per bp tag), it is quite different from what would be expected under our automated sequencing conditions. However, some unassigned tags are much higher than the expected frequency of A as the last 5 base pairs of that tag (5 of the 52 maximal unassigned tags),
It suggests that these tags were derived from transcripts containing enzyme binding sites at several base pairs from the poly A tail. The frequency of NlaIII sites in the genome was obtained (one of 309 base pairs) and about 3% of the transcripts contained their poly A tail.
It was predicted to contain an NlaIII site within 10 bp.

The data available for yeast mRNA sequences are very sparse, and efforts to date have shown a highly common polyadenylation signal (Iniger and Braus, 1994; Z
Since we were unable to identify aret and Sherman, 1982), we used a 14 bp SAGE tag (ie NlaIII site plus adjacent 10 bp to directly probe the yeast genome).
) Was used (for the yeast genome sequence, see Stanford Yeast Genome ftp on August 7, 1996).
Site (genome-ftp Stamford edu). Only the coding region is annotated in the yeast genome, and the SAGE tag can be derived from the 3'untranslated region of the gene, so if the SAGE tag matches the ORF or the 500bp region at the 3'end of the ORF, the SAGE The tag was considered to correspond to a specific gene (locus name, gene name, and ORF chromosomal coordinates were obtained from the Stanford yeast genome ftp site, and the ORF description was recorded on August 14, 1996 at the MIPS www site ( http: //www.mips.bioc
hem.mpg de /)). ORFs were considered to be genes of known function if they were associated with a three letter gene name, whereas such unnamed ORFs were considered to be featureless.

As expected, the SAGE tag matched the transcribed portion of the genome in a very non-random manner, which was 88% matched to the ORF or the flanking 3'region in the correct position (X ² assay). P value is <10 ^-30 ). In fact, one or more tags may
When F was matched, the amount of matched tags was calculated by calculating the amount. Tags that matched the ORF at the wrong position were not used in the abundance calculation. In fact, if the tag matched more than one region of the genome (eg, ORF and non-ORF regions), only the matched ORFs were considered. One of its tags, if any
The fifth base can also be used to resolve the ambiguity.

In identifying the NORF gene, only those tags that matched a portion of the genome greater than 500 bp 3 ′ to the previously identified ORF were considered and were observed at least twice in the SAGE library.

[0063]

[Table 1] Genes expressed at high levels

[0064]

[Table 2] Putative coding sequence

[0065]

[Table 3] NORF gene

[0066]

[Table 4] Additional NORF genes

[0067]

Table 5 Changes in gene expression during different cell cycle phases

[0068]

[Table 6] Analysis of NORF in intergenic region

References

[Brief description of drawings]

FIG. 1 is a schematic diagram of SAGE method and genomic analysis. To define the unique position of each transcript when applying SAGE to the analysis of yeast gene expression patterns, and BsmFI
The 3'-most NlaIII to provide a ligation site for the site-bearing linker
The site was used. Then, the type IIs enzyme BsmFI, which cleaves at a certain distance from the non-palindromic recognition site, was used to create a 15-bp SAGE tag (indicated by a black arrow) containing an NlaIII site. Automated sequencing of the linked SAGE tags can routinely identify about 1000 tags per 36 lane sequencing gel. Once sequenced, the abundance of each SAGE tag was calculated and each tag was used to identify the corresponding gene from the entire yeast genome. The lower panel shows a small region of chromosome 15. Gray arrows indicate all potential SAGE tags (NlaIII site), black arrows are most 3 '.
Indicates the SAGE tag of. The total number of tags observed for each potential tag is (+
Chain) or below (-chain). As expected, the observed SAGE tag was associated with the 3'end of the expressed gene.

FIG. 2. Sampling of yeast gene expression. In analyzes that increase the amount of tags identified, the number of uniquely expressed genes indicates a plateau. Triangles represent genes having known functions, squares represent genes predicted based on sequence information, and circles represent total genes.

FIG. 3 Virtual lot. (A) Abundance in the yeast transcriptome.
The transcript abundances are plotted in reverse order on the abscissa, and at least the percentage of total transcripts with that abundance is plotted on the ordinate. The dotted line is the curve
Three components 1, 2, and 3 are shown. This is similar to a lot curve from reassociation kinetics, where the product of initial RNA concentration and time is plotted on the abscissa and the percentage of labeled cDNA that hybridizes to excess mRNA is plotted on the ordinate. is doing. (B) Comparison of virtual lots and lot components. The migration and data from the virtual lot component are calculated from the data in Figure 3A and the lot component data is He
Obtained from reford and Rosbash, 1977.

FIG. 4: Chromosomal expression map of S. cerevisiae. Individual yeast genes were mapped on each chromosome according to the open reading frame (ORF) start coordinates. The abundance of the tag corresponding to each gene is shown on the vertical axis, the transcription from the + strand is shown on the upper side of the abscissa, and the transcription from the − strand is shown on the lower side. The yellow band at the end of the expanded chromosome represents a poorly transcribed telomeric region (see text for details).

FIG. 5. Northern blot analysis of representative genes. TDH2 / 3, TEF1 / 2, and NORF1 are all three states (lane 1, G2 / M pause; lane 2, S phase pause; lane 3,
RNR4, RNR2, and NORF5 are highly expressed in cells arrested in S phase, although they are relatively equally expressed in log phase). Expression level (number of tags) observed with SAGE
Are listed below each lane and were highly correlated with the quantification of Northern blots by Phosphorimager analysis (r ² = 0.97).

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) C12N 5/10 C12N 15/00 ZNAA C12Q 1/02 F 1/68 5/00 B (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OA (BF, BJ, CF, CG, CI) , CM, GA, GN, GW, ML, MR, NE, SN, TD, TG), AP (GH, GM, KE, LS, MW, MZ, SD, SL, SZ, TZ, UG, ZW), EA (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), AE, AL, AM, AT, AU, AZ, BA, BB, BG, BR, BY, CA, CH, N, CR, CU, CZ, DE, DK, DM, EE, ES, FI, GB, GD, GE, GH, GM, HR, HU, ID, IL, IN, IS, JP, KE, KG, KP, KR, KZ, LC, LK, LR, LS, LT, LU, LV, MA, MD, MG, MK, MN, MW, MX, MZ, NO, NZ, PL, PT, RO, RU, SD, SE , SG, SI, SK, SL, TJ, TM, TR, TT, TZ, UA, UG, US, UZ, VN, YU, ZA, ZW (72) Inventor Kinsler Kenneth Belle Air Halkirk Way, Maryland, USA 1403 F term (reference) 4B024 AA20 CA04 CA09 DA12 HA14 4B029 AA07 FA12 4B063 QA01 QA13 QQ07 QQ42 QR32 QR55 QR76 QS34 4B065 AA58X AA72X AA72Y AA90X AB01 BA01 CA24

Claims (42)

[Claims] 1. An isolated DNA molecule comprising a coding sequence for a yeast gene selected from the group of NORF genes containing a SAGE tag as set forth in SEQ ID NOs: 67-811. 2. The isolated DNA molecule according to claim 1, which is involved in cell cycle progression. 3. The NORF gene expression is altered by at least 10% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. Isolated DNA molecule. 4. The NORF gene expression is altered by at least 25% between any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. Isolated DNA molecule. 5. The NORF gene expression is altered by at least 50% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. Isolated DNA molecule. 6. The expression of the NORF gene is changed by at least 100% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase and G2 / M. Isolated DNA molecule. 7. A statistically significant difference (95% or more confidence) in the expression of the NORF gene between any two cell cycle phases selected from the group consisting of logarithmic phase, S phase, and G2 / M. Level) indicating the isolated DNA molecule according to claim 2. 8. The isolated DNA molecule of claim 7, wherein the NORF gene is selected from the group consisting of NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27. 9. The isolated DNA according to claim 2, wherein the NORF gene is not expressed in at least one phase of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M.
molecule. 10. The isolated DNA molecule according to claim 1, which is a genome. 11. The isolated DNA molecule according to claim 1, which is a cDNA. 12. A method of using the NORF gene to affect the cell cycle, comprising the steps of: any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. DNA containing the coding sequence of the NORF gene whose expression changes by at least 10% between
Administering a molecule to a cell. 13. The method of claim 12, wherein the cells are yeast cells. 14. The method of claim 12, wherein the cells are fungal cells. 15. The method of claim 12, wherein the cell is a mammalian cell. 16. The NORF gene has NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27.
13. The method of claim 12, selected from the group consisting of: 17. A method for screening an antifungal drug candidate, which comprises the steps of: contacting a test substance with a yeast cell; of a cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. Monitoring the expression of the NORF gene whose expression changes by at least 10% between any two phases, wherein the test substance which changes the expression of said yeast gene is a candidate antifungal agent. 18. The NORF gene has NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27.
18. The method of claim 17, selected from the group consisting of: 19. A method for identifying a human gene involved in cell cycle progression, comprising the steps of: log phase, S phase, and any two phases of the cell cycle selected from the group consisting of G2 / M. The step of contacting a probe containing at least 10 consecutive nucleotides of the NORF gene whose expression changes by at least 10% with human DNA, the human DNA sequence hybridizing to the probe being involved in cell cycle progression. Is identified as a candidate sequence of the human gene for 20. The NORF gene has NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27.
20. The method of claim 19, selected from the group consisting of: 21. A NORF containing a SAGE tag as shown in SEQ ID NOs: 67-811
A probe comprising at least 14 contiguous nucleotides of a gene. 22. The expression of the NORF gene is changed by at least 10% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. probe. 23. The NORF gene expression is altered by at least 25% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase and G2 / M. probe. 24. The NORF gene expression is altered by at least 50% between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. probe. 25. The expression of the NORF gene is altered by at least 100% between any two phases of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. probe. 26. The probe of claim 22, wherein the NORF gene is not expressed in at least one phase of the cell cycle selected from the group consisting of log phase, S phase, and G2 / M. 27. A statistically significant difference (95% or more confidence) in the expression of the NORF gene between any two phases of the cell cycle selected from the group consisting of logarithmic phase, S phase, and G2 / M. 23. The probe according to claim 22, which indicates the level. 28. The NORF gene has NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27.
23. The probe of claim 22, selected from the group consisting of: 29. The method of claim 17, wherein the expression monitoring step is performed with the nucleic acid molecule immobilized on a solid support. 30. The method of claim 29, wherein the nucleic acid molecules are on an array. 31. The method of claim 19, wherein the probes comprising a portion of the NORF gene are in an array on a solid support. 32. An array of probes on a solid support, wherein at least one probe comprises at least 14 contiguous nucleotides of a NORF gene containing a SAGE tag as set forth in SEQ ID NOs: 67-811. 33. The array of claim 32, wherein at least one NORF gene is involved in cell cycle progression. 34. The NORF gene has NORF numbers 1, 2, 4, 5, 6, 17, 25, and 27.
33. The array of claim 32 selected from the group consisting of: 35. The array of claim 32, comprising at least 100 probes with different sequences. 36. The array of claim 32, comprising at least 500 probes with different sequences. 37. The array of claim 32, comprising at least 1,000 probes with different sequences. 38. A method of identifying a candidate drug as a member of a class of drugs characteristically affecting gene expression in a yeast cell, the method comprising the steps of: contacting the candidate drug with a yeast cell. Monitoring the expression of at least one NORF gene in yeast cells whose expression is affected by the class of drug, the expression of at least one NORF gene in yeast cells as compared to the absence of a candidate drug If a difference is detected in, the candidate drug is identified as a member of that drug class. 39. The method of claim 38, wherein the step of monitoring expression is performed with the nucleic acid molecule immobilized on a solid support. 40. The method of claim 39, wherein the nucleic acid molecules are on an array. 41. The method of claim 38, wherein expression of two or more NORF genes is monitored. 42. The probe according to claim 21, which is immobilized on a solid support.