Screening and use of agents which block or activate intein splicing utilizing natural or homologous exteins

Abstract

In accordance with the present invention, there are provided selection systems and methods for screening for agents that control splicing of inteins in their native host protein (extein) or in homologous exteins. Specifically, there are provided positive genetic selection systems for the screening of agents which inhibit or activate protein splicing which comprise: a host cell containing a chromosomal gene encoding either a drug-resistant form of a target enzyme or a wild-type target enzyme, and a plasmid-borne gene encoding either a drug-sensitive form of the target enzyme, which is dominantly cytotoxic upon interaction with the drug, or a dominantly cytotoxic form of the target enzyme. In these systems the plasmid-borne gene contains an intein, and the inhibition or activation of splicing of the dominant cytotoxic form of the target enzyme by a given reagent results in the survival or death of the host cell. More specifically, positive genetic selection systems which utilize the M. xenopi GyrA intein or M. tuberculosis DnaB helicase intein are provided. Similar reporter systems utilizing native or homologous exteins and systems utilizing controllable inteins are provided, as are methods of controlling in vivo expression of proteins by modulating protein splicing with inhibiting or activating agents, and methods of controlling the delivery of proteinaceous drugs in vivo by modulating protein splicing.

Description

BACKGROUND OF THE INVENTION

[0002] The present invention relates to the screening for and use of agents which either inhibit or activate protein splicing of inteins (IVPS). Specifically, disclosed herein is the development of 2 specific reporter systems for Gyrase A and DnaB inteins. Agents screened for in accordance with the present invention can be used to control protein splicing for any purpose, in vivo or in vitro, including antimicrobial activity of organisms containing inteins in essential genes. More specifically, the present invention relates to the use of inteins expressed in modified or unmodified native protein splicing precursors or homologous extein systems to screen for mutations that modulate splicing or agents that inhibit or activate splicing. The present invention improves on current reporter systems used to screen for agents that can control splicing by using a modified or unmodified native precursor or precursor homolog in order to take advantage of the more native intein active site formed by natural precursors or inteins in homologous exteins, since agents that are derived from non-native precursors may not have the identical selected activity on native precursors.

[0003] Production of mature proteins involves the flow of information from DNA to RNA to protein. Precise excision of DNA and RNA elements which interrupt that information has been previously described (M. Belfort, Annu. Rev. Genet. 24:363 (1990); T. R. Cech, Annu. Rev. Biochem. 59:543 (1990); Hunter et al., Genes Dev. 3:2101 (1989)). More recently, evidence for the precise excision of intervening protein sequences has also been described for the TFPI allele from Saccharomyces cerevisiae (Hirata et al., J. Biol. Chem. 265:6726 (1990); Kane et al., Science 250:651 (1990)) and the recA gene from Mycobacterium tuberculosis (Davis et al., J. Bact. 173:5653 (1991); Davis et al., Cell 71:1 (1992)). Each of these genes contains internal in-frame peptide segments which must be removed to produce the mature protein. Expression of Tfp1 and RecA each results in two peptides: one representing the intervening protein sequence (IVPS) and the other the ligated product of the external protein sequences (EPS). In 1994, the terms “intein” and “extein” were adopted in place of IVPS and EPS, respectively (Perler, et al., Nucleic Acids Res. 22:1125-1127 (1994)). This post-translational processing event has been termed “protein splicing”. Similarly, the “Vent”® DNA polymerase gene from the hyperthermophilic archaeon Thermococcus litoralis contains two in-frame IVPS (Perler, et al., PNAS 89:5577 (1992)) and the DNA polymerase gene from the hyperthermophilic archaeon Pyrococcus species GB-D contains one intein (Xu, M., et al., Cell 75, 1371-1377 (1993)).

[0004] Over 80 inteins have been identified in bacteria, archaea and eucarya (Perler, F. B., et al. Nucleic Acids Res 25, 1087-93 (1997), Dalgaard, J. Z., et al., J Comput Biol 4, 193-214 (1997), Pietrokovski, S., Protein Sci. 7, 64-71 (1998) and Perler, F. B. Nucleic Acids Res. 27, 346-47 (1999). Four inteins have been found in Mycobacterium leprae (Davis, E. O., et al., EMBO J. 13, 699-703 (1994) and Smith, D. R., and et al. Genome Res 7, 802-19 (1997)) and three inteins in Mycobacterium tuberculosis (Cole, S. T., et al. Nature 393, 537-44 (1998)). One intein has been found in Candida tropicalis (Gu, et al., J. Biol. Chem., 268(10):7372-7381 (1993)).

[0005] Controllable IVPS (CIVPS) and methods for using the same to modify, produce and purify target proteins has been described (Comb et al., U.S. Pat. No. 5,496,714, issued Mar. 5, 1996; Comb et al., U.S. Pat. No. 5,834,247, issued Nov. 10, 1998). Methods for using inteins to screen for peptides (or derivative, analogic or mimetic thereof) or any agent that can enter cells to block or activate splicing of a natural or experimental reporter protein have also been described (U.S. Pat. No. 5,834,247, supra.. at Example 17). These methods specifically describe the screening of peptides using mycobacterial inteins as targets. The preparation of an in vivo peptide library utilizing chicken α-spectrin is also described.

[0006] While a general method of screening for antimicrobial agents using the M. tuberculosis RecA intein in a thymidylate synthetase (TS) reporter system has been described (Belfort, U.S. Pat. No. 5,795,731, issued Aug. 18, 1998), this system suffers from several limitations. Importantly, several studies of protein splicing in foreign contexts (such as the Belfort system) indicate that intein splicing is more efficient in the native extein than in foreign exteins (Xu, EMBO J. 13:5517-5522 (1994), Xu, EMBO J. 15:5146-5153 (1996), Telenti, J. Bacteriol. 179:6379-6382 (1997), Chong J. Biol. Chem, 273:10567-10577 (1998), Liu, FEBS Lett. 408:311-314 (1997), Wu, Biochim. Biophys. Acta 1387:422-432 (1998B), Nogami Genetics, 147:73-85 (1997), Kawasaki J. Biol. Chem., 272:15668-15674 (1997), Derbyshire, Proc. Natl. Acad. Sci USA, 94:11466-11471 (1997), Southworth, BioTechniques 27:110-121 (1999), FIG. 7)). For example, the use of foreign exteins yields temperature-dependent splicing of the Psp-GBD Pol, Mxe GyrA and Synechocystis DnaB inteins (Xu, EMBO J. 13:5517-5522 (1994), Xu, EMBO J. 15:5146-5153 (1996), Telenti, J. Bacteriol. 179:6379-6382 (1997), Chong J. Biol. Chem, 273:10567-10577 (1998), Liu, FEBS Lett. 408:311-314 (1997), Wu, Biochim. Biophys. Acta 1387:422-432 (1998B), Nogami Genetics, 147:73-85 (1997), Kawasaki J. Biol. Chem., 272:15668-15674 (1997) and Southworth, BioTechniques, 27:110-121 (1999), and FIG. 7).

[0007] While not wishing to be bound by theory, it is believed that such inefficient protein splicing in the foreign extein context occurs because the flanking extein is, in effect, the substrate of the intein. It is, therefore, likely that the intein may exhibit substrate specificity like all other enzymes. The substrate specificity of the intein limits acceptable extein sequences, hence the native extein sequence is the optimal substrate, whereas foreign extein sequences may not be acceptable substrates at all. For example, studies of the Sce VMA and Mxe GyrA inteins indicate that thiol induced N-terminal splice junction cleavage and splicing are, to varying extents, dependent on the single extein residue preceding the intein (Chong, J. Biochem. 273:10567-10577 (1998), Southworth, BioTechniques, 27:110-121 (1999)). Other extein residues have also been shown to affect splicing of the Sce VMA intein (Nogami Genetics, 147:73-85 (1997), Kawasaki J. Biol. Chem., 272:15668-15674 (1997)).

[0008] Additionally, exteins may affect the packing at the intein active site, or global folding of the intein and/or precursor, hence the use of a foreign extein may result in improper folding of the intein or precursor and inefficient or no splicing. Moreover, expression of an extein gene that naturally contains an intein in a foreign host, for example E. coli or yeast, may not be efficient (Perler et al. Proc. Natl. Acad. Sci. USA 89:5577-5581 (1992) and Hodges, et al., Nucleic Acid Res. 20:6153-6157 (1992)), whereas expression of the homologous endogenous extein is likely to be more efficient. For example, the Mycobacterium leprae RecA intein fails to splice in E. coli, while it splices in M. leprae (Davis, et al., EMBO J., 13:699-703 (1994)). It is possible that the M. leprae RecA intein would splice in E. coli RecA, although that has yet to be tested. In another example, the Synechocystis sp. strain PCC6803 DnaB gene, containing an intein, was unclonable in E. coli (Wu, et al., Proc. Natl. Acad. Sci. USA 95:9226-9231 (1998)). The M. leprae GyrA precursor did not splice efficiently in E. coli and was mostly insoluble, while the homologous Mxe GyrA intein spliced efficiently in E. coli GyrA.

[0009] Additionally, the use of homologous exteins would eliminate, in many instances, the need to introduce silent mutations in the reporter gene in order to insert the desired intein (see Belfort, supra., Comb, supra, Example 17). Homologous exteins may have naturally-occuring, conserved restriction enzyme sites that would allow the cloning of the intein into the homologous extein or they may have enough extein similarity to allow insertion of the intein into the homologous extein by recombination. Such systems also eliminate the need for an exogenous reporter gene, since innate extein properties of the native extein may be used for selection. Alternatively, the native extein may be mutated, either de novo or based on mutations in similar extein genes, to make the extein into a selectable marker or reporter gene.

[0010] Accordingly, the most desirable intein splicing systems would be those systems which express an intein from one organism in the homologous extein from the foreign host organism used for expression or to express the native precursor gene in a suitable foreign host organism.

[0011] Such intein systems would not only be useful in the screening of antimicrobial agents which inhibit intein splicing within a reporter gene (as described in Belfort, supra, Comb, supra..), but as controllable targets to direct expression of an extein product. Agents, for example peptides, that block intein splicing may be used to limit the expression of an extein in such systems. The suppression of such expression may be highly useful in the drug delivery context, where, for example, one wishes to turn on an enzyme which is active in killing cancer cells, or by delivering needed activity, for example insulin.

[0012] Similarly, such intein systems may utilize splicing-incompetent inteins to screen for agents with the ability to activate splicing.

SUMMARY OF THE INVENTION

[0013] In accordance with the present invention, there is provided selection systems and methods for selecting or screening for mutations or agents that control the splicing of inteins which comprise use of the intein's native host protein (extein) or a homologous extein in any host organism (FIG. 8). Specifically, in one preferred embodiment, there is provided a positive genetic selection system for the screening of agents which inhibit protein splicing which comprises: a host cell which contains (1) a copy of the extein gene (either episomal, chromosomal or synthetic) gene encoding a mutant or naturally drug-resistant form of a target enzyme (which as used herein includes not only enzymes, but proteins, peptides or the like), and (2) a wild-type or mutant form of the extein gene (either episomal, chromosomal, or synthetic) encoding a drug-sensitive form of the target enzyme which is dominantly cytotoxic upon interaction with the drug, wherein the gene encoding the drug-sensitive form of the target enzyme contains an intein, and wherein the inhibition of splicing of the drug-sensitive form of the target enzyme by a given reagent results in the survival of the host cell in the presence of the drug. In one specific embodiment, a positive genetic selection system which utilizes the M. xenopi GyrA intein is provided. This system is also applicable to any GyrA intein inserted at the same or different site in the GyrA extein gene.

[0014] In accordance with another preferred embodiment, there is provided a similar positive genetic selection system for the screening of agents which inhibit protein splicing which comprisesa host cell which contains (1) a copy of the extein gene (either episomal, chromosomal or synthetic) encoding a wild type form of a target enzyme, and (2) a gene encoding a dominant cytotoxic form of the target enzyme (either episomal, chromosomal or synthetic) wherein the gene encoding the dominantly and cytotoxic form of the target enzyme contains an intein, and wherein the inhibition of splicing of the cytotoxic form of the target enzyme by a given reagent results in the survival of the host cell. In one particularly preferred embodiment, a positive genetic selection system which utilizes the M. tuberculosis DnaB helicase intein is provided. This positive selection system may also employ any DnaB intein inserted at the same or different site in the DnaB extein gene. Similar systems and methods of screening for agents that activate protein splicing are also provided, as are reporter systems utilizing native or homologous exteins and systems utilizing inteins.

[0015] Also provided by the present invention are methods of controlling in vivo expression of proteins by modulating protein splicing with inhibiting or activating agents. Similar methods of controlling the delivery of proteinaceous drugs in vivo by modulating protein splicing are also provided.

[0016] As used herein, “agent” includes, but is not limited to, a peptide (free or displayed on a scaffold such as chicken α-spectrin), a peptide derivative, analogic or mimetic, a natural product or a synthetic molecule.

BRIEF DESCRIPTION OF THE DRAWINGS

[0017] FIG. 1 is a diagram depicting a protein splicing precursor and products and alternative names for each element or part thereof.

[0018] FIG. 2 is a diagram depicting a general scheme for the selection of peptides which block intein splicing of a dominant lethal suicide gene in vivo.

[0019] FIG. 3A is a diagram depicting the irreversible blocking of DNA replication by E. coli GyrA interaction with a drug (ofloxacin).

[0020] FIG. 3B is a diagram depicting a Mxe GyrA intein-splicing system for the selection of agents which block intein splicing. The splicing of the Mxe GyrA intein out of the homologous Eco GyrA extein produces a drug sensitive, wild-type Eco GyrA which, in the presence of ofloxacin, forms an irreversible covalent poison complex during replication that kills the cell, despite the presence of the chromosomal mutant GyrA which is drug resistant.

[0021] FIG. 3C is a diagram depicting a Mxe GyrA intein-splicing system for the selection of agents which block intein splicing: the blocking of splicing of the Mxe GyrA intein out of the homologous Eco GyrA extein results in the expression of the inactive drug sensitive Eco GyrA and the chromosomal mutant GyrA, which is ofloxacin-resistant. Hence, in the absence of an active drug sensitive Eco GyrA (due to the blockage of splicing,) the host grows.

[0022] FIG. 3D is an amino acid sequence comparison of part of the E. coli GyrA (SEQ ID NO:1) and M. xenopi GyrA (SEQ ID NO:2) sequences, indicating that the GyrA exteins are very similar in amino acid sequence, especially at the intein insertion site marked by the arrow.

[0023] FIG. 3E is a gel indicating efficient splicing of the Mxe GyrA intein in the homologous Eco GyrA extein. The position of Eco GyrA is indicated by the solid black box and the position of the precursor comprising the Mxe GyrA intein in Eco GyrA is indicated by the black and white boxed marker with the white box indicating presence of the intein.

[0024] FIG. 4A-1 is a diagram depicting the intersection of DnaB with DnaC which is required to load DnaB onto the DNA replication machinery.

[0025] FIG. 4A-2 is a diagram depicting the sequestration of DnaC by a mutant E. coli DnaB helicase which leads to disrupted DNA replication and cell death.

[0026] FIG. 4B is a diagram depicting a Mtu DnaB helicase intein-splicing system for the selection of agents which block intein splicing: the splicing of the Mtu DnaB helicase intein out of a dominant lethal mutant Mtu DnaB helicase extein produces mutant Mtu DnaB helicase which sequesters Eco DnaC and poisons replication despite the presence of the chromosomal Eco DnaB helicase, as a result, the host dies.

[0027] FIG. 4C is a diagram depicting a Mtu DnaB helicase intein-splicing system for the selection of agents which block intein splicing: the blocking of splicing of Mtu DnaB helicase intein out of a dominant lethal mutant Mtu DnaB helicase extein prevents the sequestration of Eco DnaC; chromosomal Eco DnaB helicase is expressed and the host grows.

[0028] FIG. 4D is an amino acid sequence comparison of part of the E. coli DnaB helicase (SEQ ID NO:3) and M. tuberculosis DnaB helicase (SEQ ID NO:4) sequences indicating that the amino acid sequences are very similar and that the site in E. coli DnaB that was mutated to make it cytotoxic is conserved in M. tuberculosis DnaB (first on larged sequence) and that the intein insertion site is also conserved in E. coli DnaB (marked by the arrow).

[0029] FIG. 5A is a diagram depicting the production of a combinatorial peptide library using chicken α-spectrin and the screening of these peptides for those that block Mxe GyrA helicase intein splicing. “aa” represents a portion of spectrin which can be randomized. The spectrin scaffold is represented by a trapazoid and the different amino acid sequences by various other shapes. If the spectrin binds to the precursor, splicing is blocked. The system has three components: a host cell expressing T7 RNA polymerase, the spectrin library and the intein plus GyrA gene. The latter two genes are present on a single plasmid under control of T7 RNA polymerase promoters.

[0030] FIG. 5B-1 is a flow diagram indicating multiple-round selection of combinatorial peptides that block Mxe GyrA.

[0031] FIG. 5B-2 is a flow diagram indicating Mtu DnaB helicase intein splicing.

[0032] FIG. 5C is a gel indicating that peptides p814-818 (SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9 and SEQ ID NO:10) block splicing of Mtu DnaB in E.coli. This is a Western block using anti-T7 tag antibody to detect the T7 tag at the N-terminus of each DnaB protein. In p815rev, the selected blocking peptide sequence in α-spectrin has been replaced with the wild type spectrin sequence to demonstrate that inhibition of splicing is due to the selected peptide sequence. The bands and markers on the right represent the precursor, a putative C-terminal cleavage product and the spliced DnaB exteins, respectively from top to bottom of the Western Blot. Size markers are in lane M.

[0033] FIG. 6 is a diagram depicting the production of a combinatorial peptide library using chicken α-spectrin and the screening of these peptides for those that block Mtu DnaB intein splicing in Mtu DnaB. E. coli GyrA gyrase. “aa” represents a portion of spectrin which can be randomized. The spectrin scaffold is represented by a trapazoid and the different amino acid sequences by various other shapes. If the spectrin binds to the precursor, splicing is blocked. The system has three components: a host cell expressing T7 RNA polymerase, the spectrin library and the intein plus GyrA gene. The latter two genes are present on a single plasmid under control of T7 RNA polymerase promoters.

[0034] FIG. 7 is a table showing the effect of the single amino acid preceding the Mxe GyrA intein in a heterologous extein context on splicing and N-terminal cleavage by DTT.

[0035] FIG. 8 is a flow chart for choosing native precursors, homologous exteins or heterologous exteins to develop a selection or reporter system for testing agents that inhibit or activate splicing of an intein.

[0036] FIG. 9 is a summary of the various methods of selecting agents that inhibit or activate protein splicing. Each system is based on a merodiploid cell containing an intein plus and an intein minus extein gene.

[0037] FIG. 10 depicts the scheme for creating random mutations in the Mxe gyrA intein by error prone PCR of the intein followed by cloning of the mutated intein genes into the E.coli Mxe gyrA extein.

[0038] FIG. 11 depicts the selection scheme based on the GyrA selection described in Example I in which the presence of a drug kills cells where the intein has spliced. Clones that do not splice at Temperature 1 grow, while replica plated clones that splice at lower Temperature 2 do not grow.

[0039] FIG. 12 show cell lysates from wild type or mutated intein clones were electrophoresed in SDS acrylamide gels. A temperature sensitive clone grown at 37° C. (labeled ‘A’) fails to splice, while the wild type intein clone splices (labeled ‘WT’). Wild-type levels of splicing are observed in the mutant clones (1-6) when shifted to 16° C. overnight.

[0040] FIG. 13 illustrates the Mxe GyrA intein sequence (SEQ ID NO:46) with mutations found in the temperature sensitive splicing clones indicated below the wild-type residue.

[0041] FIG. 14 illustrates the positioning of the mutations in the temperature sensitive splicing clones on the Mxe GyrA intein 3-D structure. The two panels depict opposite sides of the Mxe GyrA intein with a single Alanine preceding the intein. Double amino acid change indicates that the clone had more than one mutation.

DETAILED DESCRIPTION

[0042] The present invention is directed to methods of selecting or screening for mutations or agents that block or activate protein splicing of inteins using natural precursors or by inserting inteins in homologous extein genes. These mutations or agents can be used to activate or keep inactive enzymes in vivo or in vitro for pharmacological, chemotherapeutic, or biotechnological purposes. In contrast, these same methods can be used to select agents that block or activate splicing in a non-homologous extein if no genetic selection system or screen can be generated for the native extein protein.

[0043] The in vivo control of protein splicing mediated by a blocking or activating peptide, or other agent that can enter a cell, acting on a controllable intervening protein sequence (CIVPS) has been described (U.S. Pat. No. 5,834,247, supra. at Example 17). In the present invention, it should be noted that a non-controllable IVPS, or intein, is used to identify agents that will convert the IVPS into a CIVPS. The blocking of such splicing activity by specific agents such as peptides or natural products, and analogues thereof, is particularly useful in combating pathogens such as Mycobacterium tuberculosis, Mycobacterium leprae, Mycobacterium avium, or Candida tropicalis by blocking the splicing of essential proteins in those organisms.

[0044] Approximately 97 inteins have been identified and are available from public databases (Perler, Nucleic Acids Res. 22:1125-1127 (1994), Perler, Nucleic Acids Res. 27:346-347 (1999), Pietrokovski, S., Protein Sci., 7:64-71 (1998) and Dalgaard, et al., J. Comput. Biol., 4:193-214 (1997). Sequencing projects of small prokaryotic genomes (e.g., Mycobacterium tuberculosis, Mycobacterium leprae, and Methanococcus jannaschii) already account for the majority of published intein sequences. Host genes of these inteins are often involved in such essential cellular functions as DNA replication, expression, or various metabolic processes (compiled in: Perler, Nucleic Acids Res. 25:1087-1093 (1997), Perler, Nucleic Acids Res. 27:346-347 (1999), Pietrokovski, S., Protein Sci., 7:64-71 (1998) and Dalgaard, et al., J. Comput. Biol., 4:193-214 (1997). Hence, the disruption of these essential functions via the blocking of intein splicing by peptides, or other agents, represents a means by which to screen for anti-microbial and anti-pathogenic agents.

[0045] Generally, a positive selection system consists of a gene that is detrimental to a host organism depending on the growth media or the host strain genetic background. The gene product is toxic for the cell, inhibiting growth or killing the host unless the gene product is inactivated. In the context of a protein splicing genetic system, a positive selection system is defined as a system that allows selection against the splicing of an intein inserted in-frame into a host gene (see FIG. 2). If splicing occurs in the precursor protein containing the intein, the cytotoxic host protein will be active and inhibit cell growth or kill the cell; if splicing is disrupted the cytotoxic host protein will be inactive and allow cell growth. The same description applies to reporter systems where detection of the host protein is scored, rather than selection for organism viability. Many reporter genes are known, the most common example is the Blue/white screen involving β-galactosidase function on X-gal to produce a blue color.

[0046] In accordance with one embodiment of the present invention, there is provided a positive selection system for identifying agents which block or activate protein splicing which comprises a host cell which contains (1) a copy of the extein gene (either episomal, chromosomal or synthetic) encoding a mutant or naturally drug-resistant form of a target enzyme; and (2) a wild type or mutant form of the extein gene (either episomal, chromosomal or synthetic) encoding the target enzyme that is sensitive to the drug, into which is inserted an intein, wherein the spliced form of the intein-containing target enzyme is toxic to the host organism upon interaction with a certain drug. In this system, the host cell so transformed will express the drug sensitive enzyme if the intein is properly spliced, resulting in reduced viability of the organism because the spliced product is dominantly lethal or cytotoxic to the host organism, despite the fact that the drug-resistant homologous gene is also expressed. The intein-les copy of the gene is required to maintain viability in cells in which splicing of the plasmid borne extein gene is blocked.

[0047] In one preferred embodiment, a plasmid-encoded, drug sensitive gene is the naturally occurring intein/extein precursor or its homolog containing an intein, which intein may be either naturally occurring or inserted, and drug sensitivity may be naturally occurring in this precursor gene or introduced by mutation in the extein portion of the gene. This system results in death of all cells where splicing occurs and thus provides a system for selecting for mutations, drugs, chemicals, peptides, etc. which block splicing in vivo since cell viability requires blockage of splicing.

[0048] In a particularly preferred embodiment, the Mycobacterium xenopi GyrA intein (Mxe GyrA) (SEQ ID NO:11) (Telenti et al., J. Bacteriol. 179:6378-6382 (1997) is inserted into the homologous extein of E. coli GyrA (see FIG. 3D). In E. coli, GyrA is an essential gene that encodes for the A subunit of the E. coli gyrase hetero-tetramer protein complex. E. coli gyrase is a type II topoisomerase involved in DNA relaxation at the origin of replication of the bacterial chromosomal DNA (Swanberg and Wang, J. Mol. Biol. 197:729-736 (1987)). The wild type E. coli GyrA binds irreversibly to quinoline drugs such as ofloaxcin, preventing DNA relaxation during replication, and leading to cell death (see FIG. 3A). However, certain mutants of wild type E. coli GyrA are drug-resistant, while retaining gyrase activity. The generation of an E. coli GyrA merodiploid host cell which contains a chromosomal copy of a drug-resistant gyrA gene with a second intein-containing, drug sensitive gyrA gene results in a drug sensitive E.coli host, since in this case, drug sensitivity is dominant. The drug sensitive phenotype is dominant because the drug sensitive GyrA forms an irreversible covalent poison complex with the drug that interferes with DNA replication (FIG. 3A).

[0049] By merodiploid we mean that the cell contains an extra copy of a gene (or several genes) which has been introduced into the cell by any means known to one skilled in the art, such as transformation, infection, conjugation, plasmids, virus, phage, or by generating a transgenic strain and which may be present on either an episomal element or on the host chromosome.

[0050] In accordance with the present invention, there is further provided a similar type of positive selection system (for identifying agents which block or activate protein splicing) which comprises a host cell which contains (1) a copy of the extein gene (either episomal, chromosomal or synthetic) encoding a wild type form of a target enzyme, which expresses a non-toxic form of the extein protein; and (2) a second a extein gene (either episomal, chromosomal or synthetic) encoding a cytotoxic form of the target enzyme, into which is inserted an intein. In this system, the merodiploid host cell expresses the cytotoxic enzyme if the intein is properly spliced. Thus, cells must be treated with chemicals, agents or peptides that block splicing of the cytotoxic enzyme at all times when the cytotoxic enzyme is expressed. The cytotoxic extein must be dominantly lethal, as the intein-less copy of the extein gene is also expressed. The intein-less copy of the gene is required to maintain viability in cells in which splicing of the plasmid borne extein gene is blocked.

[0051] In one preferred embodiment, instead of using an intein inserted into a cytotoxic foreign extein homolog, the natural intein precursor may be mutated to produce a cytotoxic extein enzyme after splicing of the intein.

[0052] In a particularly preferred embodiment, the Mycobacterium tuberculosis DnaB precursor (Cole et al., Nature, 393:537-544 (1998)) is mutated in the extein region to a cytotoxic form based on the known cytotoxic mutation in E.coli DnaB, where Arg231 was mutated to Cysteine (Marszalek and Kaguni, J. Biol. Chem., 267:19334-19340 (1992) and Shrimankar, et al., J. Bacteriol., 174:7689-7696 (1992)). DNA helicases are essential proteins that unwind a DNA duplex to yield a single-stranded DNA intermediate required for replication, recombination, and repair (LeBowitz and McMacken, J. Biol. Chem., 261:4738-4748 (1986) and Lohman, Mol., Microbiol., 6:5-14 (1992)). The hexameric E. coli helicase encoded by the dnaB gene interacts with an hexameric DnaC complex and ATP. Some DnaB mutants are dominant lethal (Bouvier and Oreglia, C. R. Acad. Sci. Hebd Seances Acad. Sci. D., 280:649-652 (1975) and Maurer and Wong, J. Bacteriol., 170:3682-3688 (1988), Saluja and Godson, J. Bacteriol., 177:1104-1111 (1995) and Sclafani, et al., Mol. Gen. Genet. 182:112-118 (1981)). The R231C mutant protein is deficient in ATP hydrolysis, helicase activity, and replication activity at the chromosomal origin of replication resulting in cell death. As shown in FIG. 4D, Mtu DnaB contains this same Arginine (R227), and mutating it to Cysteine renders the Mtu DnaB gene cytotoxic. This mutation is dominantly cytotoxic in E. coli, and both the E.coli and M. tuberculosis DnaB proteins result in sequestering of the E. coli DnaC protein into inactive complexes, preventing DnaC from ‘loading’ DnaB onto the E. coli DNA replication fork (see FIG. 4A-1 and FIG. 4A-2).

[0053] Both of the positive selection systems described in the present invention utilize either native or homologous foreign exteins in order to optimize protein splicing and avoid the ineffecient splicing which can result from insertion of the intein into non-homologous foreign extein.

[0054] Co-transformation of the host cell in these positive selection systems with a plasmid engineered for the expression of an in vivo peptide library or transformation with a plasmid that contains both the selection marker and the in vivo peptide library (as in FIG. 6) allows for the direct selection of clones expressing a peptide that blocks (or alternatively activates) protein splicing. In vivo expression of peptides may be hampered by the host's efficient proteolytic degradation systems. Therefore, expression of these peptides in vivo in the context of larger proteins is preferred, especially in surface loop regions of larger proteins. In vivo expression of peptides fused to larger proteins has been achieved for example, in the catalytic loop of thioredoxin (Colas et al., Nature, 380:548-550 (1996)), and it is possible to express peptides fused within many different proteins. Peptides expressed in-frame in highly soluble, well expressed, thermostable, solvent-exposed loops of a protein are less subject to in vivo proteolysis or degradation and such fusions enhance the functional expression of peptides in a cell.

[0055] In a preferred embodiment, a combinatorial peptide library in a fragment of chicken α-spectrin is constructed, as previously described (see U.S. Pat. No. 5,834,247, supra at Example 17). The EF hand region of chicken α-spectrin was chosen because its structure is known, its EF hand domain forms a small protein with a stable structure, and it has a flexible surface loop. The structure of the chicken α-spectrin EF hand domain was elucidated by NMR analysis (Trave, et al., EMBO J. 14:4922-4931 (1995)). The term EF hand describes a type of protein tertiary structural motif consisting of a helix, a turn (loop) and a second helix. The EF hand domain of chicken α-spectrin is located at the carboxy terminus of chicken α-spectrin. Its 84 amino acid structure is arranged in two EF hand helix-turn-helix motifs separated by a 14 amino acid long flexible linker (SEQ ID NO:12). The protein is extremely soluble without any detectable precipitation or aggregation even at concentrations of up to 10 mM. The linker loop is mainly unstructured in solution and mutagenesis data show that minor deletions or insertions in the loop do not disturb the stabilizing hydrophobic interactions between the 2 EF-hand.

[0056] We have taken advantage of this last property to insert random peptides in the linker region between the chicken α-spectrin EF hands. Peptide libraries of various sizes can be investigated. It will be readily apparent to the skilled artisan that alternative methods of producing in vivo peptide libraries for screening may be utilized and are within the scope of the present invention.

[0057] Although the systems discussed above select for agents that block splicing of native or homologous exteins, it will be recognized by those of skill in the art that similar strategies can be used for screening with reporter genes to look for agents that inhibit expression of active reporter genes. It will likewise be recognized by the skilled artisan that similar strategies can be used to look for agents that activate splicing of a splicing-deficient intein in its native context or in a homologous extein gene, as long as the extein gene can be converted into a reporter. For example, in one embodiment, a reagent that activates a splicing-deficient precursor results in expression of an active reporter protein, resulting in inhibition of cell growth or detection of the active reporter protein.

[0058] Although the Escherichia coli (E.coli) GyrA and the Mycobacterium tuberculosis (M. tuberculosis) DnaB selection systems are described above, it will be recognized by the skilled artisan that any genetic selection system can also be used to isolate peptide sequences or other agents which disrupt or catalyze protein splicing. Likewise, although we describe the specific use of the M. xenopi GyrA intein (SEQ ID NO:11) and the M. tuberculosis DnaB intein (SEQ ID NO:13), the skilled artisan will recognize that this strategy is equally applicable to any intein (see, e.g., Perler, et al., Nucleic Acids Res. 27:346-347 1999)) present in its native or homologous context. It will likewise be readily apparent to those of skill in the art that alternative means of generating peptide libraries for screening may be used. It will likewise be recognized by the skilled artisan that in the absence of a selection or screening system for the native extein protein that similar strategies can be applied to splicing of inteins in less optimal heterologous extein systems.

[0059] Activating and/or inhibiting agents identified by the screening methods of the instant invention may also be used to control the in vivo expression of a target protein. Once the only copy of an active extein gene contains an intein, gene function can be inhibited if the organism is treated with an agent that blocks splicing. On the other hand, if a splicing-impaired intein is used, gene function can be activated if the organism is treated with an agent that activates splicing. The agents and splicing can be modulated at any time during the development and life of the organism by addition or removal of the splicing activating or inhibiting agent.

[0060] Similarly, controllable splicing may be used to deliver active proteins at specific times or to specific places. In many instances, therapeutic drugs can be cytotoxic to the host and would be best if only active at the target site. For example, chemotherapy drugs are often generally cytotoxic and adverse reactions in normal cells could be eliminated if the drug could be specifically activated in the tumor. If one has a drug that is at least partially proteinacious, an intein that can be activated or inhibited by a second agent, as described above, could be inserted into the protein portion of the therapeutic agent. The drug is then administered in an inactive form, and subsequently activated in the desired target tissue.

[0061] As noted above, it is believed that inefficient protein splicing in the foreign extein context occurs as inteins may exhibit substrate specificity, preferring their native extein sequence to that of foreign exteins. In accordance with another embodiment, there is provided a method of overcoming this limitation by employing an intein with one or more, preferably one to five, amino acid residues from its native extein. Inclusion of such amino acid residues may be at either or both ends of the homologous intein. Inclusion of amino acid residues from the native extein will facilitate methodologies of the present invention. Such amino acid residues from the native extein may be incorporated into the precursor by methods well known to those skilled in the art.

[0062] Insertion of a target intein in the heterologous extein may be at any of a number of sites, including but not limited to, a surface location in the extein, within a loop region of the extein, at a protease sensitive site, or a position known to facilitate insertion of additional amino acid residues without inactivating the extein.

[0063] The Examples presented below are only intended as specific preferred embodiments of the present invention and are not intended to limit the scope of the invention except as provided in the claims herein. The present invention encompasses modifications and variations of the methods taught herein which would be obvious to one of ordinary skill in the art.

[0064] The references cited above and below are hereby incorporated by reference herein.

EXAMPLE I

A Mxe GyrA Intein-Mediated Positive Selection System for Inhibition of Protein Splicing

[0065] A Positive Selection System Based on Blocking Splicing of the Mxe GyrA Intein (IVPS) in E. coli GyrA: Background

[0066] Gyrases are essential multimeric enzymes involved in DNA replication in bacteria (Swanberg and Wang, 1987). Both gyrase subunit A and B have been extensively studied as drug targets in bacterial human pathogens (e.g., Mycobacteria, Salmonella, Enterbacteriaceae, Citrobacter, Pseudomonas, Streptococcus, Staphylococcus, Yersinia, Rhodobacter, Haemophilus, Neisseria, Providencia). The GyrA subunit of gyrases can complex with quinoline drugs, such as ofloxacin, and induce cell death. The complex formation of quinolines with gyrase is followed by a rapid and irreversible inhibition of DNA synthesis, inhibition of growth, and induction of the SOS response (see FIG. 3A). At higher drug concentrations, cell death occurs as double-strand DNA breaks in the bacterial chromosome are released from trapped gyrase complexes.

[0067] In many gram-negative bacteria (e.g., E. coli), resistance to quinoline arises from mutation of the Gyrase A protein in the quinoline resistance determining region such as gyrA96 or gyrA83. Those mutations may involve Ser83 in E. coli GyrA. In a merodiploid cell containing a drug resistant gyrA gene (such as gyrA83) on the chromosome and a wild type (gyrA+) copy of gyrA on a plasmid, the wild type gene (drug sensitive) product of gyrA is dominant. By merodiploid, we mean that the cell contains an extra copy of a gene (or several genes) which has been introduced into the cell by any means known to one skilled in the art, such as transformation, infection, conjugation, plasmids, virus, phage, or by generating a transgenic strain and which may be present on either an episomal element or on the host chromosome. The wild type gyrA gene can be introduced into the cell by any means known to one skilled in the art and should not be considered limited to a plasmid. Many E. coli strains are available that contain gyrA mutants which are resistant to quinoline drugs such as ofloxacin. However, this system is also applicable to any other host system where (1) the chromosomal copy of the gyrA gene is resistant to quinoline drugs, (2) the introduced sensitive gyrA gene is present as the heterologous E. coli gyrA::Mxe gyrA intein fusion or the native Mxe gyrA or M. leprae gyrA genes, and (3) the intein containing drug sensitive gyrA gene is operably linked with the appropriate signals for expression in that host. Likewise, the Mxe GyrA intein could be inserted into the gyrA gene of any experimental host cell just as it was inserted into the E. coli gyrA gene. Likewise, any gyrA intein can be used in the above selection system, whether present at the same site as the Mxe GyrA intein or a different site.

[0068] Since sensitivity to quinoline drugs is dominantly cytotoxic, in the presence of these drugs, a gyrA+/gyrA83 host cell is not viable because wild type GyrA proteins can still bind drug molecules and poison DNA replication (see FIG. 3B). The co-expression of a chicken α-spectrin peptide library (as described in U.S. Pat. No. 5,834,247 supra. at Example 17) allows for the positive selection of peptides that can disrupt splicing of the Mxe GyrA intein. Likewise, this system can be used to screen for any agent that inhibits splicing of the Mxe GyrA intein in vivo or for Mxe GyrA intein mutations that block splicing.

[0069] Insertion of the Mxe GyrA Intein (IVPS) Gene into the Homologous Site in the E. coli GyrA Gene

[0070] In some Mycobacteria, the gyrase A subunit active site is often interrupted by a naturally occurring allelic intein (IVPS) near the active site tyrosine residue (e.g., Mycobacterium flavescens, Mycobacterium gordonae, Mycobacterium kansasii, Mycobacterium leprae, Mycobacterium malmoense and Mycobacterium xenopi, (Sander, et al., Microbiology, 144:589-591 (1998), Telenti, et al., J. Bacteriol., 179:6378-6382 (1997), Perler, et al., Nucleic Acids Res. 27:346-347 (1999), Southworth, BioTechniques, 27:110-121 (1999)). The M. xenopi (Mxe) GyrA intein (IVPS) utilized in the system described herein, lacks the endonuclease signature motifs and other sequences similar to homing endonucleases and has been extensively studied as a prototype minimal intein (IVPS) (Klabunde, et al., Nature Struct. Biol. 5:31-36 (1998), and Telenti, et al., J. Bacteriol., 179:6378-6382 (1997) and Southworth BioTechniques, 27:110-121 (1999)). The most favorable insertion site for the Mxe GyrA intein in E. coli GyrA is the homologous insertion site compared to the native Mxe GyrA extein, since it shares sequence identity with the native Mxe GyrA extein (FIG. 3D). The intein (IVPS) insertion site was chosen immediately upstream of the conserved tyrosine active site residue at position 122 in E. coli GyrA:

D S A A A M R Y122 c i t --- s h n T123 E I R L+E  A K I

(SEQ ID NO:14)

(SEQ ID NO:45)

[0071] Amino acid numbers refer to the position of the amino acids in E. coli GyrA (see SEQ ID NO:1 for a partial E. coli GyrA sequence). The underlined amino acids (single letter amino acid code) are the amino acids identical in both E. coli and Mxe GyrA exteins (see FIG. 3D). The lower case letters represent Mxe GyrA intein (IVPS) amino acids. The dashes represent the remainder of the residues of the Mxe GyrA intein (IVPS) that are not listed (see SEQ ID NO:2, for the complete Mxe GyrA intein sequence).

[0072] First, the E. coli gyrA gene was cloned by polymerase chain reaction (PCR) using E. coli K12 genomic DNA under the following experimental conditions. A forward primer 5′-GATA GGCTAGCGATGAGCGACCTTGCGAGAG-3′ (SEQ ID NO:15) and reverse primer 5′-TGAAGCAATTGAATTATTCTTCTTCTGGCTCG-3′ (SEQ ID NO:16) were used in a PCR mixture containing 20 U/ml Vent® Exo+ DNA polymerase (New England Biolabs, Inc., Beverly, Mass.), 400 μM of each dNTP, 4 nM each primer and 100 ng of E. coli K12 genomic DNA. Amplification was carried out in a Perkin-Elmer/Cetus (Emeryville, Calif.) thermal cycler 480 for 1 min at 95° C. and then cycled at 45° C., 30 sec; 72° C., 2 min and 30 sec; 95° C., 30 sec for 20 cycles. The PCR products from one 50 μl PCR reaction and 2 μg of pCYB1 (IMPACT™ I kit, New England Biolabs, Inc., Beverly, Mass.) were separately digested with 250 U/ml of NheI and 1000 U/ml of Mfel in the presence of 100 μg/ml of BSA. The digestion was performed at 37° C. for 1 hour. Digested PCR products and plasmid DNA were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 20° C. for 1 hour using a 1:4 ratio of vector to insert and 40,000 U/ml of T4 ligase. Ligation products were transformed into E. coli strain ER2267 competent cells. Recombinant plasmids were checked by NheI/MfeI digestion which results in the excision of the cloned insert in properly ligated recombinants. One of the resultant correct plasmids containing the E. coli gyrA gene placed under transcriptional control of the pCYB1 pTac promoter was named pEA500. The gyrA insert was checked by DNA sequencing to insure that no sequence errors were introduced by PCR.

[0073] Second, to facilitate cloning of the Mxe GyrA intein into E. coli GyrA, unique silent NotI and XbaI restriction enzyme sites were engineered 7 bp and 44 bp, respectively, away from each side of the E. coli GyrA active site residue, Y122 of pEA500 by site-directed silent mutagenesis. The QuickChange kit was used following the manufacturer's instructions (Stratagene, La Jolla, Calif.) with mutagenic primers: NotI oligonucleotides: 5′-CGGCGACTCTGCGGCCGCAATGCGTTATA CGG-3′ (SEQ ID NO:17) and 5′-CCGTATAACGCATTGCGGCCGCA GAGTCGCCG-3′ (SEQ ID NO:18), and XbaI oligonucleotides: 5′-GAACTGATGGCCGCTCTAGAAAAAGA GACGG-3′ (SEQ ID NO: 19) and 5′-CCGTCTCTTTTTCTAGAGCGGCCA TCAGTTC-3′ (SEQ ID NO:20). The resultant plasmid containing E. coli GyrA with NotI and Xbal restriction enzyme sites was called pEA502.

[0074] Third, a 68 bp DNA cassette with flanking NotI/XbaI restriction sites was designed to be cloned into the pEA502 unique NotI/XbaI sites. This cassette introduced a unique BlpI silent restriction enzyme site 10 bp away from Y122 which subsequently allowed cloning of any intein (IVPS or CIVPS) near the E. coli GyrA active site Y122 using NotI and BlpI restriction enzyme sites. This cassette was generated by annealing 2 complementary oligonucleotides : 5′-GGCCGCAA TGCGTTATACGGAAATCCGCTTAGCGAAAATTGCCCATGAACTGATG GCCGAT-3′ (SEQ ID NO:21) and 5′-CTAGATCGGCATCAGTTCATG GGCAATTTTCGCTAAGCGGATTTCCGTATAACGCATTGC-3′ (SEQ ID NO:22). 5 nM of each oligonucleotide was combined in 1× T4 ligase buffer (New England Biolabs, Inc., Beverly, Mass.), boiled for 5 min and cooled down to room temperature. 10 μg of pEA502 were digested with NotI and XbaI using 500 U/ml each enzyme in the presence of 100 μg/ml of BSA. The digestion was performed at 37° C. for 2 hours. The digested plasmid DNA was separated by agarose gel electrophoresis and the excised band further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation of the oligonucleotide cassette and the digested plasmid DNA was carried out at 20° C. for 1 hour using a 1:2 ratio of vector to insert and 40,000 U/ml of T4 ligase. Ligation products were transformed into E. coli strain ER2267 competent cells. Recombinant plasmids were checked by BlpI digestion which results in the linearization of the correct recombinant plasmids. One of the resultant correct plasmids was named pEA523.

[0075] Fourth, the Mxe GyrA intein (IVPS) was amplified by PCR with the addition of primer derived NotI and BlpI sites using pMIP(Mxe)#4 plasmid DNA (Telenti et al., J. Bacteriol, 179:6378-6382 (1997)) under the following experimental conditions: Forward primer 5′-CGACCCGCGCGGCCGCAATGC GTTATTGCATCACGGGAG-3′ (SEQ ID NO:23) and reverse primer 5′-GCCAAAGGCGCTAAGCGGATTTCCGTGTTGTGGCTGACGAACC CG-3′ (SEQ ID NO:24) were used in a PCR mixture containing 10 U/ml Taq DNA polymerase (Promega, Madison, Wis.), 200 μM of each dNTP, 4 nM each primer and 100 ng pMIP(Mxe)#4 DNA. Amplification was carried out in a Perkin-Elmer/Cetus (Emeryville, Calif.) thermal cycler 480 at 94° C., 30 sec; 50° C., 30 sec; 72° C., 15 sec for 10 cycles. The PCR products of one 50 μl reaction and 2 μg of pEA523 were separately digested using 1000 U/ml of NotI and 300 U/ml of BlpI in the presence of 100 μg/ml of BSA. The digestion was performed at 37° C. for 2 hours. Digested PCR products and plasmid DNA were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 20° C. for 2 hours using a 1:3 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain Top10F′ (Invitrogen, Carlsbad, Calif.) competent cells. Recombinant plasmids were checked by EcoNI restriction enzyme digestion which results in the linearization of the correct recombinant plasmids. One of the resultant correct plasmids was named pEA600 and contains the in-frame insertion of the Mxe GyrA intein (IVPS) into the E. coli GyrA extein at the active site Y122 (see FIGS. 3D and 3E).

[0076] Construction of a Vector for Co-Expression of a Peptide Library and the Mxe GyrA Intein (IVPS)::E. coli GyrA Selection System

[0077] As described above, the Mxe GyrA intein (IVPS) was inserted into the active site of E. coli GyrA. In this homologous context, the Mxe GyrA intein (IVPS) splices efficiently to produce active E. coli GyrA. As is detailed below, the E. coli gyrA::Mxe gyrA intein (IVPS) gene fusion described above was cloned under control of a T7 RNA polymerase promoter and introduced into E. coli, ER2726 (New England Biolabs, Inc., Beverly, Mass.). ER2726 expresses T7 RNA polymerase and has the gyrA83 mutation which makes the chromosomal gyrA gene resistant to quinoline drugs. In the presence of quinoline drugs such as ofloxacin, only splicing deficient clones can survive (see FIGS. 3B and 3C), since the spliced gyrA product is sensitive to ofloxacin in a dominant cytotoxic manner (see above).

[0078] The spectrin scaffold was cloned into EA600 as follows. First, a 30 bp DNA cassette with flanking PflMI/ApaI restriction sites was designed to be cloned into the unique PflMI/ApaI sites in pEA600 (which also contains the E. coli gyrA::Mxe gyrA fusion). This cassette introduced a unique SphI site in place of the lacIq gene and was synthesized by annealing 2 oligonucleotides: 5′-ATGGGCATGCATATATATA TAGGCCTGGGCC-3′ (SEQ ID NO:25) and 5′-CAGGCCTATATATAT ATGCATGCCCATTCG-3′ (SEQ ID NO:26). 5 nM of each oligonucleotide was combined in 1× T4 ligase buffer (New England Biolabs, Inc. Beverly, Mass.), boiled for 5 min and cooled down to room temperature. 5 μg of pEA600 was digested using 320 U/ml of PflMI and 800 U/ml of ApaI in the presence of 100 μg/ml of BSA. The digestion was performed at 37° C. for 2 hours. The digested plasmid DNA was separated by agarose gel electrophoresis and the excised band further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:1 ratio of vector to insert and 40,000 U/ml of T4 ligase. Ligation products were transformed into E. coli strain XL1B (Stratagene, La Jolla, Calif.) competent cells. Recombinant plasmids were checked by SphI digestion which results in the linearization of the correct recombinant plasmids. One of the resultant correct plasmids was named pEA661.

[0079] Second, unique SgfI and sites ClaI were engineered on either side of the spectrin loop region in a spectrin encoding plasmid (Trave, et al., EMBO J. 14:4922-4931 (1995)) by site-directed silent mutagenesis using the QuickChange kit as described by the manufacturer (Stratagene, La Jolla, Calif.). The SgfI oligonucleotides were: 5′-GTTTAAGTCTTGCTTGCGATC GCTTGGCTATGACCTGCC-3′ (SEQ ID NO:27) and 5′-GGGCAGGT CATAGCCAAGCGATCG CAAGCAAGACTTAAA-3′ (SEQ ID NO:28) and ClaI oligonucleotides were: 5′-GCCTGACCCCGAATTTGAATC GATTCTTGACACTGTTG-3′ (SEQ ID NO:29) and 5′-CAACAGTGT CAAGAATCGATTCAA ATTCGGGGTCAGGC-3′ (SEQ ID NO:30). The resulting plasmid was called pEA670.

[0080] Third, the SgfI/ClaI mutated spectrin gene was cloned by PCR into pEA661 under the following experimental conditions. A forward primer 5′-AATGGTGCATGCAAGGAGATGGCGCCCAAC AGTC-3′ (SEQ ID NO:31) and reverse primer 5′-GCTTTGGCTAG CTTTCCTGTGTCACCTGCTGATCATGAACG-3′ (SEQ ID NO:32) were used as described in the Expand High Fidelity PCR system (Boehringer Mannheim, Indianapolis, Ind.) in the presence of 1× buffer 2 (New England Biolabs, Beverly, Mass.) and 50 ng of pEA670 DNA. Amplification was carried out in a Perkin-Elmer/Cetus (Emeryville, Calif.) thermal cycler 480, 94° C., 30 sec; 45° C., 30 sec; 72° C., 45 sec; for 15 cycles. The PCR products of one 50 μl tube and 5 μg of pEA661 were NheI/SphI digested using 250 U/ml of each enzyme. The digestion was performed at 37° C. for 2 hours. Digested PCR products and plasmid DNA were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:5 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain Novablue DE3 (Novagen, Madison, Wis.) competent cells. Recombinant plasmids were checked by NheI/SphI digestion which results in the excision of the cloned insert. One of the resultant correct plasmids was named pEA671.

[0081] Fourth, the first E. coli gyrA Pvul site in pEA671 was eliminated by site-directed silent mutagenesis using the QuickChange kit as described by the manufacturer (Stratagene, La Jolla, Calif.) and oligonucleotides 5′-GCGTAAAGCTCGCGACC GTGCTCATATCC-3′ (SEQ ID NO:33) and 5′-GGATATGAGCACGGTC GCGAGCTTTACGC-3′ (SEQ ID NO:34), resulting in plasmid pEA681.

[0082] Fifth, the 200 bp Acc65I/HindIII fragment from pEA681 was transferred to pEA671 replacing the Acc691/HindIII fragment of EA671. Plasmids pEA671 and pEA681 were digested in 1× buffer 2 (New England Biolabs, Inc., Beverly, Mass.) using 500 U/ml of Acc65I and 500 U/ml of HindIII (New England Biolabs, Inc., Beverly, Mass.). The digestion was performed at 37° C. for 2 hours. Digested plasmids were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 3 hours using a 1:3 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain XL1B (Stratagene, La Jolla, Calif.) competent cells. Recombinant plasmids were checked by Pvul digestion. One of the resultant correct plasmids was named pEA682. This plasmid contains both the α-spectrin peptide library and the E. coli gyrA::Mxe gyrA intein-based selection system on the same plasmid, both under control of a T7 RNA polymerase promoter (FIG. 5A).

[0083] A Theoretical Screening for Peptides that Disrupt Protein Splicing of the Mxe GyrA Intein (IVPS) in E. coli GyrA.

[0084] In the theoretical embodiment detailed below, random peptides of 7-12 amino acids would be inserted in-frame into the loop between the 2 EF-hand motifs of α-spectrin, contained, as described above, on the same plasmid as E. coli gyrA::Mxe gyrA intein fusion. The resulting plasmids would be electro-transformed into strain ER2726 (New England Biolabs, Inc., Beverly, Mass.). Transformants would be selected in LB liquid growth media in the presence of ampicillin, ofloxacin (Sigma, St. Louis, Mo.) and IPTG to allow selection against the splicing proficient clones. Ampicillin selects for the presence of the plasmid and ofloxacin selects for peptides that block splicing, since the spliced E. coli GyrA protein would be sensitive to the drug and lead to cell death. Plasmid DNA would be isolated from selected clones and digested with SgfI and ClaI to isolate DNA fragments encoding the selected spectrin peptides. The spectrin DNA loop fragments would then be cloned back into the original selection plasmid. Iterative rounds of drug selection and “back-cloning” would be performed (FIG. 5B). Iterative screening helps enrich for agents that truly block splicing while eliminating clones that survived selection because of some other mutation or anomaly. Final selected clones would be grown individually in liquid culture and the plasmid-encoded E. coli GyrA specifically induced by IPTG. Crude protein cell extracts would be electrophoresed and blotted for immuno-staining. Clones in which the E. coli GyrA spliced product was not detected would be considered positives, i.e. clones in which splicing had been disrupted, potentially by the selected peptide.

[0085] The random peptide library would be synthesized in vitro using the following protocol, as was done in Example II. A single strand/double strand DNA hybrid cassette would be synthesized by annealing 2 oligonucleotides: 5′-TGTCAAGAATC GATTCAAATTCGGGGTCAGGCTCTCC((W)NN)7-12ATAGCCAAGCGA T-3′ (SEQ ID NO:35) and 5′P-CGCTTGGCTAT-3′ (SEQ ID NO:36). 5 μg of oligonucleotide SEQ ID NO:35 and 3 molar equivalents of oligonucleotide SEQ ID NO:36 would be mixed together in the presence of 0.1 M NaCl in a final volume of 50 μl. The mixture would then be boiled and immediately cool down to room temperature in the same boiler. The single stranded random nucleotide part of the DNA hybrid cassette formed by annealing of the 2 oligos would be extended using 400 μM of each dNTPs and 60 U/ml of Klenow DNA polymerase (New England Biolabs, Inc., Beverly, Mass.) in a final volume of 200 μl in 1× EcoPoI buffer (New England Biolabs, Beverly, Mass.). The extension reaction would be left 20 minutes at 37° C. and further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). 60 μg of pEA682 (50 μg/ml) would be digested in 1× Buffer 2 (New England Biolabs, Inc., Beverly, Mass.) using 250 U/ml of SgfI (Promega, Madison, Wis.) and 500 U/ml of BspDI (New England Biolabs, Inc., Beverly, Mass.) (an isoschizomer of ClaI) in the presence of 100 μg/ml of BSA. The digestion would be performed at 37° C. for 2 hours. Purified cassettes (50 μl) would then be digested in 1× Buffer 4 (New England Biolabs, Inc., Beverly, Mass.) using 500 U/ml of ClaI (New England Biolabs, Inc., Beverly, Mass.) in the presence of 100 μg/ml of BSA. Cassettes would be further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Digested plasmid DNA would be electrophoresed on 0.7% agarose gel and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation would be carried out at 16° C. for 1 hour using a 1:1 ratio of vector (2 ng/μl) to insert and 1,600 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products would then be electro-transformed into E. coli strain ER2744 (New England Biolabs, Inc., Beverly, Mass.) competent cells (109 pUC18 transformants/μg) using 1-2 μg of total ligated plasmid for each 200 μl aliquot of competent cells, at 2.5 kV/cm in a 2 mm cuvette (BIORAD, Richmond, Calif.). Cells would be allowed to recover in a shaker for 1 hour at 37° C. Recovered transformants would be inoculated at 1/100 dilution ratio into LB liquid growth media containing appropriate amounts of ofloxacin (Sigma, St. Louis, Mo.), 100 μg/ml of ampicillin and 1 mM IPTG. Transformants would be incubated overnight at 37° C. Plasmid DNA would be isolated from a 100 ml of the overnight culture using a tip100 column (QIAGEN, Studio City, Calif.), ClaI/SgfI digested as above and electrophoresed on a 4% GTG Nusieve agarose gel (FMC BioProducts, Rockland, Me.). The 57 to 72 bp spectrin loop DNA inserts (depending upon whether the peptide library contained 7 or 12 random amino acids) would be purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.) and cloned back into SgfI/ClaI digested and purified selection plasmid as described above. This protocol would be repeated 3 times to enrich the pool of transformants for peptide clones having the most biologically active sequences against the protein splicing of the Mxe intein (IVPS). Finally selected clones would be grown individually in 10 ml LB containing 100 μg/ml ampicillin at 37° C. and induced with 1 mM IPTG for 3 hours. Crude protein cell extracts would be electrophoresed on a 10-20% gradient gel (Novex, San Diego, Calif.). The gel would then be electro-blotted for immuno-staining using anti-His tag antibodies (Sigma, St. Louis, Mo.) to detect GyrA::Mxe intein (IVPS) protein splicing products. One would expect to see the absence of spliced product. The clones would then be sequenced to determine the amino acid sequences which had been selected.

[0086] Hypothetical Screening with Agents that Inhibit Splicing

[0087] At this stage, the vector, pEA600, is amenable for screening with any type of agent that blocks splicing, using a similar screening protocol as for peptides that block splicing, described above. However, in this case, pEA600 or similar plasmids can be directly screened without having to clone the peptide library contained within the chicken α-spectrin gene as described above. The protocol would involve treating individual cultures with single or pooled agents that can enter the cell and looking for cell growth, using any means known to one skilled in the art. Agents that block splicing allow the cell to grow in the presence of ofloxacin

[0088] Summary

[0089] In summary, we describe the cloning of the Mxe gyrA intein gene into the E. coli gyrA extein gene for use in selecting for agents that inhibit splicing. The Mxe GyrA intein splices well in the E. coli GyrA extein, resulting in production of active E. coli GyrA protein. The E. coli GyrA extein was used with the Mxe GyrA intein because the Mle GyrA intein did not splice efficiently in E. coli in its native context and the precursor was mostly insoluble in E. coli. Because the GyrA intein and extein sequences are very similar (Telenti, et al., J. bacteriol, 179:6378-6382 (1997) and Perler, et al., Nucleic Acids Res. 27:346-347 (1999)), mixing and matching of inteins, exteins and experimental hosts resulted in an efficient model system for examining agents that modulate splicing of GyrA inteins, using exteins that have similar insertion sites and therefore similar splicing active sites as in the native context.

EXAMPLE II

A M. tuberculosis DnaB Intein-Mediated Positive Selection System

[0090] A Positive Selection System Using the Mtu DnaB Intein (IVPS) in its Native Mtu DnaB Extein in E coli: Background

[0091] The hexameric E. coli helicase encoded by the dnaB gene interacts with an hexameric DnaC complex and ATP. Some DnaB mutants are dominant lethal (Bouvier and Oreglia, C. R. Acad. Sci. Hebd. Seances Acad. Sci D., 280:649-652 (1975), Maurer and Wong, J. Bacteriol 170:3682-3688 (1988), Saluja and Godson, J. Bacteriol. 177:1104-1111 (1995) and Sclafani, et al., Mol. Gen. Genet., 182:112-118 (1981)). By dominant or dominantly cytotoxic, we mean that the toxicity occurs even if homologous proteins are present which are not cytotoxic or resistant to the drug, i.e., the cytotoxic effect dominates irrespective of the presence of non-cytotoxic homologs. The mutant protein is deficient in ATP hydrolysis, helicase activity, and replication activity at the chromosomal origin of replication resulting in cell death (see FIG. 4A). Despite only moderate protein sequence identity between bacterial helicases, arginine 231 is located in a conserved motif proposed to interact directly with DnaC (Marszalek and Kaguni, J. Biol. Chem., 267:19334-19340 (1992) and Shrimankar, et al., J. Bacteriol., 174:7689-7696 (1992)). M. tuberculosis (Mtu) DnaB has a naturally occurring intein at the carboxy-terminus and an arginine at position 227 homologous to arginine 231 of E. coli DnaB (see FIG. 4D).

[0092] We have demonstrated proficient protein splicing of the Mtu DnaB intein (IVPS) from the Mtu DnaB precursor protein in E. coli and also have shown that the R227C mutation results in dominant lethality. Therefore, a merodiploid cell containing a wild type dnaB gene and a Mtu DnaB (R227C) gene is not viable unless protein splicing can be disrupted (see FIGS. 4B and 4C). By merodiploid we mean that the cell contains an extra copy of a gene (or several genes) which has been introduced into the cell by any means known to one skilled in the art, such as transformation, infection, conjugation, plasmids, virus, phage, or by generating a transgenic strain and which may be present on either an episomal element or on the host chromosome. The co-expression of a chicken α-spectrin peptide library (as described in U.S. Pat. No. 5,834,247 supra. at Example 17) allows for the positive selection of peptides that can disrupt splicing of the M. tuberculosis DnaB intein (see FIG. 5A). Likewise, this system can be used to screen for any agent that inhibits splicing of the Mtu DnaB intein or any other DnaB intein in vivo or for DnaB intein mutations that block splicing.

[0093] Construction of a Positive Selection System Using the Mtu DnaB Intein (IVPS) in its Native Mtu DnaB Extein

[0094] As described in detail below, the Mtu dnaB gene has been cloned by PCR under T7 RNA polymerase transcriptional control. In E. coli, the Mtu DnaB intein (IVPS) splices very efficiently from its natural precursor to produce the Mtu DnaB helicase. The Mtu dnaB gene has been mutagenized at position 227 from arginine to cysteine and the plasmid transformed into BL21(DE3)-Gold (Stratagene, La Jolla, Calif.). In the presence of the T7 RNA polymerase (induced by IPTG) only splicing deficient clones can survive (see FIG. 4C).

[0095] First, the Mtu dnaB gene was cloned by PCR using M. tuberculosis H37Ra genomic DNA under the following experimental conditions. A forward primer 5′-AGGTGAGAA TTCATGGCGGTCGTTGATGACC-3′ (SEQ ID NO:37) and reverse primer 5′-TATATAAAGCTTTCATGTCACCGAGCCATGTTGGCG-3′ (SEQ ID NO:38) were used as described in the Extend Long Template PCR system (Boehringer Mannheim, Indianapolis, Ind.) in the presence of 1× buffer 3 and 100 ng of M. tuberculosis genomic DNA. Amplification was carried out in a Perkin-Elmer/Cetus (Emeryville, Calif.) thermal cycler 480 for 2 min at 94° C. and then cycled at 45° C., 30 sec; 68° C., 2 min; 95° C., 1 min for 25 cycles. The PCR products of one 50 μl reaction and 5 μg of pET21a (Novagen, Madison, Wis.) were digested using 1000 U/ml of EcoRI and 800 U/ml of HindIII in 1× EcoRI buffer (New England Biolabs, Inc., Beverly, Mass.). The digestion was performed at 37° C. for 1 hour. Digested PCR products and plasmid DNA were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:5 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain Novablue DE3 (Novagen, Madison, Wis.) competent cells. Recombinant plasmids were checked by EcoRI/HindIII digestion which results in the excision of the cloned inserts. One of the resultant correct plasmids was named pEA807. The sequence of the dnaB insert was checked by DNA sequencing.

[0096] Second, the 1200 bp BglII/SgrAI fragment from pEA682 containing the spectrin-based peptide library was transferred to pEA807. Plasmids pEA682 and pEA807 were digested in 1× buffer 2 (New England Biolabs, Inc., Beverly, Mass.) using 500 U/ml of BglII and 240 U/ml of SgrAI. The digestion was performed at 37° C. for 1 hour. Digested plasmids were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:5 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain ER2726 (New England Biolabs, Inc., Beverly, Mass.) competent cells. Recombinant plasmids were checked by ClaI/NcoI digestion. One of the resultant correct plasmids was named pEA808.

[0097] Third, the Mtu dnaB gene was mutagenized to R227C by PCR under the following experimental conditions. A forward primer 5′-AGGTGAGAATTCATGGCGGTCGTTGATGACC-3′ (SEQ ID NO:39) and reverse primer 5′-TTTCCCACGCCCGGGCaCGCCGC CACGATGACC-3′ (SEQ ID NO:40) were used as described in the Extend Long Template PCR system (Boehringer Mannheim, Indianapolis, Ind.) in the presence of 1× buffer 3 (New England Biolabs, Inc., Beverly, Mass.) and 500 ng of pEA808 DNA. Amplification was carried out in a Perkin-Elmer/Cetus (Emeryville, Calif.) thermal cycler 480 for 2 min at 94° C. and then cycled at 45° C., 30 sec; 72° C., 45 sec; 95° C., 1 min for 20 cycles. The PCR products of one 50 μl reaction and 2 μg of pEA808 were digested overnight at 37° C. using 100 U/ml of EcoRI and 40 U/ml of SrfI in the 1× PCR-Script reaction buffer (Stratagene, La Jolla, Calif.). Digested PCR products and plasmid DNA were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:1 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain XL10-Kan (Stratagene, La Jolla, Calif.) competent cells. Recombinant plasmids were checked by NdeI/AscI. One of the resultant correct plasmids was named pEA809.

[0098] Fourth, the NotI dnaB-spectrin module was inverted on plasmid pEA809. 2 μg of pEA809 DNA was digested with 500 U/ml of NotI at 37° C. for 2 hours and digestion products split into two tubes. One tube containing 1 μg of NotI digested pEA809 was incubated further with 100 U/ml Calf Intestinal Alkaline Phosphatase (CIP, New England Biolabs, Inc., Beverly, Mass.) for 20 minutes at 37° C. Digested plasmid DNA from both tubes was separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation of the vector band from the CIP treated tube and the insert band from the CIP untreated tube was carried out at 16° C. for 1 hour using a 1:1 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain XL1-Blue (Stratagene, La Jolla, Calif.) competent cells. Recombinant plasmids were checked by SacII digest. One of the resultant correct plasmids was named pEA810.

[0099] Fifth, the lacIq gene from pEA810 was removed and replaced by a smaller DNA fragment from pBR322. pEA810 and pBR322 DNA were digested using 500 U/ml EcoRV and 500 U/ml HindIII in buffer 2 (New England Biolabs, Inc., Beverly, Mass.) at 37° C. for 2 hours. Digested plasmid DNA was separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation of the pEA810 vector band and the pBR322 insert band was carried out at 16° C. for 1 hour using a 1:1 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain XL1-Blue (Stratagene, La Jolla, Calif.) competent cells. One of the resultant correct plasmids was named pEA813.

[0100] Sixth, the first Mtu R227C dnaB AatII site of pEA813 was eliminated by site-directed silent mutagenesis using the QuickChange kit (Stratagene, La Jolla, Calif.) and oligonucleotides 5′-GCCGCCGATCCGCGACATCGTAGATTTCGGCC -3′ (SEQ ID NO:41) and reverse primer 5′-GGCCGAAATCTACGA TGTCGCGGATCGGCGGC-3′ (SEQ ID NO:42) resulting in plasmid pEA832.

[0101] Seventh, the wild type intein containing Mtu DnaB gene of plasmid pEA808 was shuffled back into pEA832. pEA808 and pEA813 DNA were digested using 1× buffer 1 (New England Biolabs, Inc., Beverly, Mass.) with 500 U/ml EcoRI and 500 U/ml HindIII at 37° C. for 1 hour. Digested plasmid DNAs were separated by agarose gel electrophoresis and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation of the pEA813 vector band and the pEA808 insert band was carried out at 16° C. for 1 hour using a 1:1 ratio of vector to insert and 40,000 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were transformed into E. coli strain XL1-Blue (Stratagene, La Jolla, Calif.) competent cells. One of the resultant correct plasmids was named pEA825.

[0102] Eighth, the AatII site elimination in pEA825 was performed identically as described for pEA813, resulting in plasmid pEA835.

[0103] Screening for Peptides that Disrupt the Mtu DnaB Intein (IVPS) Protein Splicing

[0104] The following is an actual experimental example demonstrating the use of this system to select for peptides that block splicing. As detailed below, random peptides of 7-12 amino-acids were inserted in-frame into the loop of the 2 EF-hand motif of α-spectrin, contained, as described above, on the same plasmid as the Mtu DnaB intein (IVPS) selection system. The resulting plasmids were electro-transformed into the T7 RNA polymerase E. coli strain ER2744 (New England Biolabs, Inc., Beverly, Mass.) (see FIGS. 5A and 5B). Transformants were selected in LB liquid growth media in the presence of ampicillin and IPTG to allow selection against the splicing proficient clones. Plasmid DNA was isolated from selected clones and digested to isolate DNA fragments encoding the selected spectrin peptides. The selected spectrin loop region DNA was cloned back into the original selection plasmid. Iterative rounds of selection and “back-cloning” were performed (FIG. 5B). After selection, the selected spectrin peptide were cloned into pEA825 (containing the non-toxic DnaB gene) for expression analysis. Final selected clones were grown individually in liquid culture and the plasmid-encoded Mtu dnaB gene specifically induced by IPTG. Crude protein cell extracts were electrophoresed and blotted for immuno-staining. Clones in which the Mtu DnaB spliced product was not detected were considered positives, i.e. clones in which splicing had been disrupted, potentially by a selected peptide.

[0105] The random peptide library was synthesized in vitro using the following protocol. A single strand/double strand DNA hybrid cassette was synthesized by annealing of 2 oligonucleotides: 5′-TGTCAAGAATCGATTCAAATTCGGGGTC AGGCTCTCC((W)NN)7-12 ATAGCCAAGCGATCGCAGGCAGCTTTT AAAGCCCTGATGGTTCAGACGT-3′ (SEQ ID NO:43) and 5′P-CTGAACCATCAGGGC-3′ (SEQ ID NO:44). 5 μg of oligonucleotide SEQ ID NO:43 and 3 molar equivalents of oligonucleotide SEQ ID NO:44 were mixed together in the presence of 0.1 M NaCl in a final volume of 50 μl. The mixture was boiled and immediately cooled down to room temperature in the same boiler. The single strand random nucleotide part of the DNA hybrid cassette formed by annealing of the 2 oligos was extended using 400 μM of each dNTPs and 60 U/ml of Kienow DNA polymerase (New England Biolabs, Inc., Beverly, Mass.) in a final volume of 200 μl in 1× EcoPoI buffer (New England Biolabs, Inc., Beverly, Mass.). The extension reaction was incubated 20 minutes at 37° C. and further purified using QIAEX II beads as described by the manufacturer (Qiagen, Study City, Calif.). 60 μg of pEA832 (50 μg/ml) were digested in 1× Buffer 4 (New England Biolabs, Inc., Beverly, Mass.) using 400 U/ml of AatII (New England Biolabs, Inc., Beverly, Mass.) and 500 U/ml of ClaI (New England Biolabs, Inc., Beverly, Mass.) in the presence of 100 μg/ml of BSA. The digestion was performed at 37° C. for 2 hours. Synthesized random cassettes (20 μg/ml) were digested in 1× Buffer 4 (New England Biolabs, Inc., Beverly, Mass.) using 500 U/ml of ClaI (New England Biolabs, Inc., Beverly, Mass.) in the presence of 100 μg/ml of BSA. Cassettes were further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Digested plasmid DNA was electrophoresed on a 0.7% agarose gel and the excised bands further purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.). Ligation was carried out at 16° C. for 1 hour using a 1:1 ratio of vector (2 ng/μl) to insert and 1,600 U/ml of T4 ligase (New England Biolabs, Inc., Beverly, Mass.). Ligation products were electro-transformed into E. coli strain ER2744 (New England Biolabs, Inc., Beverly, Mass.) competent cells (competency of 1×109 pEA835 transformants/μg) using 2 μg of total ligation product for each 200 μl aliquot of competent cells, at 2.5 kV/cm in a 2 mm cuvette (BIORAD, Richmond, Calif.). Cells were allowed to recover in a shaker for 1 hour at 37° C. Recovered transformants were inoculated at 1/100 dilution into LB liquid growth media containing 100 μg/ml of ampicillin and 1 mM IPTG. Transformants were incubated overnight at 30° C. Plasmid DNA was isolated from the overnight culture using tip100 columns (QIAGEN, Studio City, Calif.)), AatII/ClaI digested as above and electrophoresed on a 4% GTG Nusieve agarose gel (FMC BioProducts, Rockland, Me.). The 57 to 72 bp spectrin loop DNA inserts were purified using QIAEX II beads as described by the manufacturer (Qiagen, Studio City, Calif.) and cloned back into AatII/ClaI digested and purified selection plasmid (pEA832) as described above. This protocol was repeated 3 times to enrich the pool of transformants for peptide clones having the most biologically active sequences against the protein splicing of the Mtu DnaB intein (IVPS). Finally selected spectrin modules were cloned into a pEA832 homologous plasmid containing the wild type Mtu dnaB gene (pEA825) and grown individually in 10 ml LB containing 100 μg/ml ampicillin at 37° C. and induced with 1 mM IPTG for 3 hours. Crude protein cell extracts were electrophoresed on a 10-20% gradient gel (Novex, San Diego, Calif.). The gel was further electro-blotted for immuno-staining using anti-T7 tag antibodies (Novagen, Madison, Wis.) to detect Mtu DnaB protein splicing products (FIG. 5C). Lane Eco DnaB contains extracts of T7-tagged E. coli DnaB without the intein. pMtuDnaB contains extracts from a clone expressing only Mtu DnaB. Lanes p814, p815, p816, p817, and p818 contain extracts of the clones pEA814, pEA815, pEA816, pEA817, and pEA818, respectively, encoding peptides selected for inhibition of splicing. To demonstrate that the inhibition of splicing was due to the peptide inserted into the chicken α-spectrin loop, the selected sequence of pEA818 was replaced with the spectrin sequence, DLPMVEE (SEQ ID NO:10) to generate clone pEA818rev, and extracts loaded on lane p818rev. Splicing of pEA818rev occurred as efficiently as with the pMtu DnaB clone that expresses the wild-type spectrin protein. Note that in the absence of splicing, much of the DnaB precursor undergoes cleavage at the intein C-terminal splice junction.

[0106] The sequence of the inserted peptides in these clones is as follows:

	pEA814	TVQSTKR	(SEQ ID NQ:5)
	pEA815	RPAPRPL	(SEQ ID NO:6)
	pEA816	PTARTYE	(SEQ ID NO:7)
	pEA817	PTRPTAPPLNFS	(SEQ ID NO:8)
	pEA818	HPNPHPTLSGQR	(SEQ ID NO:9)
	pEA818rev	DLPMVEE	(SEQ ID NO:10)

[0107] We have thus demonstrated that this system can be used to select for peptides that block splicing of the Mtu DnaB intein. This system is amenable to selection of any modulators of splicing of the Mtu DnaB intein or other DnaB inteins, as long as the agent can enter a cell.

EXAMPLE III

In Vivo Control of Protein Splicing for Chemotherapeutic Purposes or to Make Controllable Gene Knockouts

[0108] The selection and screening systems described for selection of agents that modulate protein splicing can also be applied to intein-less versions of the extein gene to select for agents that inhibit or activate the extein gene product. All of the selection and screening systems described in this patent are based on the activity or inactivity of the extein portion of the precursor. If one deletes the intein from the intein-containing gene by methods known to one skilled in the art, then one can select for agents that block or activate extein activity also using the methods described for inhibiting or activating splicing of the intein containing precursor, since these latter methods involve assaying extein function. For example, if one deletes the intein from the Mtu DnaB gene by methods known to one skilled in the art, then one can select for agents that block activity of the cytotoxic Mtu DnaB protein using the methods described for inhibiting splicing of the DnaB intein. M. tuberculosis can then be attacked using a cocktail of two agents that block activity of the essential DnaB protein, making it more difficult for the organism to develop resistance to these agents.

[0109] We have previously described the insertion of a CIVPS or IVPS into a foreign gene. In these cases, protein splicing could be controlled by temperature, mutation, pH, photo-activated blocking groups, phosphorylation or peptides (Comb, et al., U.S. Pat. No. 5,834,247 and Comb, et al., U.S. Pat. No. 5,496,714). In this Example we describe a general method for selecting specific protein splicing inhibiting or activating agents that are capable of controlling protein splicing in vivo or in vitro. The methods are equally applicable to genetic selection systems or reporter systems. By genetic selection, we mean, in this Example, that viability or growth rate of the test organism is monitored during the experiment, while a reporter system in this Example refers to the monitoring of a marker, such as color detection, fluorescence, phenotype, etc., rather than cell viability. Genetic selection or reporter systems are used to identify agents that can either disrupt or catalyze protein splicing of a given intein, depending on the context of the experiment. Any genetic selection or reporter system known to one skilled in the art can be used to isolate agents which disrupt or catalyze protein splicing. This strategy is equally applicable to any intein present in a foreign context or in its native or homologous context (e.g., the insertion of an intein at the same position in an homologous extein). However, use of the native extein is preferable because it best represents the enzyme target of the intein. If the native precursor does not express well or splice well in the experimental host organism, then the intein can be inserted into the same site in that host organism's homolog of the native extein or in another extein homolog with desired properties for testing, using any method known to one skilled in the art or described in the previous Examples. This method of finding agents that modulate splicing is applicable to any host, as long as the protein splicing precursor is operably linked to the appropriate control signals for transcription and translation in that host. As the target organism may not be an easy experimental model for identifying agents that modulate protein splicing, the agent may first be identified in a model system and then tested in the final target organism. This strategy is summarized in FIG. 8.

[0110] Experiments involving inhibition of splicing start with a precursor that contains a fully active intein that may or may not be controllable. The goal of this experiment is to find agents that can be used to control splicing of this intein. In experiments involving activation of splicing, a CIVPS (controllable intein) or an inactive intein is required, as the goal is to find agents that activate the previously inactive intein. The intein may be inactivated by any means known to one skilled in the art, such as temperature sensitive inteins, inteins with mutations in amino acids known to be involved in catalysis that slow down or block splicing (including the conserved amino acids at both splice junctions and in intein Block B, (Perler, Nucleic Acids Res. 25:'087-1093 (1997), Pietrokovski, Protein Sci., 3:2340-2350 (1994)) inteins which have been randomly mutated and selected for inhibition or blockage of splicing.

[0111] A positive selection system is preferred. In general, a positive selection system consists of a gene that is detrimental to a host organism depending on the growth media or the host strain genetic background. The gene product is static or lethal for the cell, killing the host or preventing growth unless the gene product is inactivated. The gene product may be directly cytotoxic to the host in a dominant manner, as in the DnaB example (Example II) or it may be dominantly cytotoxic in response to a drug which the chromosomal copy of the gene is resistant to, as in the GyrA example (Example I). By dominant or dominantly cytotoxic, we mean that the toxicity occurs even if homologous proteins are present which are not cytotoxic or resistant to the drug, i.e., the cytotoxic effect dominates irrespective of the presence of non-cytotoxic homologs. In the context of a protein splicing inhibition system, positive selection involves a system that allows selection against the splicing of an IVPS or intein. If splicing occurs, the cytotoxic extein protein will be active and kill the cell or inhibit growth; if splicing is disrupted, the cytotoxic extein protein will be inactive and cells will grow. Cell growth can be monitored by any means known to one skilled in the art, including, but not limited to obervation of a colony on solid media, optical density, monitoring of fluorescent reporters of cell growth such as green fluorescent protein of luciferase activity. The extein gene may be an unrelated reporter system or the natural extein of the intein (either using the natural precursor or inserting the intein into a homologous extein context) (FIG. 8). In this context, selection systems have the advantage that only agents that inhibit splicing allow cell growth and are thus easily found amongst the background of agents that have no effect on splicing or are directly toxic to the cell. If the agents to be tested are also expressed in the host cell, then one examines the colonies that survive on the plate. If the agent to be tested is not expressed in the host cell, but is instead added to the media, then aliquots of host cells must be arrayed for testing with individual agents or pools of agents in any number of devices, such as microtiter dishes. In such cases, cell growth may be more easily measured if the cells expres a protein that leads to fluorescence, such as green fluorescent protein or luciferase.

[0112] When selecting for agents that activate splicing, the intein is already present or is inserted into a gene whose protein product is required for cell growth. In the absence of splicing, the cell fails to grow or dies. In order to practically employ this selection system, a second gene is present which can rescue the cell in the absence of splicing. This second copy of the gene should be controllable, by methods such as a temperature sensitivity or controllable promoters, to allow cell growth until the agent which activates splicing is applied or induced in the cell. The cells are treated with the splicing activator and then moved to the nonpermissive condition for activity of the second gene product that does not contain the intein or expression of this gene is turned off. Cell growth will then require splicing since the second gene product lacking the intein is no longer active.

[0113] Another method for identifying agents that modify protein splicing involves screening rather than genetic selection. Screening systems employ reporter genes whose products can be readily assayed, but do not necessarily affect cell growth. Many reporter systems are known, such as the blue/white β-galactosidase screening system. β-galactosidase acts on X-gal, for example, to generate a blue color; in the absence of β-galactosidase activity, the X-gal remains uncolored or ‘white’. Other reporters include those described in Burns and Beacham, Gene, 27:323-325 (1984) and Mechulam, et al., J. Bacteriol., 163:787-791 (1985). One can use native precursors if reporter systems are available for those extein genes or the intein can be cloned into the reporter gene (β-galactosidase in this Example) (see for example, Belfort, U.S. Pat. No. 5,795,731 and Comb, et al., U.S. Pat. No. 5,834,247). Agents that inhibit splicing of an otherwise active intein will block reporter protein functions, such as β-galactosidase action on X-gal, resulting in white instead of blue clones. Agents that activate splicing of an otherwise inactive or slowly acting intein restore reporter protein functions, resulting in blue clones using the β-galactosidase system as an example. If the agents to be tested are also expressed in the host cell, then one examines the colonies that survive on the plate. If the agent to be tested is not expressed in the host cell, but is instead added to the media, then aliquots of host cells must be arrayed for testing with individual agents or pools of agents in any number of devices, such as microtiter dishes. Unlike selection, all cells grow in reporter systems and one must determine whether the read out is positive or negative for each colony or microculture.

[0114] Previous Examples have described genetic selection systems based on the pheS non-homologous selection system and the gyrA and dnaB intein/extein systems. This Example describes how one would screen for agents that modulate splicing using any selection or reporter system. Note that the selection or screening systems may not have been originally identified in the organism containing the intein. However, if a selection or screening system has been described for the extein homolog, it can be adapted to the intein containing homolog. As in the case of DnaB, the same mutation can be made in the intein containing homolog to generate a selectable phenotype for the intein containing extein gene. As in the case of GyrA, the screening system can involve a chromosomal mutation that leaves the host resistant to a drug; all that need be done is to show that the intein containing homolog is also sensitive to the drug.

[0115] Iterative screening (FIG. 5B) provides a method of identifying lead compounds and reducing background and can be used in any of the schemes described below. Iterative screening involves repeated cycles of testing of the agent on fresh extein genes. It helps insure that the agent is not acting on a mutated extein, which could also be a by-product of screening.

[0116] Positive Selection Systems for Inhibition or Activation of Protein Splicing of an Intein in its Natural Precursor or an Extein Homolog

[0117] In this case, the intein of interest is naturally found in a target gene which can naturally serve as a selectable marker or reporter or which can be converted into a selectable marker or reporter. Initial experiments may be performed in the target organism or an experimentally more amenable model host such as bacteria, E. coli, yeast, mammalian cells, insect cells, etc. The decision as to whether to use the natural splicing precursor to select for agents that block splicing or to first insert the intein gene into a homologous extein gene from a model organism depends on the similarity amongst the extein genes, the ability of the natural precursor or recombinant precursors to express in the model hosts used for selection or screening, and the ability of each precursor to splice in the model hosts. (FIG. 8) These parameters will have to be experimentally determined, although the more similar the extein sequences, the more likely that splicing will work in the homologous extein protein from the model organism. Sequence comparison will indicate the appropriate homologous intein insertion site in the homologous extein gene from the model organism.

[0118] Next, one has to determine by a literature search whether any genetic selection systems or screens are available for the target extein in any organism and whether the extein gene is essential for cell growth in any organism. If the target gene is essential, but no genetic selection or screens are available, it can be mutagenize directly or in model systems to attempt to generate a selection or reporter system. If the target gene product is essential to the cell, under defined conditions, the host gene can be either knocked out and replaced by a controllable copy of the gene or mutated to generate a temperature sensitive activity. The intein containing gene must the produce an active product when the host gene homolog is inactivated. A temperature sensitive phenotype can easily be generated by random or rational mutation by one skilled in the art. Once a selection system has been identified and the best splicing precursor has been determined (selecting from the naturally occurring precursor, or after inserting the intein into the homologous extein from the target or selection organism), testing for agents that block splicing can begin in either a model organism or the target organism, depending on ease of use. Some of the possible schemes for identifying agents that block or activate splicing are shown in FIG. 9.

[0119] Scheme 1 is a method for selecting for agents that inhibit splicing. The selection system involves a dominant cytotoxic phenotype in response to a drug. By dominant cytotoxic we mean that the spliced product is toxic to the cell irrespective of expression of a resistant copy of the extein gene. The GyrA system described in Example I is an example of this type of scheme. The selection host organism contains a chromosomal copy of the extein gene that is resistant to the drug and allows growth of the organism in the presence of the drug. First, a merodiploid is made containing a gene which is sensitive to the drug and contains the intein, and a gene which is resistant to the drug and does not contain an intein. Second, the host containing the resistant extein gene and the intein containing sensitive extein gene is then treated with agents that can enter the cell or by induction of expression of agents within the cell. Finally, the selection drug is added to the cells. If the intein splices, the drug sensitive target protein kills the cell or inhibits growth when the drug is present. If any agent blocks splicing, no drug sensitive extein protein is made and the organism grows. Usually, one tests a library of compounds of any type, rather than a single agent, and one uses small cultures, as in microtiter dishes, for example. Any type of agent can be used, as long as it can enter the cell. Alternatively, the agent can be cloned and expressed in the target cell and clones can be tested for growth on plates or in liquid media. Expression of combinatorial peptide libraries would be an example of such an agent that is expressed in the cell.

[0120] Scheme 2 is a second method for selecting for agents that inhibit splicing. The selection system involves a dominant lethal phenotype in the absence of exogenous drug treatment that is inherent in the intein containing extein protein or can be introduced into the extein protein. The DnaB system described in Example II exemplifies this type of system. The selection host organism contains a wild type gene that is not toxic to the cell and allows growth of the organism. First, a mero-diplid is made containing a gene which is toxic to the cell, but contains an intein and an intein-less extein gene which is not toxic. Next, this host is treated with agents that can enter the cell before the cytotoxic precursor gene is expressed. Finally, expression of the intein containing cytotoxic extein gene is induced. If the intein splices, the cytotoxic target protein kills the cell or inhibits growth. If any agent blocks splicing, no cytotoxic target protein is made and the organism grows. Usually, one tests a library of compounds of any type, rather than a single agent, and one uses small cultures, as in microtiter dishes, for example. Any type of agent can be used, as long as it can enter the cell. Alternatively, the agent can be cloned and expressed in the target cell and clones can be tested for growth on plates or in liquid media. Expression of combinatorial peptide libraries would be an example of such an agent that is expressed in the cell.

[0121] Scheme 3 selects agents that inhibit splicing of an essential gene. In this case, the chromosomal copy of the gene, or its equivalent, is either temperature sensitive, sensitive to a drug in a recessive manor, or under some type of expression control. Alternatively, the chromosomal copy of the extein gene is inactivated or knocked out. The cells can grow under conditions where the gene product is not needed. The cells are then shifted to conditions which require the extein protein for survival. An example of this type of extein is a metabolic enzyme. When cells are grown in rich media, they can grow. However, when cells are grown in minimal media or media lacking the downstream product of the extein blocked metabolic pathway, the cells fail to grown. The intein containing target gene is not temperature sensitive or is resistant to the drug. If splicing occurs under non-permissive conditions for the chromosomal extein homolog, then the cells live. This system requires assay of cell growth in isolated containers, such as microtiter dish wells, for example. If the agent blocks splicing, then the cells will not grow under non-permissive conditions for activity of the intein-less copy of the extein protein. Cell growth can be determined by any means known to one skilled in the art, including, but not limited to measuring optical density or presence of a fluorophore generated in the cell. First an experimental host must be found that contains a controllable copy of the extein gene or its equivalent. It is propagated under permissive conditions for expression of active intein-less extein protein. Second, this host is transformed with a vector containing a wild type extein gene or extein homolog gene containing the intein. Third, merodiploid cells containing the intein-plus and intein-minus copies of the extein gene, or its equivalent, are treated with agents to block splicing and are also shifted to non-permissive conditions for activity of the intein-less extein protein. This may involve a shift to a temperature at which the intein-minus protein is inactive, removal of inducers for expression of the intein-minus shifting to different media, or addition of a drug which inactivates the intein-minus protein. If splicing occurs, the cells will continue to grow using the intein-plus gene product. However, if the agent inhibits splicing, products of both copies of the gene are inactivated and the cells die. Alternatively, the agent can be cloned and expressed in the target cell. However, in this case, each clone must be copied or replica plated to maintain a living copy of the library and a copy to be tested for inhibition of splicing. Expression of combinatorial peptide libraries would be an example of such an agent that is expressed in the cell.

[0122] Schemes 4-6 are methods of selecting for agents that activate splicing rather than inhibit it. The precursor contains an inactive intein which is introduced into the cell on any type of vector. The agent(s) may be added individually or in pools to isolated cultures. Alternatively, the agent can be cloned and expressed in the target cell. However, in this latter case, each clone must be copied or replica plated to maintain a living copy of the library and a copy to be tested for activation of splicing. Expression of combinatorial peptide libraries would be an example of such an agent that is expressed in the cell.

[0123] Scheme 4 is identical to scheme 1. In the presence of the drug, an agent that activates splicing kills the host since the intein-plus drug sensitive copy of the gene is active and dominantly cytotoxic. One assays for the absence of growth in isolated cultures, such as microtiter dish wells, for example.

[0124] Scheme 5 is similar to scheme 1. An agent that activates splicing kills the host since the dominantly cytotoxic extein is active after splicing of the intein, irrespective of the presence of the wild type extein protein derived from the intein-minus gene. One assays for the absence of growth in isolated cultures, such as microtiter dish wells, for example.

[0125] Scheme 6 is similar to scheme 3, except that the selection system requires expression of the spliced target gene for cell growth and selects for agents that activate splicing. In this type of system, the intein-minus copy of the target extein gene, or its equivalent, is either temperature sensitive, sensitive to a drug in a recessive manor, or under some type of expression control. Alternatively, the chromosomal copy of the extein gene is inactivated or knocked out. The cells can grow under conditions where the gene product is not needed. The cells are then shifted to conditions which require the extein protein for survival. An example of this type of extein is a metabolic enzyme. When cells are grown in rich media, they can grown. However, when cells are grown in minimal media or media lacking the downstream product of the extein blocked metabolic pathway, the cells fail to grow. The intein containing target gene is not temperature sensitive or is resistant to the drug. The target gene containing the intein is introduced into the cell by any means known to one skilled in the art. In this case, the intein has been modified so that it can not splice under the assay conditions. The host copy of the gene is expressed in an active form under permissive conditions (permissive temperature, in the absence of drug, rich media under permissive expression conditions, etc.), allowing the cells to grow. The intein-plus copy of the target extein gene, containing the inactive intein, is introduced into the cell. After expression of the intein precursor is established, agents are added externally or peptide libraries are expressed internally to induce splicing. After allowing the agent to activate splicing, the cells are shifted to the nonpermissive condition (non-permissive temperature, in the presence of drug, minimal media under non-permissive expression conditions, etc.). The only cells that can grow are those in which splicing activity has been restored by the agent. If an external agent is to be tested, then the agent is added to cells in isolated containers, such as microtiter dish wells. Alternatively, the agent can be cloned and expressed in the target cell. In this case, the library of agents can be directly tested for cell viability on plates. Expression of combinatorial peptide libraries would be an example of such an agent that is expressed in the cell.

[0126] Reporter Systems for Inhibition or Activation of Protein Splicing of an Intein in its Natural Precursor or an Extein Homolog

[0127] Any extein that can be converted into a tractable phenotype can be used in a reporter system screen. This type of system requires the ability to differentiate between active and inactive extein by any direct or indirect means. Once the reporter system is available, the intein containing gene is introduced into the cell by any method known to one skilled in the art and agents that inhibit splicing are added or induced as above. Alternatively, an inactive intein is introduced into a cell and agents that activate it are added or induced as above. One then examines individual clones and determines whether the extein is active or not.

[0128] Systems for Inhibition or Activation of Protein Splicing of an Intein in an Unrelated Extein Context

[0129] The scenario for this method of identifying agents that inhibit or activate splicing is the same as schemes described above, except that the intein is placed in an unrelated extein. However, one must first determine that the intein splices in the non-homologous extein (FIG. 8). To improve the probability that an intein will splice in a non-homologous foreign context, the intein insertion site should be as similar to the natural extein sequence as possible for at least 1 and up to 5 or more extein residues. If the intein is inserted into a nonessential region of the target protein, one could possibly modify the sequence of the target protein at the intein insertion site to be the same as the native extein sequence of that intein. The intein must be cloned prior to a Ser, Thr or Cys with the amino acid naturally following the intein being the best choice or the Ser, Thr or Cys codon must be inserted into the extein along with the intein sequence. To improve folding, surface locations on the protein would be preferable since they are more likely to allow the extein to fold independently of the intein. If the structure of the target protein is unknown, protease sensitive sites on the target protein should be good positions to insert the intein.

[0130] Since splicing can be sequence dependent, it is optimal to experimentally identify agents that modify splicing in the same target protein that one wants to finally control. However, agents could possibly also control splicing of that intein in any extein. New exteins may have to be treated experimentally.

[0131] Controllable Knockouts

[0132] Once an agent has been found which can inhibit or activate splicing, the homologous extein gene in the target organism is replaced by the homologous gene containing the intein by methods known to one skilled in the art. for example, this may be performed in a one step process by inserting the intein-containing gene directly into the chromosomal copy of the extein gene by homologous recombination. Alternatively, the intein containing gene is introduced into the organism and the non-intein containing homolog is inactivated either concurrently or separately and in any order of event. Once the only copy of the active extein gene contains a intein, gene function can be inhibited if the organism is treated with an agent that blocks splicing. On the other hand, if a splicing impaired intein is used, gene function can be activated if the organism is treated with an agent that activates splicing. The agents and splicing can be modulated at any time during the development and life of the organism by addition or removal of the splicing activating or inhibiting agent. For example, a gene for mouse embryogenesis can be replaced by an intein containing gene homolog and the product of that gene can be activated or inactivated at various times to determine when the gene product is required and if it is required during multiple stages of development or growth. In a second example, a gene product thought to be required for passage through a specific stage of the cell cycle could be replaced with an intein containing copy that would allow study of exactly when the gene product is required or to synchronize the culture by arresting all cells at the same point in the cell cycle to study the effect of any agent, etc., on a synchronized culture of cells.

[0133] Use of Controllable Inteins (CIVPS) in Therapeutics

[0134] Several options can be envisioned for the use of controllable splicing to deliver active proteins at specific times or to specific places. In many instances, therapeutic drugs can be cytotoxic to the host and would be best if only active at the target site. For example, chemotherapy drugs are often generally cytotoxic and adverse reactions in normal cells could be eliminated if the drug could be specifically activated in the tumor. If one has a drug that is at least partially proteinacious, an intein that can be activated or inhibited by a second agent, as described above, could be inserted into the protein portion of the therapeutic agent. The drug is then administered systemically in an inactive form. The drug could then be specifically activated in the tumor or target organ by (1) injecting the activating agent into the tumor, (2) exposing the tumor to laser treatment to increase the temperature of the tumor and thus induce splicing of a temperature sensitive intein, (3) use gene therapy to target the inactive cytotoxic precursor to the tumor cells and then add the splicing activator systemically or (4) use gene therapy to target the activating peptide to the tumor and add the inactive intein containing drug systemically. In the Examples described above, the inactive cytotoxic precursor or the activating peptide, respectively, could be transformed systemically with a vector that is not tissue or cell specific, and only expressed in specific target cells by operably linking these genes to tissue specific promoters.

EXAMPLE IV

Methods for Generating Temperature Controllabel Inteins

[0135] The methods used for identifying agents that inhibit or activate splicing can also be used to identify inteins that are active at one temperature and inactive at a second temperature (referred to as temperature sensitive inteins). Instead of adding an external agent or expressing an internal agent, the intein is randomly mutated by any method known to one skilled in the art, such as error prone polymerase chain reaction (FIG. 10) or use of combinatorial DNA sequences at specific regions in the intein. Alternatively, one can specifically mutate residues thought to function in or assist the chemical reactions, such as the C-terminal splice junction residues, the intein N-terminus, the intein penultimate residue, the residues in intein Block B (Perler, Nucleic Acids Res., 25:1087-1093 (1997); Perler, Nucleaic Acids Res., 27:346-247 (1999); Pietrokovski, supra), residues proximal to the intein active site as determined crytallographically (Duan, et al., Cell, 89:555-564 (1997); Klabunde, et al., Nat. Struct. Biol., 5:31-36 (1998)), etc. The mutated intein gene is then introduced into a cell and examined for the ability to splice under permissive and non-permissive temperatures as chosen by the researcher, and can be any combination of temperatures (FIG. 11). Splicing is assayed as in Examples I through III as long as the chromosomal or intein minus extein gene is not similarly temperature sensitive.

[0136] Using the Mxe GyrA intein in the E.coli GyrA extein and expressing the fusion in E.coli cells (Example I), we have identified several polymerase chain reaction generated mutations that render splicing of the Mxe GyrA intein temperature sensitive (FIGS. 10, 11, 12 and 13). These precursors splice at 19° C., but not at 37° C. Moreover, these mutations concentrate in the beta-sheet that includes intein Block B (FIGS. 12 and 13).

[0137] Screening for Temperature Sensitive Mxe GyrA Intein Mutants

[0138] The gyrA selection system described in Example I, can also be used to screen for temperature sensitive splicing mutants of the Mxe GyrA intein in the ofloxacin sensitive E. coli GyrA extein. Experiments were performed with a vector similar to pEA600. A splicing proficient clone and a splicing deficient clone (containing mutation of the intein Cys1 to Ala and Asn198 to Ala) were plated on solid media containing various concentrations of ofloxacin to determine the appropriate drug concentration to allow growth of the splicing deficient clone while blocking growth of the splicing proficient clone. The Mxe gyrA intein gene was then amplified by PCR (FIG. 10) using mutagenic strategies known to one skilled in the art and inserted into the E. coli gyrA gene. Libraries were plated on solid media containing ofloxacin at the predetermined concentration, replica plated and grown at either 37° C. or 16° C. (FIG. 11). Only splicing defective clones survived and grew on the plates. The replica plates were compared to identify clones that grew at 37° C., but not at 16° C. Such clones were picked and retested for temperature dependent splicing. Alternatively, the libraries of mutated Mxe GyrA inteins in E. coli GyrA were grown at 37° C. and then streaked onto a second plate to test for lack of growth at the splicing permissive temperature of 16° C. Splicing of the GyrA precursor was examined in clones that failed to grow at 16° C. by incubating in the absence of ofloxacin at 37° C. for 3 hours and then shifting to 16° C. overnight. Cell lysates were electrophoresed in SDS-PAGE gels that were then stained with Coomassie blue. Spliced GyrA was observed in several clones, although splicing was not complete (FIG. 12).

[0139] The Mxe gyrA intein gene was sequenced from several of these temperature sensitive clones and found to have one or more mutations which are summarized in FIG. 13. The 3-D structure of the Mxe GyrA intein is known (Klabunde, et al., Nature Struct. Biol. 5:31-36 (1998))., GyrA enabling us to place these mutations on the Mxe GyrA intein structure (FIG. 14). We found that many of the mutations were in the beta-sheet including intein Block B (FIGS. 13 and 14), specifically in Mxe GyrA intein beta-strand B8 and the loop between beta-strands B8 and B9 (Klabunde, supra; Perler Cell 92:1-4 (1998)). intein Block B contains conserved intein residues thought to assist in the autocatalytic reactions at the intein N-terminal splice junction (Klabunde, supra; Noren, C. J., et al. Angewandte Chemie (in press)). Mutation in residues proximal in space to intein Block B, as found in this selection for temperature sensitive Mxe GyrA intein mutants, may slightly perturb the position of Block B residues, resulting in the temperature sensitive phenotype.

[0140] We suggest that mutation of the amino acids in the analogous beta-strand and loop in other inteins may generate temperature sensitive mutants of any intein. Homologous regions in other inteins can be easily identified due to the structural similarity of known intein splicing domains and intein multiple sequence alignments. To date, the 3-D structure of the Mxe GyrA intein (Klabunde, supra), the Sce VMA intein (Duan, et al., Cell 89:555-564 (1997)) and the Drosophila hedgehog protein autoprocessing domain (Hall, et al. Cell, 91:85-97 (1997)) have been determined. The splicing domain of both inteins and the N-terminal part of the hedgehog autoprocessing domain have the same protein fold; the alpha carbon trace of most of the amino acids in each of these 3 structures are superimpossible (Klabunde, supra; Perler, supra (1998)). Intein amino acid sequence similarity comparisons have also been described in the literature (Perler supra (1997), Pietrokovski, supra(1994), Pietrokovski, Protein Sci. 7:64-71 (1998), Dalgaard, et al., J. Comp. Biol. 4:193-214 (1997)).

[0141] Given the similarity in intein splicing domain structure and sequence, one skilled in the art should easily be able to identify regions in any intein that are analogous to the Mxe GyrA intein beta-strand B8 and the loop between beta-strands B8 and B9, and using this information, mutate this region to specifically generate temperature sensitive protein splicing mutants.

Claims

1. A positive genetic selection system employing a precursor comprising a native intein in its natural or homologous modifiable extein context for the screening of agents which inhibit or activate protein splicing, said selection system comprising: (1) a host cell which contains a first gene encoding a non-selectable form of a target enzyme, and (2) a second gene encoding a selectable form of said target enzyme which is dominantly cytotoxic upon interaction under predetermined selection conditions, said second gene containing an intein, wherein the inhibition or activation of said selectable form of said target enzyme by a given agent affects the viability of said host cell.

2. The positive genetic selection system of claim 1, wherein the activation of said selectable form of said target enzyme by a given agent results in the death of the host cell.

3. The positive genetic selection system of claim 1, wherein the inhibition of said selectable form of said target enzyme by a given agent results in the viability of the host cell.

4. The positive genetic selection system of claim 1, wherein said host cell contains a first gene encoding a drug-resistant form of the target enzyme and a second gene encoding a drug-sensitive form of the target enzyme which is dominantly cytotoxic upon interaction with said drug.

5. The positive genetic selection system of claim 1, wherein said first gene encodes a wild type form of said target enzyme and said second gene encodes a dominant cytotoxic form of said target enzyme.

6. The positive genetic selection system of claim 4, wherein in the absence of a silent mutation of said extein, said intein is selected from the group consisting of an intein inserted into the drug-sensitive form of said target enzyme and a natural intein in a mutated native or homologous extein, wherein said mutation renders the extein cytotoxic upon interaction with said drug.

7. The positive genetic selection system of claim 1, wherein said agent comprises an in vivo peptide library or derivatives thereof.

8. The positive genetic selection system of claim 4, wherein said intein comprises the M. xenopi GyrA intein, said homologous extein comprises E.coli GyrA and said host cell comprises E. coli.

9. The positive genetic selection system of claim 8, wherein said drug-resistant form of said target enzyme is the Ser83 mutant of E.coli GyrA.

10. The positive genetic selection system of claim 7, wherein said peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

11. The positive genetic selection system of claim 5, wherein in the absence of a silent mutation in said extein, said intein is selected from the group consisting of an intein inserted in to the cytotoxic form of said target enzyme and a natural intein in a mutated native or homologous extein, wherein said mutated native extein is cytotoxic.

12. The positive genetic selection system of claim 11, wherein said agent comprises an in vivo peptide library or derivatives thereof.

13. The positive genetic selection system of claim 11, wherein said natural intein is the M. tuberculosis DnaB intein and said mutated native extein is the M.tuberculosis R231C mutant, and wherein said host cell is E. coli.

14. The positive genetic selection system of claim 13, wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

15. A method of screening for agents which inhibit protein splicing, said method comprising the steps of: (a) creating a positive selection system comprising a host cell containing a gene encoding a dominantly cytotoxic protein containing an intein; and (b) culturing the host cell of step (a) in the presence of test agents, wherein the inhibition of splicing of said cytotoxic protein results in viability of said host cell.

16. The method of claim 15, wherein said agent is expressed within the host cell as a protein or portion thereof and said agent is identified by the gene encoding said agent from said surviving host.

17. The method of claim 15, wherein said positive selection system of step (a) comprises the positive genetic selection wherein said host cell contains a first gene encoding a drug-resistant form of the target enzyme and a second gene encoding a drug-sensitive form of the target enzyme which is dominantly cytotoxic upon interaction with said drug, and wherein step (b) further comprises culturing said host cell of step (a) in the presence of said drug.

18. The method of claim 15, wherein said positive selection system of step (a) comprises a first gene encoding a wild type form of said target enzyme and a second gene encoding a dominant cytotoxic form of said target enzyme.

19. The method of claims 17 or 18, wherein said host cell of step (a) expresses an in vivo peptide library or derivatives thereof, and wherein said test agents of step (b) comprise said in vivo peptide library or derivatives thereof.

20. The method of claim 19 wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

21. A method for screening of agents which activate protein splicing, said method comprising the steps of: (a) creating a positive selection system comprising a host cell containing a gene encoding a dominantly cytotoxic protein, said gene containing an inactive intein; and (b) culturing said host cell of step (a) in the presence of individual test agents, wherein the activation of splicing of said cytotoxic protein results in host cell death.

22. The method of claim 21, wherein said agent is controllably expressed within the host cell as a protein or portion thereof and said agent is identified by the identification of the gene encoding said agent in a parallel sample of said host cell in which expression of said agent was not activated.

23. The method of claim 21, wherein said positive selection system of step (a) comprises the positive genetic selection system wherein said host cell contains a first gene encoding a drug-resistant form of the target enzyme and a second gene encoding a drug-sensitive form of the target enzyme which is dominantly cytotoxic upon interaction with said drug, and wherein step (b) further comprises culturing said positive selection system of step (a) in the presence of said drug.

24. The method of claim 21, wherein said positive selection system of step (a) comprises the positive genetic selection system wherein the drug-resistant form of said target enzyme is the Ser83 mutant of E.coli GyrA.

25. The method of claims 21 or 22, wherein said host cell of step (a) expresses an in vivo peptide library or derivatives thereof, and wherein said test agents of step (b) comprise said in vivo peptide library or derivatives thereof.

26. The method of claim 25, wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

27. A positive genetic selection system for the screening of agents which inhibit protein splicing, said selection system comprising: (1) a host cell which contains a first gene encoding a controllable form of a target enzyme which is required for cell viability, and (2) a second gene encoding an expressed form of said target enzyme, said second gene containing an intein, wherein the inhibition of splicing of said target enzyme by a given agent results in the reduced viability or death of said host cell under conditions which do not permit the expression of said controllable first gene of said target enzyme.

28. The selection system of claim 27, wherein said controllable form of said target protein is selected from the group consisting of a drug-sensitive target protein, an inducer-sensitive target protein, a temperature-sensitive target protein, and a target protein operably linked to a controllable promoter.

29. The selection system of claim 27, wherein said intein is selected from the group consisting of a foreign intein inserted into the homologous extein of said target enzyme and a natural intein in a native extein.

30. A positive genetic selection system for the screening of agents which activate protein splicing, said selection system comprising: (1) a host cell which contains a first gene encoding a controllable form of a target enzyme required for cell viability, and (2) a second gene encoding an expressed form of said target enzyme, said second gene containing an inactive intein, wherein the activation of splicing of said target enzyme by a given agent results in the survival of said host cell under conditions which do not permit the expression of said controllable first gene of said target enzyme.

31. The selection system of claim 30, wherein said controllable target protein is selected from the group consisting of a drug-sensitive target protein, an inducer-sensitive target protein, a temperature-sensitive target protein, and a target protein operably linked to a controllable promoter.

32. The selection system of claim 30, wherein said inactive intein is selected from the group consisting of a foreign inactive intein inserted into the homologous extein of said target enzyme and a natural inactive intein in a native extein.

33. The method of claims 27 or 30, wherein said host cell of step (a) expresses an in vivo peptide library or derivatives thereof, and wherein said test agents of step (b) comprise said in vivo peptide library or derivatives thereof.

34. The method of claim 33, wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

35. A method of screening for agents which inhibit protein splicing, said method comprising the steps of: (a) culturing the positive selection system of claim 27 in the presence of test agents under non-permissive conditions; and (b) identifying non-surviving host cells from step (a), wherein said agent inhibits protein splicing.

36. The method of claim 35, wherein said agent is expressed within the host cell as a protein or portion thereof and wherein said non-surviving host cells of step (b) contain test agents which inhibit protein splicing.

37. A method of screening for agents which activate protein splicing, said method comprising the steps of: (a) culturing the positive selection system of claim 30 in the presence of test agents under non-permissive conditions; and (b) identifying surviving host cells from step (a), wherein said agent activates protein splicing.

38. The method of claim 37, wherein said agent is expressed within the host cell as a protein or portion thereof and wherein said surviving host cells of step (b) contain test agents which activate protein splicing.

39. The method of claims 36 or 38, wherein said host cell of step (a) expresses an in vivo peptide library or derivatives thereof, and wherein said test agents of step (a) comprise said in vivo peptide library or derivatives thereof.

40. The method of claim 39, wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

41. A reporter system for the screening of agents which inhibit protein splicing, said reporter system comprising a host cell which contains a reporter gene encoding a non-essential protein, said reporter gene containing an intein, wherein said intein is selected from the group consisting of a foreign intein inserted into the homologous or non-homologous extein of said reporter gene in the absence of a silent mutation of said extein and a natural intein in a native reporter extein, and wherein the inhibition of splicing of said non-essential protein by a given agent results in a specific detectable phenotype of said host cell.

42. A reporter system for the screening of agents which activate protein splicing, said reporter system comprising a host cell which contains a reporter gene encoding a non-essential protein, said reporter gene containing an inactive intein, wherein said inactive intein is selected from the group consisting of a foreign inactive intein inserted into the homologous or non-homologous extein of said reporter gene in the absence of a silent mutation of said extein and a natural inactive intein in a native reporter extein, and wherein the activation of splicing of said non-essential protein by a given agent results in a specific selectable phenotype of said host cell.

43. The method of screening for agents which inhibit protein splicing, said method comprising the steps of: (a) culturing the reporter system of claim 41 in the presence of test agents; and (b) identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype are in the presence of test agents which inhibit protein splicing.

44. The method of claim 43 wherein said agent is expressed within the host cell as a protein or portion thereof and wherein step (b) comprises identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype contain test agents which inhibit protein splicing.

45. A method of screening for agents which activate protein splicing, said method comprising the steps of: (a) culturing the reporter system of claim 42 in the presence of test agents; and (b) identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype are in the presence of test agents which activate protein splicing.

46. The method of claim 45 wherein said agent is expressed within the host cell as a protein or portion thereof and wherein step (b) comprises identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype contain test agents which activate protein splicing.

47. The method of claims 44 or 46, wherein said host cell of step (a) expresses an in vivo peptide library or derivatives thereof, and wherein said test agents of step (a) comprise said in vivo peptide library or derivatives thereof.

48. The method of claim 47, wherein said in vivo peptide library comprises a combinatorial peptide library in a fragment of chicken α-spectrin.

49. A method of controlling gene expression in vivo, said method comprising the steps of: (a) replacing a homologous extein gene in a host cell with a gene containing an intein; and (b) modulating the splicing of the intein-containing gene of step (a) with agents which inhibit or activate said splicing.

50. The method of claim 49, wherein said replacement of step (a) comprises inserting said intein gene into said homologous extein gene by homologous recombination.

51. The method of claim 49, wherein said replacement of step (a) comprises inactivating said homologous extein gene and said intein-containing gene is introduced as a second gene in the cell.

52. A method of controlling the delivery of a drug that is at least partially proteinaceous in vivo, said method comprising the steps of: (a) inserting an intein into the protein portion of said drug to create an inactive drug; (b) administering said inactive drug of step (a); and (c) activating protein splicing of said inactive drug to produce an active drug.

53. The method of claim 52, wherein step (b) further comprises utilizing gene therapy to target said inactive drug to a desired tissue.

54. The method of claim 52, wherein said intein of step (a) comprises a temperature-sensitive intein, and wherein said activation of step (c) comprises exposing a desired target tissue to any treatment which increases or decreases temperature of said target tissue thus inducing splicing of said temperature-sensitive intein.

55. The method of claim 52, wherein said activation of step (c) comprises injecting a desired target tissue with an agent which activates splicing.

56. The method of claim 52, wherein step (a) further comprises utilizing gene therapy to target an agent which activates splicing to a desired tissue, and wherein said administration of step (b) comprises systemic administration.

57. The method of claim 52, wherein said administration of step (b) comprises systemic transformation with a non-cell specific vector containing said inactive drug of step (a) operably linked to a desired tissue-specific promoter, and wherein said inactive drug is expressed only in cells which can activate said tissue-specific promoter.

58. A method for generating temperature sensitive mutants of the Mxe GyrA intein in E.coli GyrA, said method comprising the steps of: (a) identification of the region containing the Mxe GyrA intein beta-strand B8 and the loop between Beta-strands B8 and B9 in E.coli GyrA; (b) mutating said region; and (c) introducing the mutated intein gene into a cell and examining the ability to splice under permissive and non-permissive temperatures.

59. A method for generating temperature sensitive mutants of an intein, said method comprising the steps of: (a) identification of a region homologous to the Mxe GyrA intein beta-strand B8 and the loop between beta-strands B8 and B9 in an intein; (b) mutating said homologous region of said second intein; and (c) introducing the mutated intein gene into a cell and examining the ability to splice under permissive and non-permissive temperatures.

60. A method of screening for temperature-sensitive inteins, said method comprising the steps of (a) creating a positive selection system comprising a host cell containing a gene encoding a dominantly cytotoxic protein, said gene containing an intein, and wherein said intein is mutagenized; and (b) culturing the host cell of step (a) at a range of temperatures, wherein at a predetermined temperature, the protein fails to splice and results in viability of said host cell.

61. A method of screening for temperature-sensitive inteins, said method comprising the steps of (a) creating a positive selection system comprising a host cell containing a gene encoding a dominantly cytotoxic protein, said gene containing an inactive intein, and wherein said intein is mutagenized; and (b) culturing the host cell of step (a) at a range of temperatures, wherein at a predetermined temperature, the intein splices and results in the death of the host cell.

62. A positive selection system for the screening of temperatures which inhibit protein splicing in temperature-sensitive inteins, said selection system comprising (1) a host cell which contains a first gene encoding a controllable form of a target enzyme which is required for cell viability, and (2) a second gene encoding an expressed form of said target enzyme, said second gene containing an intein, wherein said intein is mutagenized, and wherein the inhibition of splicing of said target enzyme by a given non-permissive temperature results in the reduced viability or death of said host cell under conditions which do not permit the expression of said controllable first gene of said target enzyme.

63. A positive selection system for the screening of temperatures which activate protein splicing in temperature-sensitive inteins, said selection system comprising (1) a host cell which contains a first gene encoding a controllable form of a target enzyme which is required for cell viability, and (2) a second gene encoding an expressed form of said target enzyme, said second gene containing an intein, wherein said intein is mutagenized, and wherein the activation of splicing of said target enzyme by a given permissive temperature results in the viability of said host cell under conditions which do not permit the expression of said controllable first gene of said target enzyme.

64. A method of screening for temperatures which inhibit protein splicing in temperature-sensitive inteins, said method comprising the steps of: (a) culturing the positive selection system of claim 62 in the presence of non-permissive temperatures; and (b) identifying non-surviving host cells from step (a), wherein said temperature inhibits protein splicing.

65. A method of screening for temperatures which activate protein splicing in temperature-sensitive inteins, said method comprising the steps of: (a) culturing the positive selection system of claim 63 in the presence of permissive temperatures; and (b) identifying surviving host cells from step (a), wherein said temperature activates protein splicing.

66. A reporter system for the screening of temperatures which inhibit protein splicing in temperature-sensitive inteins, said reporter system comprising a host cell which contains a reporter gene encoding a non-essential protein, said reporter gene containing an intein, wherein said intein is selected from the group consisting of a foreign intein inserted into the homologous or non-homologous extein of said reporter gene in the absence of a silent mutation of said extein and a natural intein in a native reporter extein, and wherein said intein is mutagenized, and wherein the inhibition of splicing of said non-essential protein by a non-permissive temperature results in a specific detectable phenotype.

67. A reporter system for the screening of temperatures which activate protein splicing in temperature-sensitive inteins, said reporter system comprising a host cell which contains a reporter gene encoding a non-essential protein, said reporter gene containing an intein, wherein said intein is selected from the group consisting of a foreign intein inserted into the homologous or non-homologous extein of said reporter gene in the absence of a silent mutation of said extein and a natural intein in a native reporter extein, and wherein said intein is mutagenized, and wherein the activation of splicing of said non-essential protein by a permissive temperature results in a specific detectable phenotype.

68. A method for screening for temperatures which inhibit protein splicing in temperature-sensitive inteins, said method comprising the steps of: (a) culturing the reporter system of claim 66 in the presence of a range of temperatures; and (b) identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype are in the presence non-permissive temperatures which inhibit protein splicing.

69. A method for screening for temperatures which activate protein splicing in temperature-sensitive inteins, said method comprising the steps of: (a) culturing the reporter system of claim 67 in the presence of a range of temperatures; and (b) identifying host cells from step (a) having a specific detectable phenotype, wherein said host cells with detectable phenotype are in the presence permissive temperatures which activate protein splicing.

70. The positive genetic selection system of claim 6, 11, 29 or 32, wherein the extein is a heterologous extein and the intein is inserted into the extein at a site which is substantially identical to the homologous extein from about one to about five amino acid residues at either or both ends of the intein.

71. The reporter system of claim 41, wherein the extein is a heterologous extein and the intein is inserted into the extein at a site which is substantially identical to the homologous extein from about one to five amino acid residues at either or both ends of the intein.

72. The positive selection system of claim 6, 11, 29 or 32, wherein the extein is a heterologous extein and one to five amino acid residues of the native extein are present at one or both ends of the intein and said one to five amino acid residues are inserted into the heterologous extein along with the intein.

73. The reporter system of claim 41, wherein the extein is a heterologous extein and one to five amino acid residues of the native extein are present at one or both ends of the intein and said one to five amino acid residues are inserted into the heterologous extein along with the intein.

74. The positive genetic selection system of claims 6, 11, 29 or 32, wherein the insertion site is selected from the group consisting essentially of a surface location on the extein, a loop region of the extein, a protease sensitive site within the extein, or at a position known to permit insertion of one or more amino acid residues in the extein without inactivating the extein.

75. The positive genetic selection system of claim 70, wherein the insertion site is selected from the group consisting essentially of a surface location on the extein, a loop region of the extein, a protease sensitive site within the extein, or at a position known to permit insertion of one or more amino acid residues in the extein without inactivating the extein.

76. The positive genetic selection system of claim 72, wherein the insertion site is selected from the group consisting essentially of a surface location on the extein, a loop region of the extein, a protease sensitive site within the extein, or at a position known to permit insertion of one or more amino acid residues in the extein without inactivating the extein.

77. The reporter system of claim 41 or 71, wherein the insertion site is selected from the group consisting essentially of a surface location on the extein, a loop region of the extein, a protease sensitive site within the extein, or at a position known to permit insertion of one or more amino acid residues in the extein without inactivating the extein.

78. A positive genetic selection system for screening of agents which inhibit protein splicing, said selection system comprising: i) a host which contains a first gene encoding an inactivated form of a target enzyme which is required for cell viability under predetermined conditions; and ii) a second gene encoding an expressed form of said target enzyme, said second gene containing an intein, wherein the inhibition of splicing of said target enzyme by a given agent results in the reduced viability of said host cell under said predetermined conditions wherein expression of said target enzyme is required for viability or growth.

79. A positive selection system for the screening of temperatures which inhibit protein splicing in temperature sensitive inteins, said selection system comprising: i) a host cell which contains a first gene encoding an inactivated form of the target enzyme which is required for cell viability under predetermined conditions; and ii) a second gene encoding an expressed form of said target enzyme, said second gene containing an intein, wherein said intein is mutagenized and wherein the inhibition of splicing of said target enzyme at one or more predetermined temperatures results in the reduced viability of said host cell under said predetermined conditions wherein expression of said target enzyme is required for viability or growth.

80. A method of identifying an agent for antimicrobial activity against a microbial pathogen that naturally has an intein in an essential gene comprising screening for agents that block splicing of that intein in its native context or a homologous extein context.

81. A method of identifying an agent for antimicrobial activity against a microbial pathogen that naturally has an intein in an essential gene comprising screening for agents that block splicing of that intein in a heterologous extein context that includes one or more native extein residues flanking one or both ends of the intein.

82. A method of identifying agents with antimicrobial activity against Mycobacterium tuberculosis comprising screening for agents that inhibit splicing of the Mycobacterium tuberculosis DnaB intein.

83. A method of identifying agents with antimicrobial activity against Mycobacterium leprae comprising screening for agents that inhibit splicing of the Mycobacterium xenopi or Mycobacterium leprae GyrA inteins.

84. A method of identifying lead compounds with antimicrobial activity comprising identifying agents that inhibit splicing of an intein which is naturally present in that organism.