JP2001524317A - Methods for identifying and inhibiting functional nucleic acid molecules in cells - Google Patents

Abstract

  (57) [Summary] Two approaches are provided; the first provides a means to quickly and effectively identify essential and functional genes; and the second provides a biologically active nucleic acid molecule (ribozyme, EGS). , And antisense), which can be used to inactivate functional genes. In the first method, a library of EGS is prepared based on all possible known compositions. In a preferred embodiment, the EGS is 12-mer or 13-mer to the target bacterial RNAse for cleaving the substrate. This library is added to cells containing the gene to be screened (eg, E. coli). Cells in which EGS causes a loss of viability or other phenotype are identified. EGSs responsible for the loss of viability are analyzed and the resulting sequence information is used to identify genes within known genomic sequences. In a second method, a nucleotide molecule with optimal biological activity (eg, directing the cleavage of the gene of interest by RNase P) is used to generate two reporter genes (first in the phase with the gene of interest, and in the second). 2 are rapidly identified through the use of a vector containing the vector (either to prove the presence of the vector in the cell or as a control to aid in selection of cells containing the vector). Those cells in which the gene of interest is cleaved by the functional oligonucleotide molecule can then be identified by reference to reporter gene 1. The responsive functional oligonucleotide molecule is then isolated and characterized. These methods are powerful tools for identifying essential genes whose sequences are known only as part of the genome with unknown function, as well as identifying functional oligonucleotide molecules useful as diagnostic reagents and therapeutics. To provide the means for:

Description

DETAILED DESCRIPTION OF THE INVENTION

CROSS REFERENCE TO RELATED APPLICATIONS This application was filed by US Patent Application No. 08 / 976,220 (November 21, 1997 by Timothy W. Nilsen, Hugh D. Robertson and Thomas J. Kindt). No. 60 / 079,851 (March 3, 1998).
On October 0, Timothy W.M. (Filed by Nilsen).

BACKGROUND OF THE INVENTION The present invention is generally in the field of biologically active nucleic acid molecules (eg, EGS, ribozymes, and antisense RNA), and in the broad field of functional analysis of complex genomes, And more specifically, the use of biologically active nucleic acids (RNA) to specifically down-regulate the expression of messenger RNA encoding proteins essential for viability.

[0003] The recent development of automated DNA sequencing techniques and their application to the genomes of many organisms has resulted in the accumulation of vast amounts of nucleotide sequence information. At present, the genomic sequences of 16 bacteria, including several important pathogens, most recently those causing tuberculosis and syphilis, have been determined in their entirety. further,
The complete sequence of the "simple" prokaryote (Saccharomyces cerevisiae) is known and the nematode C. elegans. A similar analysis of elegans is easily published. More genomic sequences, both prokaryotic and eukaryotic, are likely to be presented in a short period of time.

A major challenge is the functional analysis of available and upcoming genomic information (ie,
Determination of the biological role of a gene represented by sequencing). In particular, it is important to identify genes encoding proteins essential for viability. Such proteins are clearly important in the development of effective chemotherapeutic agents targeting pathogenic organisms. Currently, there are several strategies available, alone or in combination, for functional genomic analysis, including bioinformatics, expression analysis, and target gene disruption. Informatics alone is unlikely to provide invaluable new gene function insights. For example, E. Although E. coli is the most studied organism to date, its genomic sequence has revealed that the function of about 40% of the gene is unknown. The expression profile provides mainly inferred information, and disruption of the target gene is crucial but intensive labor and time consuming.

[0005] Thus, identifying all or most essential genes in a particular organism,
It is highly desirable to have a powerful and high throughput method.

Accordingly, it is an object of the present invention to provide effective methods and compositions for identifying genes in bacterial and eukaryotic cells that encode proteins essential for survival.

It is a further object of the present invention to provide methods and compositions for reducing or inactivating the expression of such genes.

[0008] Ribonucleic acid (RNA) molecules serve as carriers of genetic information (eg, genomic retroviral RNA and messenger RNA (mRNA) molecules) and essential structures for protein synthesis (eg, transfer RNA (tRNA) and ribosomal RNA ( rRNA) as well as enzymes that specifically cleave nucleic acid molecules. Such a catalytic RNA molecule is called a ribozyme. Although ribozymes can theoretically cleave any desired site in an RNA molecule, in practice not all sites are cleaved efficiently by ribozymes designed to cleave those sites . This is in vivo
This is particularly true, where many experiments describe sites that are inefficiently cleaved by the target ribozyme. The problem is the determination of which sites can be most efficiently cleaved among many possible sites, rather than the entire deletion site, in the RNA molecule of interest. This is important because it is often desirable to identify the most effective cleavage site and not just any site that can be cleaved. The process of targeting one or a few sites in an RNA molecule essentially randomly and then testing for cleavage is unlikely to identify the most effective sites. Comprehensive testing of all sites is based on each ribozyme or external guide sequence ("EG
S ") is not practical due to the amount of work involved in making and testing. Innovi
WO96 / 21731 includes 80 different EGs targeting different sites.
The creation and testing of S describes the selection of efficient cleavage sites in this manner. However, this displayed only the possible sites. Techniques using similar labor intensive methods to identify accessible sites for cutting are described in US Pat. Nos. 5,525,468 and 5,496,698.

[0009] Kawasaki et al., Nucl. Acids Res. 24 (15): 301
0-3016 (1996), adenovirus E1 to assess the efficiency of cleavage of p300 RNA by hammerhead ribozymes in vivo.
The use of transcription encoding a fusion between the A-associated 300 kDa protein (p300) and luciferase is described. A few hammerhead ribozymes targeting sites with GUX triplets, which are required for cleavage by hammerhead ribozymes, were designed and expressed from intracellular vectors. Each vector expressed the p300 luciferase fusion RNA. The cleavage site in the transcribed part of p300 is evaluated by measuring luciferase activity.
Kawasaki et al. Tested each ribozyme separately. Therefore, those methods also do not solve the need for a quick and efficient selection process.

The presence or absence of secondary or tertiary structure as a substitute for practical testing for individual cleavable sites, or prior to such testing, based on theoretical considerations or at sites in the RNA molecule Attempts have also been made to predict which sites are available by testing empirically. See, for example, Ruffner et al., Biochemistry 29: 10695-10702 (1990), Z.
oumadakis and Tabler, Nucl. Acids Res. 23
1192-1196 (1995); Shiyamayama et al., Biochemi.
story 34: 3649-3654 (1995), Haseloff and G.
erlach, Nature 334: 585-591 (1988), and L.
IEber and Strauss, Mol. Cell. Biol. 8: 466-
472 (1995) describe an attempt to use the rules of RNA structure formation to predict cleavable sites. However, the structure of an RNA molecule cannot be accurately predicted from theoretical considerations, and determining the actual secondary and tertiary structure of an RNA molecule requires extensive experimentation. Also, identifying ribozymes and other biologically active molecules that function in cells is difficult because not all such biologically active molecules that are functional in vitro are functional in cells. Can be This is because those molecules are incorrectly located, sequenced, or bound, for example, by intracellular proteins.

[0011] Accordingly, an object of the present invention is to provide a biologically active RNA molecule (eg, ribozyme, EGS to ribozyme,
And antisense RNA).

[0012] It is a further object of the present invention to provide methods and compositions for identifying the most accessible sites in RNA, or nucleotide molecules involved in target RNA expression, for altered expression in vivo. It is to be.

[0013] It is a further object of the present invention to provide a functional oligonucleotide molecule for a site identified as an accessible site.

SUMMARY OF THE INVENTION Two approaches are provided; the first provides a means to quickly and effectively identify essential and functional genes; and the second provides a biological It provides a means to obtain active nucleotide molecules (ribozymes, EGS, and antisense), which can be used to inactivate functional genes.

In a first method, a library of EGS is prepared based on all possible known compositions. This is easily calculated by knowing the minimum required sequence required for targeting and cleavage by EGS length based on the predicted size of RNase P and the genome being screened. In a preferred embodiment, EGS is used to target bacterial RNase P to cleave substrates.
-Mer or 13-mer. This library of EGSs is added to cells containing this gene (eg, E. coli) for screening. Additional methods can be used to amplify cells that have survived exposure to this library or EGS. Cells in which the EGS causes a loss of viability or other phenotype are identified. EG responsible for loss of viability
The S (s) are analyzed and the resulting sequence information is used to identify genes within known genomic sequences.

In a second method, a nucleotide molecule having optimal biological activity (eg, directing the cleavage of a gene of interest by RNase P) is used to identify two reporter genes (first in frame with the gene of interest). , And secondly, as evidence of the presence of the vector in the cell or as a control to aid in selection of cells containing the vector). For example, this vector encodes any protein that confers drug resistance (target gene) in bacteria in a frame having a β-galactosidase gene (reporter gene 1) and a gene encoding antibiotic resistance (reporter gene 2). May be included. In a preferred embodiment, the vector comprises one of many possible functional oligonucleotide molecules (eg, EGS), although it may also be provided in an individual vector. This vector (s) is E. coli. Cells containing this vector can be isolated by treatment with an antibiotic that kills all cells that do not express the gene for antibiotic resistance. Then
Those cells in which the gene of interest is cleaved by the functional oligonucleotide molecule can be identified with reference to reporter gene 1. These identified cells can then be expanded. In one preferred embodiment, the gene of interest is essential for viability. In this case, the plate with the bacteria is first replicated,
Cells that are killed by cleavage of the mRNA of interest are then identified, and responsive functional oligonucleotide molecules are isolated from duplicate plates.

These methods are powerful tools for identifying essential genes whose sequences are known only as part of the genome with unknown function, and functional oligonucleotide molecules useful as diagnostic reagents and therapeutics Provides a means for identifying

DETAILED DESCRIPTION OF THE INVENTION Two general methods are described. The first is a method for the identification of essential or functional genes in the known genome using EGS, and the second is EGS, ribozymes, which are active on the desired RNA molecule. And methods for identifying functional oligonucleotide molecules, including antisense.
In a preferred embodiment of the latter method, the functional oligonucleotide molecule is
EGS, and the desired RNA molecule is an RGS transcribed from a known gene.
NA.

I. Definitions and General Elements of Methods and Reagents Ribonucleic acid (RNA) molecules are carriers of genetic information (eg, genomic retroviral RNA and messenger RNA (mRNA) molecules), and protein synthesis. Not only as an essential structure (eg, transfer RNA (tRNA) and ribosomal RNA (rRNA) molecules) but also as an enzyme that specifically cleaves nucleic acid molecules, or an element that indicates an enzyme that specifically cleaves nucleic acid molecules Can work as Any of these RNAs can be a target for cleavage or inactivation by a functional oligonucleotide molecule.

A key advantage of the disclosed methods and vectors is that they are the same or similar to settings using identified functional oligonucleotide molecules, or effector oligomers based on such identified RNA molecules. Evaluation of changes in expression of RNA of interest in an in vivo setting. Another advantage of the disclosed method is that all or a substantial number of accessible sites of the RNA of interest can be determined in a single assay. Such sites determined to be accessible for one type of functional oligonucleotide molecule may be accessible for another type of functional oligonucleotide molecule. In the case of ribozymes and EGSs, the disclosed method not only evaluates the cleavage of the RNA of interest, but also the ultimate desired phenotype (ie, the expression supported by the RNA of interest) as a result of such cleavage. Loss of type).

A. RNA molecule of interest The RNA molecule of interest can be any RNA molecule or part of an RNA molecule that can be transcribed. The target RNA molecule is preferably an RNA molecule involved in the expression of a target gene whose expression is inhibited. The RNA molecule can be an mRNA, a portion of an mRNA, a pre-mRNA, containing multiple introns or one intron. Alternatively, the RNA molecule can be a viral RNA.

Important pathogens include: the bacterium Pseudomonas aerugino
sa, Mycobacterium tuberculosis, Hemoph
ilus influenzae, Staphylococcus aureu
s, Mycoplasma pneumoniae, Escherichia
coli, Streptococcus pneumoniae, Neisse
ria gonorrhoeae, Streptococcus virid
ans, Streptococcus pyogenes, Proteus m
irabilis, Proteus vulgaris, Salmonella
typhimurium, Shigella dysentereae, Cl
ostridium difficile, and Klebsiella pn
eumoniae, and the fungus Candida albicans, Asper
gillus flavus, Aspergillus fumagatas,
And Histoplasmatus capsulatum.

B. Functional Oligonucleotide Molecules Functional nucleotide (typically RNA) molecules are designed to alter, or preferably inhibit, the expression of an RNA of interest. These molecules can be ribozymes, EGS for RNase P, or antisense RNA. Ribozymes and EGSs inhibit the expression of RNA molecules by cleaving or mediating the cleavage at the target site. Antisense RNA or DNA inhibits expression of an RNA molecule by sequence-specific interaction with the RNA molecule.

External Guide Sequence (“EGS”) Any RNA sequence in a prokaryotic or eukaryotic cell may be an RNase P
Can be converted to a substrate for In bacteria, the substrate comprises at its 5 'end a nucleotide complementary to the nucleotide 3' to the cleavage site in the RNA to be cleaved, and at its 3 'end the nucleotide NCCA (where N is any nucleotide)
It is manufactured using EGS having This is described in US Pat. No. 5,168,053, WO 92/03566 and Forster and Altman, Science.
ce 238: 407-409 (1990).

EGS to promote RNase P-mediated cleavage of RNA is also described in US Pat. No. 5,624,824, Yuan et al., Proc. Natl. Acad. Sci.
USA 89: 8006-8010 (1992), WO93 / 22434, WO
It has been developed for use in eukaryotic systems as described by 95/24489 and WO 96/21731. As used herein, “external guide sequence” and “EGS” refer to RNas in combination with a target RNA.
Refers to any oligonucleotide or oligonucleotide analog that forms a substrate for eP. Using EGS technology successfully, bacteria (Altman et al., (1993)) and mammalian cells in tissue culture (Yuan et al., Proc. N).
atl. Acad. Sci. USA 89: 8006-8010 (1992);
Liu and Altman, Gene Dev. 9: 471-480 (1995)
)) Reduced the level of gene expression.

The requirements for EGS to be functional with respect to prokaryotic RNase P are:
Less stringent than the requirements for eukaryotic EGS. The critical components of prokaryotic EGS are: (1) specifically binds to a targeted RNA substrate;
A nucleotide sequence that produces a short sequence base pair 3 ′ to the cleavage site of the substrate RNA, and (2) a terminal 3′-NCCA (where N is any nucleotide, preferably a purine is there). This sequence generally has no less than four, and most often six to fifteen, nucleotides complementary to the targeted RNA. It is not important that all nucleotides be complementary. The rate of cleavage is dependent on RNase P, the secondary structure of the hybrid substrate containing the targeted RNA, and the presence of 3'-NCCA in the hybrid substrate. Eukaryotic EGS also promotes prokaryotic RNase P cleavage and can be used for this purpose.

EGS for promoting cleavage by eukaryotic RNase P is referred to herein as eukaryotic EGS, and is secondary to the target RNA and similar to some of the tRNA molecules. Includes sequences that form structures and tertiary structures. Preferred forms of eukaryotic EGS are those that base pair with the target sequence 3 'to the intended cleavage site to form a structure similar to the aminoacyl receptor stem (A-stem).
Contains one nucleotide, base-paired to form a stem and loop structure similar to the T-stem and loop, followed by at least three nucleotides to base-pair with the target sequence to form a dihydroxyuracil stem-like structure. . Another preferred form of eukaryotic EGS is referred to herein as the short external guide sequence (SEGS), which, when hybridized to a target molecule, RNase
Provides a minimal structure that is recognized as a substrate by P. SEGS / Target RNA
The complex contains structures similar to the A and T stems of tRNA, the natural substrate for RNase P.

Ribozymes Ribozymes include any trans-cleaving catalytic nucleic acid. Several classes of such ribozymes are known and have been adapted or designed for cleavage of RNA molecules in a site-specific manner. Intron-derived ribozymes are described in U.S. Pat. No. 4,987,071, WO 88/04300, and Cech,
Annu. Rev .. Biochem. 59: 543-568 (1990), derived from a self-excised intron found in Tetrayhymena RNA. Hammerhead ribozymes are derived from self-cleaving RNA molecules present in certain viruses (Buzyan et al., Proc. Natl. Ac).
ad. Sci. USA 83: 8859-8862 (1968); Forste.
r and Symons, Cell 50: 9-16 (1987)). Target RNA
The design of hammerhead ribozymes for specific cleavage of molecules and their use is described in US Pat. No. 5,254,678, WO 89/05852, EP 3210.
21 and U.S. Patent No. 5,334,711. Derivatives of hammerhead ribozymes are described in US Pat. No. 5,334,711; WO 94/13789;
And WO 97/18312. Axhead ribozymes are derived from the self-cleaving domain in some viroid RNAs, such as hepatitis delta virus (US Pat. No. 5,527,895; US Pat. No. 5,225,337; WO
91/04319, and WO 91/04324). Ribozymes have also been developed using in vitro evolution techniques (WO 95/24489 and US Pat. No. 5,580,967).
No.).

Antisense molecules typically target RN via Watson-Crick base pairing.
A or a single-stranded DNA molecule or a substituted analog thereof that can bind to A and prevent translation of their RNA (Stein C
A, Antisense Nucleic Acid Drug Dev 8 (
2): 129-32 (1998); Crooke ST, Antisense.
Nucleic Acid Drug Dev 8 (2): 115-22 (19
98); Akhtar S, J Drug Target. 5 (4): 225-
34 (1998); Mizuno, T .; Proc. Natl. Acad. S
ci. USA, 81, (1983); Crooke ST, Biotechno.
l Genet Eng Rev 15: 121-57 (1998); Zame.
Nik, Projects for Antisense Nucleic
Acid Therapy of Cancer and Aids, Wick
Strom, Wiley-Liss, New York)). They are usually 15-30 nucleotides in length and are widely used to inhibit the expression of various proteins (Zamecnick, PC and Steven).
son, M .; L. Proc. Natl. Acad. Sci. USA, 75, 28
0 (1978); Agrawal, S .; Proc. Natl. Acad. Sc
i. USA, 85, 7089 (1988)). In addition to inhibiting translation of mRNA,
DNA-based antisense can also inhibit protein expression by presenting a DNA-RNA hybrid as a target for cleavage by the endogenous RNase H enzyme (Crooke ST, Antisense Nuclei).
c Acid Drug Dev 8 (2): 133-4 (1998); Cas
elmann et al., Intervirology 40 (5-6): 394-9 (
1997); Giles, R .; V. And Tidd, D.A. M. , Nucleic
Acid Res. 20, 763 (1992)), thereby destroying the target RNA. Antisense molecules can be made more resistant to nucleases by introducing a phosphorothioate diester bond instead of a chemical modification such as a 2 'modification and a phosphodiester bond (Agrawal S and Zhao Q, Antisense) Nucleic Acid Drug D
ev 8 (2): 135-9 (1998); Agrawal, S .; Proc
. Natl. Acad. Sci. USA, 85, 7089, (1988)),
And the duplex of these molecules with RNA is recognized by RNaseH.

C. Vectors The disclosed methods generally involve reporter genes, genes for selection of cells with or without functional oligonucleotide molecules, and expression and replication of vectors. A vector having a specific element containing a sequence necessary for is used. The disclosed vectors are generally of two types, each type being:
It is adaptable to assays for identifying essential and functional genes, or assays for identifying functional oligonucleotide molecules. These are described generally below.

(Vector for Use in Identification of Essential Gene or Functional Gene) A vector for use in the method for identifying an essential gene or a functional gene is a functional oligo containing a degenerate targeting sequence. Encodes a nucleotide molecule. By including a degenerate targeting sequence, the set of functional oligonucleotide molecules encoded includes functional oligonucleotide molecules that target all possible sequences. The length of the target sequence is preferably chosen to match the length of the unique sequence present in the genome of the cell being assayed. This relationship is described in further detail below. The effect is to obtain a set of vectors that collectively encode a set of functional oligonucleotide molecules that target all possible unique sequences in the genome of the cell of interest. The set of vectors is transformed or transfected into cells, and the cells are screened or selected for cell death or phenotypic alteration of interest. The selected cells are
Includes functional oligonucleotide molecules that are targeted to essential or functional genes in the cell. A functional oligonucleotide molecule can then be identified by characterization of the vector in the cell. Genes targeted by a functional oligonucleotide molecule can be identified by correlating the targeted sequence in the functional oligonucleotide molecule with a known genomic sequence.

(Vector for Use in Identification of Functional Oligonucleotide Molecule) A vector for use in a functional oligonucleotide molecule assay comprises a fusion transcript containing an RNA molecule of interest and an RNA encoding reporter protein A. Includes reporter gene 1 encoding. Inactivation of the RNA molecule of interest alters reporter protein A expression. This vector also contains a second reporter gene 2, which encodes a second reporter protein B. Expression of the second reporter protein B is used to detect transformation or transfection of the vector into cells and to control the effect on the expression of the first reporter protein not due to inhibition of expression of the RNA molecule of interest. As both. This vector also encodes a functional oligonucleotide molecule targeted to the RNA of interest. The method preferably uses these sets of vectors, wherein each vector in the set encodes a different functional oligonucleotide molecule and is targeted to a different site of the RNA molecule of interest, respectively. . This set of vectors is transformed or transfected into appropriate cells, and the cells are transformed into a second reporter protein B
Are screened or selected for expression. Cells expressing the reporter protein B are then screened or selected for cells that do not express the first reporter protein A or that express only low levels of the reporter protein A. These cells carry the most efficient functional oligonucleotide molecules that can then be identified by characterizing the vector within the cell.

The vector can be an autonomously replicating vector, a viral vector, a nucleic acid that integrates into the host chromosome, and a nucleic acid molecule that can be primarily expressed. The reporter gene can be expressed using any suitable expression sequence. Numerous expression sequences are known and can be used for expression of a reporter gene. Expression sequences can be generally classified as promoters, terminators, and enhancers for use in eukaryotic cells. Expression in prokaryotic cells also requires a Shine-Dalgarno sequence, immediately upstream of the coding region, for proper translation initiation. An inducible promoter is preferred for use with the first reporter gene. This is because the expression of the first reporter gene is preferably regulatable.

It is preferred that plasmid vectors containing promoter and control sequences derived from species compatible with the host cell be used with these hosts. Preferably, the vector carries a replication sequence. The vector can be used to transiently transfect or transform a host cell, or can be integrated into the host cell chromosome. However, preferably, the vector contains sequences that permit replication of the vector as well as stable or semi-stable maintenance of the vector in the host cell. Many such sequences are known for use in a variety of eukaryotic and prokaryotic cells, and their use in vectors is conventional. Usually, it is preferable to use a replication sequence known to function in a target host cell. For example, the use of a vector-derived origin of replication such as pBR322 and pUC19 is preferred for prokaryotic cells, a vector-derived origin of replication such as YEP24 and YRP17 is preferred for fungal cells, and replication from SV40 and pEGFP-N. The origin is preferred for eukaryotic cells.
All these examples are commercially available (New England Biolab)
s; Clontech).

[0035] A preferred vector for use in prokaryotic cells is Bluescript-
SK ⁺ (Stratagene). A preferred vector for use in eukaryotic cells is the shuttle vector pEGFP-N (Clontech).
This vector encodes a green fluorescent protein (GFP) that has been optimized for maximum activity in mammalian cells and is designed for expression of a GFP fusion protein. This vector also contains a multiple cloning site (MCS) 5 'to the GFP sequence designed for the generation of a fusion protein in all three reading frames. The MCS contains D encoding the RNA of interest.
The NA can be inserted and used to produce a gene encoding a fusion transcript encoding the fusion protein.

The reporter protein can be any protein that can be detected either directly or indirectly. These are enzymes that can produce specifically detectable products (
For example, β-galactosidase, luciferase, and alkaline phosphatase), and proteins that can be directly detected. Virtually any protein can be detected directly, for example, by use of a specific antibody against the protein. A preferred reporter protein that can be detected directly is green fluorescent protein (GFP). GFP (Jellyfish Aequorea victoria)
Origin) produce fluorescence when exposed to ultraviolet light without the addition of substrate (Chalfi).
e et al., Science 263: 802-5 (1994)). Many modified GFPs have been made that produce fluorescence approximately 50-fold stronger than wild-type GFP under standard conditions (Cormac et al., Gene 173: 33-8 (1996)
); Zolotukhin et al. Virol 70: 4646-54 (199
6)). This level of fluorescence allows detection of low levels of expression in the cells.

Reporter proteins that produce a fluorescent signal are useful. Because such signals allow cells to be sorted using FACS. Another method of sorting cells based on the expression of the reporter protein involves using the reporter protein as a hook to bind the cells. For example, cell surface proteins such as receptor proteins can be bound by specific antibodies. Cells expressing such a reporter protein are captured, for example, by using an antibody bound to a solid substrate, by using an antibody bound to magnetic beads, or by capturing an antibody bound to the reporter protein. obtain.
Many techniques for the use of antibodies as capture agents are known and can be used with the disclosed methods. When the second reporter protein is CD4,
The preferred form of cell surface protein for use as the first reporter protein is CD8, otherwise CD4 is preferred.

The reporter protein can also be a protein that regulates the expression of another gene. This allows for the detection of reporter protein expression by detecting the expression of the regulated gene. For example, a repressor protein can be used as a reporter protein. Inhibition of reporter protein expression in turn results in deregulation of the regulated gene. This type of indirect detection allows for positive detection of inhibition of reporter protein expression by a functional oligonucleotide molecule. One preferred form of this type of regulation is the use of an antibiotic resistance gene that is regulated by a repressor protein used as a reporter protein. Exposure of the host cell to the antibiotic derepresses expression of the antibiotic resistance gene, so that only cells in which reporter gene expression has been blocked grow.
Preferably, the second reporter protein is a protein that confers antibiotic resistance on its host cell or cell surface protein. The use of antibiotic resistance proteins is preferred in prokaryotic host cells, and the use of cell surface proteins is preferred in eukaryotic host cells. The most preferred cell surface protein for use as a second reporter protein is CD4.

A gene encoding a protein for use as a selection agent can also be inserted into the vector. Exemplary materials are shown in Table 1. The substance may confer resistance to antibiotics or other substances that are toxic to the cell, or may aid in the selection of cells containing a functional oligonucleotide molecule.

[0040]

[Table 1] The vector can be constructed using well-established recombinant DNA technology (
See, for example, Sambrook et al., Molecular Cloning: A La.
The Bounty Manual, Second Edition, Cold Spring Harb
or Laboratory Press, New York (1990)). Preferably, the base vector is prepared first. The DNA encoding the RNA molecule of interest can then be inserted into this base vector to form a second base vector. A different second base vector can be constructed for each RNA molecule of interest. Finally, a library of DNA encoding a functional oligonucleotide molecule can be inserted into a suitable second base vector. Similar base vectors can be readily used with any RNA molecule of interest, and similar second base vectors can be used with any suitable library of functional oligonucleotide molecules. For example, a similar second base vector can be used for a library of ribozymes, a library of EGSs, and a library of antisense RNA molecules.

[0041] Host cells can be transformed with the disclosed vectors using any suitable means, and are modified with conventional nutrients as appropriate for the introduction of a promoter, selection of transformation, or detection of expression. It can be cultured in a medium. Culture conditions (eg, temperature and pH) suitable for host cells are well-known. The concentration of the plasmid used for transfection of the cells is preferably titrated to reduce the likelihood of expression of multiple vectors encoding different functional oligonucleotide molecules in the same cell.

Preferred prokaryotic host cells for use in the disclosed methods are those described in E. coli. co
Li cells. Preferred eukaryotic host cells for use in the disclosed methods include the monkey kidney CVI strain (COS-7, ATCC) transformed with SV40.
CRL 1651); human embryonic kidney strain (293, Graham et al., J. Ge.
n Virol. 36:59 [1977]); Infant hamster kidney cells (BHK
, ATCC CCL 10); Chinese hamster ovary cell DHFR (CH
O, Urlaub and Chasin, Proc. Natl. Acad. Sci
. (USA) 77: 4216, [1980]); mouse Sertoli cells (TM4
, Mather, Biol. Reprod. 23: 243-251 [1980]
); Monkey kidney cells (CVI ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical cancer cells (HELA, ATCC CCL 2); dog kidney cells (MDCK, ATCC CCL).
CL 34); Buffalorat hepatocytes (BRL 3A, ATCC CRL)
1442); human lung cells (W138, ATCC CCL 75); human liver cells (hep G2, HB8065); mouse mammary adenocarcinoma (MMT 060562).
, ATCC CCL 51); TRI cells (Mather et al., Annals N).
. Y. Acad. Sci 383: 44-68 (1982)); human B cells (D
Audi, ATCC CCL 213); human T cells (MOLT-4, ATCC
CRL 1582); and human macrophage cells (U-937, ATCC
CRL 1593).

II. Methods for Identifying Essential or Functional Genes Methods have been developed that employ EGS combinatorial libraries to identify mRNAs that encode proteins essential for cell viability or other functions. . The example illustrates a method as used in bacteria (eg, E. coli).

EGS is a short oligoribonucleotide of general composition N _n ≧ ₇ NACCA. When such a sequence anneals to a complementary sequence in the second RNA,
The duplex serves as a substrate for RNase P. RNase P is a ubiquitous enzyme whose normal role in cellular metabolism is the maturation of the 5 'end of transfer RNA. All characterized RNases P consist of RNA and protein components. In bacteria, there is a single protein subunit and a single RNA subunit; the RNA component is catalytic. tRNA
At maturation, RNase P recognizes determinants in the receptor stem, D-loop, and unduplexed CCA; in fact, the CCA can base pair with the RNA subunit of the enzyme. It is known.

EGS binds to another RNA (its target) via base pairing. That RNA
-The RNA duplex mimics the natural tRNA substrate of RNase P and recruits its enzymes, which cleave the target RNA at the single / double stranded boundary. There are no known sequence or base composition restrictions in the helical region;
In theory, RNase P cleavage of any RNA in the cell, including mRNA, could be dictated. The ability of RNase P to cleave RNAs other than natural substrates serves as the rationale underlying the methodology described below.

Summary of Experimental Strategies for Identification of mRNA Encoding Essential Genes The assumptions underlying the design of experiments are that all (or most) mRNA
Can be cleaved by RNase P in the presence of the appropriate EGS sequence.
Thus, if all possible EGS sequences can be expressed, some of them will cause cleavage of mRNA encoding proteins essential for cell viability. Restoration and characterization of these "active" EGS sequences allows the identification of their targeted mRNA (gene) through various strategies. The prerequisites for a successful application of this idea are the ability to synthesize and express all or most of the possible EGSs, and to isolate those EGSs that trigger the desired phenotype (eg, cell death) Method. In addition, there must necessarily be an EGS capable of causing almost complete destruction of that message of interest.

In this regard, it is not currently possible to predict which EGS is most effective. Using a rational design, Altman and colleagues were able to achieve only modest reductions in specific mRNA using a single EGS sequence (Guerrie
r-Takeda, C.I. Proc. Natl. Acad. USA 94,8
468-8472, 1997). If the EGS sequence is used in combination with either (ie, multiple EGSs targeted to the same mRNA), or
When the GS / target ratio was extremely high, a biologically significant decrease was observed. To confirm that it could be expected or rational that a single EGS could severely down-regulate the expression of specific mRNAs, the 52 chromamphenicol acetyltransferase mRNAs targeted A library of EGS sequences was designed and it was found that only one of the 52 EGSs caused a ≧ 90% decrease in gene expression. These experiments are described in detail in Example 4 below. The results of these examples have three important implications for the design of methods to identify essential or functional genes: 1) Single E
It is reasonable to expect that GS can lead to complete or near-complete degradation of specific mRNAs; 2) Almost only one out of 50 EGS sequences appears to achieve the required level of inhibition. And 3) EG likely to be active in vivo
Current methods of predicting S are unreliable. This last consideration makes it very unlikely that the proper EGS sequence can be designed to evaluate the function of each known gene in the bacterial genome in a systematic way. Furthermore, the fact that only a small subset of EGSs targeted to a particular gene is likely to elicit a biological effect affects the design of extensive screenings of the genome and interpretation of results.

Based on calculations as set forth in the Examples, random or overlapping sequences in the EGS for use in identifying essential or functional genes are
Typically between _N10-13 ; in fungi, typically it is between _N10-15 , and in mammalian cells, typically it is between _N10-18 , Subsequent sequences had to target RNase P of the host cell, which cleaves the targeted RNA. In the case of prokaryotic RNase P, this is as simple as ACCA; the structure required for eukaryotic RNase P is more complex, and can be designed as needed as described above, A structure similar to a part of the structure is manufactured.

This method is better understood by reference to the following non-limiting examples.

Example 1 First Proof of the Principle of Extensive Genomic Functional Analysis Using a Library of EGS Sequences EGS Design The first experiment fixed two positions at the 3 ′ tip as cytosines 13 mer E
The GS sequence was used. Therefore, the EGS has the sequence N ₁₁ CC, ACCA continues sequence: having a 5'- N ₁₁ CCACCA-3 '. I chose a 13 mer. This is because any particular 13-mer has a complexity of about 4 × 10 ⁶ base pairs, or 8 × 10 ⁶ when the two strands are considered separately. This should be statistically unique in the E. coli genome. Note that the effective complexity of this genome is 4 × 10 ^{6 because} only one strand of DNA is transcribed. The probability of finding any particular 13-mer sequence is about 1/3 × 10 ⁷ . The second thought about selecting a 13-mer sequence was that 13
At the time of this experiment, the shortest EGS known to elicit an effect in vivo
Was the fact that

Conditional expression of EGS sequences The overall goal of this experimental design was to encode proteins that are essential for viability.
Identification of EGS sequences that trigger RNA cleavage. A direct and obvious consequence of this design is that EGS expression must be controllable;
If EGS expression was constitutive, cells with the desired sequence would die and consequently the appropriate cells would be lost upon expansion of the library. E. FIG. In E. coli, there are many known regulatable promoters. Particularly preferred promoters initiate transcription at the insertion site of the foreign sequence (
For example, a controllable arabinose promoter). This example uses the ara C-pBAD system for conditional expression of EGS sequences. This is a tightly regulated system where transcription is repressed in the absence of arabinose and induced in the presence of low levels of arabinose.

FIGS. 2 a-c show the constructs and plasmids used in the experiments. ar
The ac-pBAD promoter is immediately adjacent to unchanged A (to facilitate the initiation of transcription), followed by the EGS-ACCA sequence. This sequence is then followed by a T7e transcription terminator. Transcripts from such a construct are identified in this EGS-ACC
A sequence and this terminator. Altman previously had an EGS-AC
It was observed that the sequence 3 'of CA did not interfere with EGS function. In fact, the presence of this terminator sequence with an associated 3 'stem loop structure indicates that EGS
Enhances mediating activity. Perhaps because the stem loop makes this short transcript less likely to disappear. Thus, the EGS sequence associated with this terminator is the EGS sequence alone (ie, the EGS sequence produced by the cis-acting ribozyme).
Sequence)) to a higher steady state level.

This expression was evaluated by RNA blot analysis of the EGS sequence under various conditions. As expected, when no arabinose is added to the growth medium, no expression is observed in cells with the EGS plasmid. In the presence of arabinose,
EGS expression is easily detected and high steady state levels of transcripts are induced 3
Observe after 0 minutes. This level of expression is maintained for several hours.

(Design of Combined EGS Library) A library of randomized EGS was created by ligating the appropriate DNA insert to the pBAD plasmid to express all possible EGS sequences. The ligated plasmid was then used to transform E. coli. E. coli was transformed. Sufficient amounts of independent transformants were obtained, ensuring that the entire library was represented (the complexity of the EGS sequence was about 4 × 10 ⁶ ; the resulting independent transformants were about 2 × 10 ⁷ ; The number of transformants indicates 95% confidence that all members of this library are represented). Following transformation, this library (each bacterium containing one plasmid, and therefore one EGS) was amplified via liquid growth under non-induced conditions. Then, many (50 or more) individual plasmids were purified and sequenced,
The sequence of the insert was confirmed to be correct and the absence of nucleotide bias in the randomized (EGS) region.

Initial Testing of the Efficiency of a 13-mer EGS Library Identification of EGS sequences that, when expressed, cause lethality usually involves the analysis of thousands of individual bacterial colonies. Therefore, a technique was developed to prove the usefulness of the EGS library via a simplified approach. The method is based on the fact that some chemicals are only used when certain non-essential genes are expressed. Take advantage of the fact that it is toxic to E. coli. Therefore, the survival of bacteria in the presence of such substances depends on the absence of a functional gene product. For example, high levels of cobalt
E. FIG. fatal to E. coli. This lethality results from the accumulation of cobalt ions within the cell. If cobalt is prevented from entering the cell (ie, due to inactivation of the ion transporter), the cell survives. Several such toxicants were selected (Table 1) and studies were performed to determine the appropriate EG
Surviving cells were harvested with this toxic treatment if and only if S was expressed.

The colonies were then screened for arabinose (inducer) specific survival. Inducer-dependent survivors that are not arabinose could result from spontaneous mutations in the relevant gene. An outline of the design of this experiment is shown in FIG. Briefly, cells were grown to log phase and then exposed to arabinose. At this point, EGS expression was induced. The cells were then left in arabinose for several generations before being treated with the toxic agent. This incubation time is important.
This is because this EGS only affects mRNA levels and does not affect pre-existing proteins. Thus, the pre-existing protein must disappear (or be diluted via growth) prior to sorting.

Following appropriate incubation, the population of cells is individually treated with a selection agent,
The surviving strain was recovered. After removal of the selection, cells were plated on non-selective media so that independent colonies could be analyzed. At this point individual clones were then assayed for their ability to survive the selection treatment in the presence or absence of arabinose. Importantly, several arabinose-dependent (ie, EGS-dependent) viable strains were recovered for each of the seven different selections. Equally important, the observed resistance phenotype was highly specific; that is, EGS that confers resistance to one selected substance is resistant to any other such substance. Did not give. Ultimately, clear evidence linking the resistant phenotype to a particular EGS was obtained by retransforming native bacteria with plasmids expressing the appropriate EGS sequences.

The nucleotide sequence of each of the active clones was then determined and co
li's published nucleotide sequence. This active EGS sequence was expected to be complementary to the transcribed region of the genome. But,
In each case, perfect complementarity (Watson-Crick) was not observed. These observations have shown, among other possibilities, that 13 nucleotides of the EGS sequence may have more information than is required for biological function. These experiments with the selection material clearly demonstrated the following: 1. High penetrance, high specificity, conditional phenotype is induced by the EGS sequence Active EGS sequences can be efficiently and rapidly isolated from complex libraries.

Example 2: Isolation and characterization of EGS sequences that, when expressed, cause growth impairment or lethality. The experiments described in Example 1 demonstrate that EGS in the presence Demonstrated survival of dependence. To more directly demonstrate the utility of an EGS-mediated approach to identifying essential genes, Es that cause loss of viability when expressed
A study was performed to obtain a GS sequence. Although similar, these experiments differ somewhat in design from the experiment outlined in Example 1. That is, there is no method for selecting for dead or dying cells. Therefore, it is necessary to screen for EGS whose expression leads to a proliferative disorder.

The simplest method for screening for lethal phenotypes is by the “replica plating method”. In such an approach, individual clones (
The colonies arising from a single bacterium are arrayed in any of a variety of ways, and a replicate array is generated. In its simplest form, comparing the growth of colonies in either the presence or absence of the inducer (and therefore EGS) reveals the EGS that causes the growth defect. Confirmation that the deficiency is EGS-mediated can be made in a variety of ways, including retransformation.

Screening for Lethal EGS Sequences Using a Simple Replica Plate Method
Utility depends on the complexity of the library and the expected frequency of positives.
Exist. With the limited data available, the expected number of “positive” clones can be predicted. E. FIG. The E. coli genome is 4.6 × 10 ⁶ bp of complexity. Assuming a transfer density equal to almost 100%, and little, if any, symmetrical transfer, the transferred complexity is still greater.
4.6 × 10⁶It is. Further, a reasonable prediction of an essential gene is about 10%. If there is no significant difference between essential and essential genes in average size
Assuming that, the effective target size of EGS to induce a lethal phenotype is
, 4.6 × 10^FiveApproximately 400 genes of bases or 1 kb (polar effects and the fact that the majority of genes in E. coli are transcribed as part of the operon)
Is not taken into account). Using the second method as shown in Embodiment 4
From the resulting data set, only about 2% of the targetable nucleotides
Available ", ie, about 10,000 targets located in the genome
It is possible. Therefore, if you use a truly randomized library,
, Expect about 10,000 active EGSs. This N₁₁For CC library
Only 1/16 of the potential sites can be targeted and the number of active EGSs
Reduce to a few hundred. Assuming these conditions (and accepting this assumption)
, This N₁₁CC library (4.6 × 10⁶Yields one positive clone per 10,000 clones tested.

One out of 10 ⁴ is an unacceptably low number for the purpose of direct replica plating assays, and therefore the frequency of active clones is a classical technique. Increased using penicillin concentration. Penicillin enrichment has been used in microbiological research for decades and aids in the recovery of rare mutations. This technique is based on the fact that penicillin, or its various derivatives, including ampicillin, kills only actively growing cells; non-proliferating cells escape penicillin killing. This enrichment strategy is called lethal EG
It is fairly simple to apply for the isolation of S. Briefly, after induction of EGS expression with arabinose, cells with "effective" EGS (ie, cells targeted to an essential gene) cease to grow and thus become resistant to penicillin killing.
Importantly, the time at which individual cells stop growing is a function of how long it takes for the pre-existing protein of interest to decrease below a viable threshold of activity (via turnover or dilution). is there. Thus, m encoding different essential proteins
To enrich for lethal EGS targeted to RNA, it is necessary to treat with penicillin at various time points after induction (see FIG. 4).

Enrichment experiments were performed as outlined above and following amplification of viable strains, cells harvested at various time points after induction were assayed for inducer-dependent lethality via replica plating. did. Several clones that showed obvious growth defects only in the presence of the inducer were recovered. As described above, the appropriate plasmid is recovered, sequenced,
And transformed natural bacteria. In each case, the phenotype (slow or no growth) was shown to be due to the presence and expression of specific EGS sequences. For the EGS recovered from the above selection, its nucleotide sequence was
E. coli genome sequence. Also, there was no perfect (Watson-Crick) complementarity between the recovered EGS and the published sequence.

These experiments show that, when expressed, E. coli that it is possible to isolate specific EGS sequences that elicit a distinct growth deficient phenotype in E. coli.
Proven directly. In addition, they have demonstrated the utility of using penicillin treatment to enrich for active EGS sequences. In summary, the above experiments show that: 1. A library of EGS sequences can be constructed and expressed under tight regulation; These complex libraries can be "classified" so that specific EGSs that confer various phenotypes, including lethality, can be recovered and characterized.

Example 3: Analysis of an EGS library with a targeting (guide) sequence of less than 13 nucleotides Principle As discussed in Example 2, a specific sequence consisting of a 13 base guide sequence and ACCA EGS is based on E.I. It has been shown to induce a high penetrance phenotype in E. coli. However, using conventional (Watson-Crick) base pairing rules, the EGS sequence and E. It was not possible to find perfect complementarity with the E. coli genomic sequence. These results can be explained in various ways, listed below: This 13 mer provided "excess" information. In such a case,
Some mismatch between the EGS and the target is tolerated. However, there is no a priori way to distinguish which positions can be mismatched.

2. The most “effective” EGS is selected based on non-Watson-Crick interactions with its target. Such interactions are associated with fluctuation pairing and specific 3
The following structure may be included. Also, these types of interactions would not be possible to predict or rule out without prior knowledge of the target.

Given this uncertainty associated with linking this EGS sequence to a particular target, it is essential to systematically investigate the minimum length required for EGS activity in vivo. Further, it may be desirable to determine the absolute activity (ie, in the absence of selection or enrichment) of each specific library. The assay selected for activity was of the inducible lethal (proliferation deficient) phenotype described in Example 2. These experiments were performed in complete medium to avoid collecting inducible (EGS-dependent) auxotrophs (a significant portion of the E. coli genome becomes essential in minimal medium). Four libraries of EGS sequences (ie, N ₉ ACCA, N ₁₀ ACCA, N ₁₁ AACCA and N ₁₂ ACCA) were prepared with progressively varying lengths of the guide from 9 to 12, and were required in vivo. Length was explored. Unlike the 13-mer library described above, none of the bases in the guide sequence retained invariance. In all cases, sufficient amounts of transformants (5 × complexity) were obtained to ensure a proper representation.

(N₉ACCA Library and N_TenAnalysis of ACCA library) This N₉Library and N_TenTo assess the activity of the library, approximately 45,000 individual clones from each with or without EGS expression
Colonies (ie, in the presence or absence of arabinose)
These libraries were assayed for performance. N₉The complexity of the library is about 2.6 × 10^FiveAnd N_TenLibrary complexity is about 1 × 10⁶Is
. With both libraries, only a few positive clones were found. N ₉ The estimated number of “active” molecules in the library is more than 50 and N_Ten The estimated number of "active" molecules in the library was about 100. N₉Further analysis of the library is via penicillin enrichment to facilitate recovery of active clones.
Most clones that induce phenotypes have a single nucleotide sequence.
Revealed that it can be aligned as a part. The 10 mer library is
Did not analyze them.

The fact that both the 9-mer and 10-mer libraries are essentially inactive is significant. Because these results determine a lower limit for the length of the biologically active EGS sequence in vivo. Furthermore, these results are significant when aligned with the findings using a 13-mer library. Clearly, the 13-mer contains all possible 9-mers and 10-mers therein, but the 1-mer
A number of distinct clones conferring different phenotypes using the 3-mer library could be recovered. A reasonable explanation for the two sets of results is that the 13 mer could tolerate at most two mismatches while retaining activity. Ultimately, and unimportantly, the results with this short EGS sequence were This suggests that the minimum length of biologically active EGS in E. coli is approximately the length of a sequence that needs to be statistically unique in the bacterial genome.

(N₁₁Analysis of ACCA Library The absolute, ie, non-enriched, activity of the 11-mer library was
Determined by analysis of 00 independent colonies. Active (proliferation inhibitor) in this population
Hazard) Estimated from the number of EGS sequences, the 11-mer library had a total of about 4 × 10 ⁶ Out of the complexity of contains between about 1,200 and 1,400 active molecules. 11-mer library compared to 9-mer and 10-mer libraries
The increase in activity in the abdominal region indicates that the total complexity of 11 elicits significant biological activity.
It is suggested that it is the minimum value. In addition, the number of “active” EGS sequences
The second method, exemplified by Example 4, was greater than the total estimated number of
That was well below the conservative estimate of the number of total active EGSs to be estimated. these
, A significant number (greater than 50) of the active 11 mer
Analyzed via sequencing. Many, but not all, of these sequences
. Perfect (Watson-Crick) complementation with the transcribed region of the E. coli genome
Had the property. Some of the sequences having such complementarity are described in E. coli. in E. coli
MR which is known to encode proteins essential for viability
Others presumed to inactivate NA and others identified mRNAs of unknown function (Table 2).

[0071]

[Table 2] Whether this observed complementarity is biologically significant is not yet known, but at least two ideas suggest that the 11-mer is the next best for genomic screening. First, the number of active clones is small. Second
In addition, nucleotide sequence analysis revealed that the active 11-mer had significant sequence bias therein. In this regard, the 50 EGs sequenced
Of the S, more than 20 share a common 6 nucleotides, and there were many examples of shared 7 and 8 nucleotides. Fifty eleven
The linear sequence complexity of the mer is 550 nucleotides. The probability of finding a particular 6 nucleotides to appear is 1/4096, and the probability of finding a longer specific oligomer is correspondingly lower. One possible explanation for these results is that at short EGS lengths, RNase P has strong sequence selection. If true, an 11 mer may have unacceptably severe target site restrictions. As shown below, neither of these restrictions (low activity and target site restriction) is observed with the 12-mer.

Analysis of N ₁₂ ACAA Library Approximately 50,000 independent clones were evaluated for EGS-dependent activity in the _12- mer library as described above. There was a dramatic increase in active clones compared to the 11-mer library. About 13,000 active molecules in 12m
The evaluation was performed using the er library. This number of active clones is significant. This is because it is much more than the number of putative essential genes, and matches well with the putative level of activity estimated by the second method. The nucleotide sequence of the 14 active 12-mers is shown in Table 3.

[0073]

[Table 3] The six most significant of these sequences bind to the target sequence with complete Watson-Crick base pairing. Some of these targets have been individually identified as essential genes by other techniques.

Importantly, these sequences do not display any 11-mer non-random features. However, the "active" 12-mer that has been characterized to date is described in E. coli. It does not display the unique 12-mer statistically significant features present in the transcribed region of the E. coli genome. These results indicate that optimal EGS activity in vivo is not unconditionally linked to conventional (Watson-Crick) base pairing with the target sequence. One conclusion of these results is that a simple (ie, simple homology matching) computer-based search can be performed to determine the activity E related to biological targets.
In GS it is unlikely to be decisive.

(Strategies for De-Overlapping, ie, Clear Link between Active EGS Sequences and These Targets)
acillus subtilis, Borrelia burgdorfer
ri, Chlamydia trachmatic, Escherichia
E. coli K-12 MG 1655, Haemophilus influ.
enzae Rd, Helicobacter pylori 16695, M
ycobacterium tuberculosis, Mycoplasma
gentitalium G37, Mycoplasma pneumoni
ae M129, Rickettsia prowazekii, Trepon
ema pallidum has been completely sequenced and the other 38 bacterial genomes (Aquifex aeolicus, Arcanobacterium p.
yogenes, Bacillus pumilus, Bacillus su
btilis, Borrelia burgdorferi, Buchnera
aphidicola, Chlamydia trachomatis, Cl
ostridium MCF-1, Corynebacterium glut
amicum, Enterobacter aerogenes, Escher
ichia coli, Haemophilus influenzae, He
licobacter pylori, Lactobacillus delb
rueckii, Lactobacillus helveticus sub
sp. jugurti, Lactobacillus reuteri, Lac
tococcus lactis, Mycobacterium tuberc
ulosis, Mycoplasma genitalium, Mycopla
sma pneumoniae, Pantoea citrea, Pasteu
rella multicida, Prevotella ruminicol
a, Rhizobium sp. NGR234, Rhodothermus m
arinus, Rickettsia prowazekii, Ruminoc
occus flavorefaciens, Salmonella berta,
Salmonella typhimurium, Staphylococcus
s aureus, Streptococcus agalactiae, S
treptomyces clavuligerus, Streptomyces
s nigrifaciens, Streptomyces phaeochr
omogenes, Synechocystis PCC6803, Trepo
nema pallidum, Yersinia pestis, Zymomo
nas mobilis) is ongoing. In the simplest embodiment, EGSs that cause the cleavage of RNA encoding essential or functional genes are identified by comparison to known sequences, searching for complementary regions by Watson-Crick base pairing.

The examples show that a significant number (eg, 6 of 14 shown in Table 3) of “active” E
GS can be matched through homology search of unique transcribed regions of the bacterial genome,
Indicates not all. However, this result does not rule out or even suggest that EGS activity is not mediated by base pairing. Probably "fluctuation"
Some would be rather limited to binding (i.e., G: U), but some were EGS-
It appears to be acceptable in the target duplex. If accurate, it does not exclude, but complicates, simple computer-based deconvolution. It is expected that active EGS will reveal regions of the mRNA that are available for hybridization. These regions are unlikely to be restricted to a single nucleotide. Therefore, if enough EGSs are analyzed, clusters of non-identical but overlapping active clones should be observed. Analysis of these clusters
The actual target EGS of the GS should be revealed, and at the same time, by comparing active EGS with themselves, it should be revealed if wobble binding is tolerated.

Although computer-based deconvolution is desirable, it does not prove an association between the EGS and the suspected target. At least two strategies can be used to provide definitive evidence of EGS-mediated inactivation of specific targets. The first is a "genetic" test. Assume that the EGS of interest is believed to target mRNA (base pairing) within the coding region. By exploiting the degeneracy in the genetic code, it is possible to change the nucleotide sequence of an mRNA while retaining the ability of the mRNA to encode the same protein. Changed mR
Ectopic (ie, plasmid-based) expression of NA should be immune from an EGS attack, thereby rescuing the EGS-dependent phenotype. The second approach is based on direct analysis of RNA expression patterns in EGS expressing cells. The availability of the entire genomic sequence results in the generation of an ordered array of DNA sequences. Such arrays can be made on traditional membranes or miniaturized microchips. When total RNA from a cell population is used to probe such an array, it is possible to obtain an expression profile of all genes simultaneously. In EGS expressing cells, specific m
The RNA is cleaved, and in response, its expression should be reduced in array hybridization. Lack of hybridization reveals the targeted sequence. When cells die or cell growth slows down, the expression of many genes changes. This does not present a problem for the application of expression profiles in identifying EGS targets. This is because EGS-mediated effects become evident only after dilution of already existing proteins. This phenotypic lag allows unambiguous determination of the resulting mRNA.

Other Applications of EGS-Mediated Targeting Recovered effective EGS sequences are themselves valuable reagents for attenuating the expression of specific genes. Thus, by inducing the expression of a particular EGS, it is possible to determine, via various phenotypic analyzes, how cells die. These EGS sequences can be administered exogenously to a bacterial population. With a suitable delivery vehicle, the EGS itself can be used as an antibacterial therapeutic.

II. Methods for Identifying Optimal Functional Oligonucleotide Molecules Methods for Identifying Functional Oligonucleotide Molecules, such as Ribozymes, EGS, and Antisense RNA, that Inhibit Expression of Target RNA Molecules Is disclosed. This method identifies functional oligonucleotide molecules by selecting RNA molecules that alter the expression of the fusion transcript, including the RNA molecule sequence of interest from a library of potential functional oligonucleotide molecules. I do. Inhibition of the expression of the fusion transcript prevents expression of the reporter protein. This allows inhibition of the monitored expression by directly or indirectly detecting the expression of the reporter protein. Alternatively, expression can be increased in cells that do not contain the optimal functional oligonucleotide molecule in proportion to the expression of the molecule. Inhibition is achieved by the interaction of nucleic acid molecules involved in the expression of an RNA of interest having a functional oligonucleotide molecule. Ribozymes and E
GS cleaves the fusion transcript and antisense RNA inhibits expression via hybridization to nucleic acid molecules involved in the expression of the fusion transcript.

A vector for use in a functional oligonucleotide molecule assay includes a first reporter gene, a second reporter gene, and a target gene. Reporter gene 1 encodes an RNA molecule of interest and an RNA molecule containing a sequence encoding reporter protein A. The fusion transcript contains the sequence of the RNA molecule of interest in the 5 'region of the transcript, as well as the sequence encoding the first reporter protein in the 3' region of the transcript. The sequences are linked so that the fusion transcript encodes a fusion protein that is a fusion between the protein encoded by the sequence of the RNA molecule of interest and the reporter protein. This configuration is for the desired RNA
Results in expression of the reporter protein, which is dependent on the expression of the reporter protein. Reporter gene 1
Also contains expression sequences necessary for the expression of the gene in a suitable host cell. Reporter gene 2 encodes a different reporter protein B. The vector also encodes a functional oligonucleotide molecule that either specifically targets the RNA of interest or contains a degenerate or partially degenerate target sequence.

[0081] Expression of reporter protein A, or loss of expression, is used to evaluate the effect of a functional oligonucleotide molecule. Reporter protein B expression is used both to detect transformation or transfection of the vector into cells, and as a control for the effect on reporter protein A expression that is not due to cleavage of the RNA molecule of interest. obtain.

[0082] The three components: the first reporter gene, the second reporter gene, and the target gene are preferably contained in a single nucleic acid molecule, but the reporter gene and the target gene are on separate molecules I can do it. When the reporter gene and the target gene are present on separate molecules, it is desirable that the molecule containing the reporter gene be integrated into the host chromosome. This allows for easy maintenance of cell lines containing the appropriate reporter gene, and conveniently tests different sets of vectors encoding different libraries of functional oligonucleotide molecules against the same reporter gene. To be able to

Cells expressing reporter gene 2 are identified by directly or indirectly detecting the presence of reporter protein B. Reporter gene 2 is used to ensure cells containing the vector and generally control any factors that can affect expression. Without such control, loss of expression of reporter gene 1 can be misinterpreted. The level of reporter gene 2 expression measured is not critical. The reporter protein B is preferably an essential protein for the cell, such as a protein that confers antimicrobial resistance or a protein that produces essential nutrients that are not present in the culture medium, to facilitate selection.

The cells that alter the expression of the first reporter gene can be either directly or indirectly measured by measuring the level of reporter protein A expression, or can be cells based on the level of reporter protein A expression. It can be identified by fractionation. The preferred method of detection depends on the nature of the reporter protein used. For example, if a reporter protein is used that produces a detectable signal proportional to the level of expression, the cells can be sorted or collected based on the level of signal produced. Reporter proteins such as β-galactosidase and green fluorescent protein fall into this category. When a cell surface protein is used as reporter protein A, the cell separation techniques described above can be used. Cells can also be classified by FACS when using green fluorescent protein as reporter protein A. This is because this protein produces a fluorescent signal.

[0085] The selection process described above involves isolating the vector from the selected cells and reintroducing the resulting vector into a new cell until cells bearing a uniform population of plasmids can be isolated. Preferably, it can be repeated several times. Following the final isolation of the cells, the vector can be isolated and amplified as described below, and the sequence of the functional oligonucleotide molecule encoded in each preparation of the vector can be determined.

Identification of a Functional Ribozyme or EGS A functional oligonucleotide molecule that is effective in inhibiting reporter gene 1 expression can be identified using any suitable technique. Preferably, the sequence of the functional oligonucleotide molecule is determined by sequencing the vector in the selected cell. Numerous techniques for sequencing vector sequences from clones are known and can be used in published methods. For example, Hirt supernatants of selected cells can be made, and plasmids are extracted from these cells. A preferred method for identifying the sequence of a functional oligonucleotide molecule in an isolated vector is single cell PCR amplification of a functional nucleotide region followed by sequencing. Another preferred method for identifying the sequence of a functional oligonucleotide molecule in an isolated vector is to lyse the cells, extract the plasmid,
Sequencing of the amplified plasmid, which amplifies the plasmid in bacteria and identifies the sequence of functional oligonucleotide molecules associated with the cell population.

A functional oligonucleotide molecule identified using the disclosed methods is targeted to the same site as the functional oligonucleotide molecule.
Can be used to design oligomers.

(Transformation into a Single Plasmid) A first reporter gene encoding GFP as reporter protein A,
A set of vectors encoding a second reporter gene and a target gene that encodes a library of EGSs or ribozyme molecules as functional oligonucleotide molecules was obtained from E. coli. Amplify by growing the mixed population in E. coli. A fixed concentration of plasmid is complexed with a suitable carrier (eg, lipid, calcium phosphate, DEAE dextran), and delivered to mammalian cells. On the day of peak expression (usually day 2), the expression levels of GFP and the second reporter are measured by FACS separation. Expression of the second reporter (eg, CD4) is measured at a wavelength that does not overlap with the GFP fluorescence spectrum. Typically, an antibody conjugated to a fluorescent tag is used and directed against the second reporter protein to monitor the expression level of the second reporter. The antibody is incubated with the cells, excess antibody is washed away, and fluorescence is monitored at a different wavelength than GFP. The ratio of GFP expression to second reporter expression is used as a measure to determine the degree of inhibition of target sequence expression. Cells were lysed, plasmids were extracted, amplified in bacteria, and sequenced to identify EGSs or ribozymes associated with the cell population.

Transformation into Two Individual Plasmids In another embodiment, two individual plasmids are transformed into E. coli. used to transform E. coli. The first plasmid encodes the fusion protein (target-GFP), and the second plasmid encodes the second reporter and a target gene encoding an EGS or ribozyme library. Plasmids encoding the EGS / ribozyme library are grown in bacteria and mixed plasmids are prepared as described above.

A fixed concentration of mixed plasmids (each encoding a separate EGS or ribozyme) is combined with a fixed concentration of target plasmid (Target-G
FP fusion protein). This mixture is complexed with a commercial lipid or calcium phosphate preparation and transfected into 96-well plated cells. At the peak of GFP expression, the GFP-fluorescence level and the expression level of the second reporter are measured, and the ratio of GFP expression to the second reporter is used to determine EGS or ribozyme efficacy. The ratio of EGS to target can be altered to change the level of EGS or ribozyme expression for this target.

This method for identifying functional oligonucleotide molecules is described in (particularly RNa
EGS with enhanced or optimal activity in the induction of cleavage of an RNA molecule of interest by se P) is further understood by reference to the following non-limiting examples.

Example 4: Selection of functional EGS from EGS pool Prokaryotic-based vectors containing the CAT-β-galactosidase fusion protein expressed dark blue colonies. A DNA library encoding the 55EGS sequence complementary to the CAT gene was inserted into the target gene of the base vector.

[0093] Two libraries were made. The first library (Library A)
The EGS was coded following the T7 terminator. The second library is EG
Self-cleaving hammerhead ribozyme to complete the 3 'end of S (self-
clearing hammerhead ribozyme) followed by EGS
Was coded. Expression in the second library was lower, probably due to lower stability.

Library A was plated on X-gal plates. Light blue and dark blue colonies were counted. Light blue colonies were presumed to show EGS-mediated disruption of CAT expression. Expanded colonies from the EGS library provided about 5% light blue colonies as compared to less than 1% light blue colonies in control plates (grown colonies from the library without EGS insertion). This total number of positives was consistent with a few EGS sequences from the primary library that were effective. Therefore, strict grouping of sequences was expected. Therefore, pale blue colonies were picked and replated. The pale blue colonies were preserved. Numerous pale blue colonies, β
-Assayed for galactosidase activity and revealed an 80-90% inhibition of enzyme activity.

[0095] DNA from four pale blue colonies was isolated and sequenced. Each colony encoded the same EGS. This EGS was inserted into the base vector. About 90% inhibition of CAT activity was observed. Qualitatively, less inhibition was confirmed in the second library.

As a control, the opposite experiment was performed. EGS was removed from the base vector. Wild-type levels of β-galactosidase expression were observed. These data indicate that functional EGS can be sorted from a large pool. EGS1
And EGS2 (Guerrier-Takeda et al., Proc. Natl. A
cad. USA 1997) (previously identified, functional EGSs targeting CAT RNA) show little or no activity in these assays.

[0097] Thirty-nine positives from library A were selected in a tertiary screen. D
NA was prepared and sequenced. Only eight of the first 53 EGS sequences were found (EGS # 52 appeared 23 times). As a control, 48 colonies were randomly selected from library A regardless of β-galactosidase expression level. In this random set, 28 EGS sequences were found (one did not appear more than 5 times). Table 4 shows the distribution of the EGS sequence.

[0098]

[Table 4] Therefore, according to these criteria, EGS52, EGS36, and EGS
Twenty were identified as the most frequently selected EGS molecules. These same EGS molecules should be most effective at inhibiting CAT gene expression. To test this, EGS52, EGS36, and EGS20 were expressed in cells expressing the CAT gene, and the cells were tested for chloramphenicol. The results are shown in FIGS. 9a and 9b. Cells with a control plasmid lacking any EGS exhibit chloramphenicol resistance, while all three of the selected EGS molecules have a significant effect on chloramphenicol resistance.

FIG. 10 illustrates further data. Each X represents an EGS that is complementary to a specific sequence of the mRNA transcript of the CAT gene. X indicates that EGS was restored in this experiment and is representative of the library. The same EGS was found from a number of different clones selected by the clones pale blue or bluish white. Equally important, most EGS randomly collected from non-indexed plates differed from active EGS selected from indexed plates.
The most likely site in the mRNA molecule was inaccessible to apparently complementary EGS action. However, certain mRNA sites were identified as being accessible to EGS.

The data in FIG. 10 also shows the relative potency of each EGS by knocking down target gene expression. Using each set of populations, β-
4 shows the observed inhibition of galactosidase activity. EGS was most effective
-52, resulting in a 92% decrease in β-galactosidase activity.
Three separately selected EGSs (EGS-31, EGS-20, and EGS-3
6) reduced β-galactosidase activity by 20% and 30%, respectively.

The documents and materials cited herein are specifically incorporated by reference.

One skilled in the art can ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the following claims.

[Brief description of the drawings]

FIG. 1 is a diagram of the mechanism of action of EGS in cleavage of target mRNA.

FIG. For use in identifying essential genes in E. coli,
FIG. 2 is a diagram of the structure used in the example for creating an RA-N11 library (13-mer EGS, N ₁₁ CCACCA).

FIG. 2b is a schematic diagram of the pARANx (A / K) EGS transcription vector.

FIG. 2c is a schematic of the pARAN vector and library construct.

FIG. 1 is a flow chart of a method for inducing and selecting an AraN11 library in E. coli on a solid medium. Only cells expressing the appropriate EGS survive and are amplified.

FIG. 4 is a flowchart of an ampicillin enrichment method for selecting an EGS target essential gene.

FIG. 5a is a graph showing the growth of ampicillin selection clone E8-1 with arabinose (circles) and without arabinose (squares).

FIG. 5b is a graph showing the growth of controls with arabinose (circles) and without arabinose (squares).

FIG. 6 shows a library of EGSs of different sizes (9-mers, 10-mers, 11-mers).
-Mer, and 12-mer).

FIG. 7 shows an example of a vector for use in a method for identifying a functional oligonucleotide molecule containing EGS, ribozyme, and antisense. Reporter gene 1 is composed of target RNA and reporter protein (
It encodes a fusion transcript consisting of RNA encoding reporter protein A). The fusion transcript encodes a fusion protein consisting of the protein encoded by the RNA of interest and reporter protein A. Reporter gene 2
Encodes reporter protein B. The target gene encodes one of the functional oligonucleotide molecules to be tested.

FIG. 8 shows an example of a vector for use in a method for identifying a functional oligonucleotide molecule. Reporter gene 1 encodes a fusion transcript (reporter protein A) consisting of RNA encoding chloraphenicol acetyltransferase (CAT) and RNA encoding β-galactosidase. This fusion transcript encodes a fusion protein consisting of CAT and β-galactosidase. Reporter gene 2 is an ampicillin resistance gene. The target gene is an EGS cassette that encodes one of a library of 50 EGS molecules, each of which is targeted to a different site in the CAT RNA.

FIG. 9a is a graph of cell culture density (A ₆₀₀ ) versus time (minutes) for cells in the presence of 5 μg / ml chloramphenicol. The cells contained a non-EGS encoding vector (circle), EGS36 similar to the vector shown in FIG.
, A vector encoding EGS20 (an inverted triangle),
Alternatively, vectors (diamonds) encoding both EGS52s were included.

FIG. 9b is a graph of cell culture density (A ₆₀₀ ) versus time (minutes) for cells in the presence of 25 μg / ml chloramphenicol. The cells contained a non-EGS encoding vector (circle), EGS3 similar to the vector shown in FIG.
Vector encoding 6 (triangle), vector encoding EGS20 (reverse triangle)
, Or both EGS52-encoding vectors (diamonds).

FIG. 10 shows EGS-CAT-1, EGS-CAT-2, EGS-20, and EGS.
FIG. 4 is a graph of the percentage inhibition of chloramphenicol acetyltransferase (CAT) activity by -31, EGS-36 and EGS-52. Each X is C
EGS complementary to the specific sequence of the mRNA transcript of the AT gene. This multiple X indicates that the EGS was restored and the library was represented in this experiment. Circles indicate the relative frequency at which each EGS knocks down the expression of the targeted gene.

──────────────────────────────────────────────────続 き Continuation of front page (81) Designated country EP (AT, BE, CH, CY, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE ), AU, CA, JP (72) Inventors Quint, Thomas Jay. United States Maryland 20817-2727, Bethesda, Still Spring Court 8313

Claims (40)

[Claims] 1. A nucleic acid molecule comprising a first reporter gene, a second reporter gene, and a target gene, wherein the first reporter gene is a protein of interest or a part thereof and a first reporter protein And the second reporter gene encodes a second reporter protein; and the protein of interest or a portion thereof is encoded by RNA of interest; and the target gene Encodes a functional oligonucleotide molecule selected from the group consisting of EGS, ribozyme, and antisense RNA, and the functional oligonucleotide molecule encodes a site in the RNA molecule of interest or an RNA molecule of interest. At a site in a portion of the first reporter gene Matoka is is, the nucleic acid molecule. 2. The nucleic acid molecule according to claim 1, wherein the functional oligonucleotide molecule encoded by the target gene is EGS. 3. The nucleic acid molecule of claim 1, wherein said nucleic acid molecule is a vector that is expressed and replicated in prokaryotic cells. 4. The nucleic acid molecule of claim 1, wherein said vector is expressed and replicated in eukaryotic cells. 5. The set of nucleic acid molecules according to claim 1, wherein each nucleic acid molecule in the set is the same except for the encoded functional oligonucleotide molecule, and in the set. The functional oligonucleotide molecule encoded in the nucleic acid molecule may be directed to a partially or completely non-overlapping oligonucleotide sequence site on the RNA molecule of interest or to a first reporter gene encoding the RNA molecule of interest. A set of nucleic acid molecules targeted to partially or completely non-overlapping oligonucleotide sequence sites in said portion. 6. The set of claim 5, wherein the functional oligonucleotide molecule encoded by the target gene on each of the nucleic acid molecules in the set is EGS. 7. The set of claim 6, wherein said functional oligonucleotide molecule encoded by said target gene on each said nucleic acid molecule in said set is antisense. 8. The set of claim 5, wherein said functional oligonucleotide molecule encoded by said target gene on each said nucleic acid molecule in said set is a ribozyme. 9. The set of claim 5, wherein said set comprises five or more nucleic acid molecules. 10. The set of claim 5, wherein said set comprises a library of partially or completely randomized oligonucleotide molecules. 11. The method according to claim 11, wherein the expression of the RNA of interest is reduced or the RNA of interest is reduced.
A method for identifying a functional oligonucleotide molecule that inactivates a cell, comprising the steps of: (a) introducing a set of nucleic acid molecules into a cell, wherein after introduction of the nucleic acid molecule, each cell A reporter gene, a second reporter gene, and a target gene, wherein the first reporter gene encodes a fusion protein comprising the protein of interest or a portion thereof and a first reporter protein; and the second reporter gene The gene encodes a second reporter protein and the protein of interest or a portion thereof
A, and wherein the target gene encodes a functional oligonucleotide molecule comprising a target sequence selected from the group consisting of ribozyme, EGS, and antisense, and each nucleic acid molecule in the set is encoded The same except for a functional oligonucleotide molecule, and where the functional oligonucleotide molecule in the set encoded in each nucleic acid molecule is partially or completely non-overlapping on the RNA of interest, or Being targeted to partially or completely non-overlapping sites in the portion of the first reporter gene encoding the RNA molecule of interest, (b) expressing the second reporter protein; Identifying these cells from step (a) showing reduced expression of said first reporter protein, And (c) identifying the functional oligonucleotide molecule encoded by the nucleic acid molecule present in the cell that expresses the second reporter protein and exhibits reduced expression of the first reporter protein. The method, wherein the functional oligonucleotide molecule identified herein is a functional oligonucleotide that reduces the expression of the RNA of interest. 12. The method of claim 11, wherein the cell that expresses the second reporter protein and exhibits reduced expression of the first reporter protein is the cell from step (a). Screening for cells that express a second reporter protein, or selecting cells that express the second reporter protein from the cells from step (a); and Screening the cells that express, or selecting from the cells that express the second reporter protein, cells that exhibit reduced expression of the first reporter protein. The described method. 13. The method according to claim 11, wherein said cells are bacteria. 14. The method of claim 11, wherein said cells are eukaryotic cells. 15. The method of claim 14, wherein said cells are fungi. 16. The method of claim 14, wherein said cells are mammalian cells. 17. The method according to claim 16, wherein the cells are Pseudomonas aeruginos
a, Mycobacterium tuberculosis, Hemophili
rus influenzae, Staphylococcus aureus
, Mycoplasma pneumoniae, Escherichia c
oli, Streptococcus pneumoniae, Neisser
ia gonorrhoeae, Streptococcus virida
ns, Streptococcus pyogenes, Proteus mi
rabilis, Proteus vulgaris, Salmonella
typhimurium, Shigella dysentereae, Clo
stridium difficile, Klebsiella pneumo
niae, Candida albicans, Aspergillus fl
avus, Aspergillus fumagatas, and Histop
2. The composition of claim 1, wherein the composition is selected from the group consisting of:
2. The method according to 1. 18. A functional oligonucleotide molecule is identified by the step of identifying the nucleic acid molecule present in the cell that expresses the second reporter protein and exhibits reduced expression of the first reporter protein. The method of claim 11. 19. The method of claim 18, further comprising determining a partial sequence of said target gene encoding said target sequence. 20. A functional oligonucleotide molecule for reducing expression of a target RNA obtained by the method according to any one of claims 10 to 19. 21. An essential or functional gene comprising mixing an external guide sequence library targeting cleavage by RNase P RNA molecules transcribed from the gene with cells containing the gene having a known nucleotide sequence. A method for identifying a gene. 22. The method according to claim 19, wherein the external guide sequence is a prokaryotic RNase P.
22. The method of claim 21, which targets cleavage of the NA molecule, and wherein said cell is a bacterial cell. 23. The method according to claim 23, wherein the external guide sequence is an Eukaryotic RNase P
22. Targeting the cleavage of a NA molecule, and wherein said cell is a eukaryotic cell.
The method described in. 24. The method according to claim 24, wherein the external guide sequence is an eukaryotic RNase P
22. The method of claim 21, which targets cleavage of the NA molecule, and wherein said cell is a fungal cell. 25. The cell according to claim 15, wherein the cell is Pseudomonas aerugino.
sa, Mycobacterium tuberculosis, Hemoph
ilus influenzae, Staphylococcus aureu
s, Mycoplasma pneumoniae, Escherichia
coli, Streptococcus pneumoniae, Neisse
ria gonorrhoeae, Streptococcus virid
ans, Streptococcus pyogenes, Proteus m
irabilis, Proteus vulgaris, Salmonella
typhimurium, Shigella dysentereae, Cl
ostridium difficile, Klebsiella pneum
oniae, Candida albicans, Aspergillus f
lavus, Aspergillus fumagatas, and Histo
22. The method according to claim 21, wherein the method is selected from the group consisting of plasmascapsatum. 26. The library comprising all possible combinations of nucleotide sequences of a length effective to specifically direct RNase P cleavage at any cleavage site in any RNA in a prokaryotic cell. 22. The method of claim 21 constructed to include. 27. The external guide sequence comprises a N ₁₀ ~ ₁₃ oligomers The method of claim 26. 28. The external guide sequence comprises N ₁₀ ~ ₁₅ oligomer, and wherein said cell is a fungus, The method of claim 21. 29. The external guide sequence comprises N ₁₀ ~ ₁₈ oligomer, and wherein said cell is a mammalian cell, the method according to claim 21. 30. The method of claim 21, wherein said external guide sequence is on a vector capable of being expressed and replicated in said cell. 31. The vector has an external guide sequence that induces cleavage of a reporter gene or a selection gene that can be used to screen cells harboring the vector, or an RNA encoding an essential or functional gene. Item 30. The method according to Item 30, 32. The method of claim 21, wherein said RNA is selected from the group consisting of viral RNA, messenger RNA, transfer RNA, and ribosomal RNA. 33. RNA encoded by an essential or functional gene
Further comprising selecting cells containing an external guide sequence that induces cleavage of
The method of claim 31. 34. The cell is first replicated and then exposed to an agent that kills or alters the phenotype of the cell that expresses an external guide sequence that cleaves an RNA encoding an essential or functional gene. 34. The method according to claim 33. 35. The method according to claim 26, wherein the cells expressing the vector are exposed to an agent that kills or alters the phenotype of cells expressing an external guide sequence that induces cleavage of RNA encoding an essential or functional gene. 4. Amplified before
4. The method according to 4. 36. The method of claim 33, further comprising identifying a sequence of said external guide sequence that induces phenotypic killing or alteration of said cell. 37. The method of claim 36, further comprising determining the sequence of a gene whose encoded RNA is cleaved by RNase P when targeted by said external guide sequence. 38. The method of claim 37, wherein the sequence of the gene is determined by analysis of a plurality of external guide sequences that kill or alter the phenotype of the cell when exposed to a predetermined agent. . 39. An isolated gene sequence identified by the method of claims 21-38. 40. A library of external guide sequences for use in the method of claims 21-38.