AU1009499A - Eukaryotic cell-based system for identifying gene modulators - Google Patents

Description

-1-

AUSTRALIA

PATENTS ACT 1990 COMPLETE

SPECIFICATION

FOR A STANDARD

PATENT

ORIGINAL

0

R

Name of Applicant/s: Johnson Johnson Research Pty. Limited Actual Inventor/s: Gregory Martin Arndt; David George Atkins- Address for Service: BALDWIN SHELSTON WATERS MARGARET

STREET

SYDNEY NSW 2000 nvention Title: "EUKARYOTIC CELL-BASED SYSTEM FOR IDENTIFYING GENE

MODULATORS"

The following statement is a full description of this invention, including the best method of performing it known to mFile: 21703.00 File: 21703.00 FIELD OF THE INVENTION The present invention relates generally to eukaryotic cell-based systems for screening molecules which are capable of modulating gene expression. More particularly, the present invention relates to high throughput phenotypic screening systems for detection of candidate suppression molecules, which systems rely on the use of target-reporter gene sequence fusion constructs.

BACKGROUND OF THE INVENTION :i Advances in procedures for genetic mapping and large scale DNA sequencing have already led to the determination of entire genome sequences of several •10 microorganisms and it is likely that the human genome will be sequenced within the next decade. A common observation following the analysis of sequenced genomes is that a -00..

large proportion of putative genes have no assigned cellular function. The magnitude of the challenge of ascribing biological function to these sequences has necessitated the development of novel practical methods for determining gene function. This field of 15 inquiry, dubbed 'functional genomics', includes the study of protein-protein interactions, °the construction of gene expression profiles and the use of single-cell and simple animal models to attempt to assign positive functions to these so-called 'orphan genes'.

A valuable complementary approach to gene function identification involves the study of cells in which the expression of a putative gene has been eliminated or suppressed. The elimination of gene expression is most commonly achieved through the use ofmutagenesis or gene disruption. These techniques can be laborious and are usually restricted to organisms with well-defined genetics. In contrast, the use of trans-acting strategies such as antisense RNA, ribozymes, and dominant negative proteins have been shown to be widely applicable in suppressing specific gene expression. However, the full potential of these technologies has yet to be realised due to the inability to select the most effective agents of this type for in vivo studies. To overcome this limitation, cellbased systems have been developed in which random gene fragments are expressed in cells followed by screening for a target gene-specific phenotypic modification (Holzmayer TA et al, Nucleic Acids Res, 1992, 20:711-717). These methods have been used to identify antisense RNAs and trans-dominant peptides for specific mammalian -3genes (Gudkov AV et al, Proc Natl Acad Sci USA, 1993, 90:3231-3235), to clone tumour suppressor genes (Gudkov AV et al, Proc Natl Acad Sci USA, 1994, 91:3744- 3748), and to effect genetic immunisation against specific viruses in mice (Barry MA et al, Nature, 1995, 377:632-635). The primary advantage of using such libraries is the isolation of the most effective gene suppression molecule in vivo amongst a large pool of different gene constructs. These methods using mammalian cells are time-consuming, costly and limited.

The ideal screening system should be based on an organism that is easy to handle, readily transformable, inexpensive and has the potential for application to high throughput screening. The candidate that satisfies these requirements are the yeasts and several groups have made progress in defining conditions in which yeasts can be used as hosts for studying gene-delivered modes of gene suppression (Arndt G.M et al, Mol. Gen. Genet.

1995, 248:293-300; Ferbeyre G et al, Gene, 1995, 155:45-50). A requirement of any yeastbased screening system is the use of a target gene that when down-regulated confers a 15 readily scorable cellular phenotype. Ideally, a quantitative relationship would exist between the degree of target gene suppression and the expression of the external phenotype. It has been shown previously that the lacZ gene, used both as a target gene sequence and a reporter sequence, could be regulated using lacZ gene antisense RNA in the.fission yeast Schizosaccharomyces pombe (Arndt G.M et al, Mol. Gen. Genet. 1995, 248:293-300). Using this system, it is possible to suppress P-galactosidase activity by to 45% using different lacZ antisense RNAs. However, due to the low steady-state level of lacZ target mRNA, it was not possible to adequately test the relationship between antisense RNA effectiveness and the external blue colony colour phenotype encoded by the JacZ gene.

Another yeast cell-based system for identifying gene suppression constructs (GSCs) encoding trans-acting molecules capable of suppressing the expression of lacZ gene, again used both as a target and a reporter sequence, has been described in the International Patent Application PCT/AU95/00235, published as WO 95/29254. In one version of that system, antisense sequences derived from the lacZ gene were used to transform a yeast strain expressing target lacZ gene at a high level and effective GSCs identified by screening transformants for changes in the lacZ gene-encoded blue colour -4colony phenotype. This system however is not capable of usefully identifying differences in the intensity of blue colony colour nor is it capable of being applied equally to transformants expressing low and high levels of 1-galactosidase, to the extent that it can be effectively used for high through-put screening of expression libraries.

Thus, there is a need for an improved phenotypic screening system for identifying candidate molecules capable of modulating expression of any target gene, which is high through-put and which is equally applicable to low and high levels of expression of the reporter gene sequences.

SUMMARY OF THE INVENTION 10 It has now been found that the expression of a desired phenotype by yeast cell o* o transformants harbouring a target/reporter gene fusion construct, is dependent on the nature of the target sequence fused to the reporter sequence. Low level of expression of the reporter sequence is not as amenable to phenotypic detection, and in particular to phenotype-based high throughput screening procedures, as transformants expressing high levels of the reporter sequence. By pre-determining for example the time of 0 •incubation, the concentration of the reporter gene product substrate and/or other conditions and components of the medium for expression of the reporter sequence for each individual fusion construct, a universally applicable, high throughput screening system can be achieved which has significant advantages over the existing systems.

Other embodiments and advantages of the present invention will be clear from the-following description.

BRIEF DESCRIPTION OF FIGURES Figure 1 shows a strategy for construction of random fragment expression libraries and screening in fission yeast. This schematic uses the SV40 promoter-driven ura4-lacZ fusion gene as an example (top of the diagram). The fusion gene is used as a template for PCR to amplify the SV40 promoter-ura4 target sequence (step This DNA is randomly fragmented using DNaseI and the resulting fragments size-selected, end-filled and ligated to Bgll linkers (step These fragments are then cloned into the kanR plasmid pBGS to construct an intermediate library (step and then sub-cloned from this library into the yeast expression plasmid pGTI 18 to produce the yeast expression library (step This library DNA is then used to transform the SUL strain, containing the SV40 promoter-ura4-lacZ fusion gene integrated at the ura4 locus, and transformants selected for leucine prototrophy and screened for their blue colony colour phenotype following overlay with agarose-Xgal medium (step The light blue transformants are assayed for b-galactosidase activity and whole cells used as templates to amplify inserts by PCR for sequencing (step The resulting plasmid with insert is referred to as a gene suppression construct

(GSC).

Figure 2 shows inserts contained on GSCs derived from the ura4 gene-specific 10 expression library. The top of the schematic shows the region of the ura4-lacZ fusion gene containing the SV40 early promoter and the ura4 sequence. The inserts contained on GSCs derived from the DL6 library are aligned below the target gene. To the left of each insert is the transformant designation, while to the right is indicated the degree of a suppression of P-galactosidase activity in comparison to the vector control. The direction 15 of the arrows represents the orientation of the insert in the GSC. The indicates the transcription initiation site and represents the ATG codon.

SFigure 3 shows alignment of the c-myc-specific inserts contained on GSCs identified from the DL7A library screen in yeast. At the top of the schematic is indicated Sa. the region of the c-myc-lacZ fusion gene used as a template for construction of the DL7A library. The exon 2 and exon 3 boundaries are marked. The inserts contained on each GSC are aligned below the target gene. All abbreviations are as referred to in Figure 2.

Figure 4 shows single and chimeric inserts on GSCs derived from the DL7B cmyc-specific expression library. The target gene region at the top of the diagram is as described in Figure 3. The inserts from the DL7B library are indicated relative to the target gene and categorised into 4 sub-groups as indicated. For the single fragment inserts, the arrows indicate the orientation of this DNA in the GSC. The regions X and Y represent potential sub-regions of accessibility in the c-myc mRNA. The number and letter designation within square brackets above or below fragments in the chimeric inserts define the position and orientation, respectively, within the GSC insert (with A indicating antisense and S representing sense orientations). Hatched boxes differentiate this fragment from others derived from the same target site in the chimeric inserts.

Figure 5 shows the alignment of inserts derived from the ura4 gene-specific GSCs identified in yeast transformants using 5-FOA selection. The schematic at the top of the diagram represents the region of the ura4-lacZ fusion gene containing the early promoter and the ura4 sequences. The inserts contained on the GSCs from the DL6 random expression library identified in yeast using 5-FOA selection are aligned below the target gene sequences. These inserts are categorised as antisense or sense according to their orientation in the yeast expression plasmid pGT118. All abbreviations 10 are as referred to in Figure 2.

~Figure 6 shows the inserts contained on the GSCs specific for Hradl and their alignment with the target gene sequence and the level of suppression of the fusion gene target in yeast. The labels "ATG" and "TGA" rerfer to the translation initiation and stop codons, respectively.

Figures 7 and 8 summarise the arrangement of inserts derived from the Hrad27 GSCs and the level of suppression of the fusion gene target in yeast. In both of these figures, the target gene sequence fused to the amino terminal end of the lacZ reporter is indicated at the top of the diagram. The labels "ATG" and "TGA" refer to the translation initiation and stop codons, respectively. "EVS" indicates the penultimate codon relative to the stop codon DETAILED DESCRIPTION OF THE INVENTION The screening method of the present invention is particularly but not exclusively useful for screening gene-encoded suppression molecules such as for example antisense, sense or ribozyme constructs, or transdominant polypeptides and small peptides.

However, it will be appreciated by those skilled in the art that other classes of suppression molecules may also be advantageously screened by the methods of the present invention. Further, the methods of the present invention may be used to screen for molecules which modulate target gene expression in ways other than suppression.

The rationale behind the most preferred format of the method of the present invention is the production in the yeast cell of a single mRNA transcript encoding both the target nucleic acid product or a portion thereof and a reporter molecule or a sequence encoding a reporter molecule. Generally, the reporter molecule will be encoded by a nucleotide sequence placed downstream of the nucleotide sequence encoding the target gene product. A molecule such as a genetic sequence capable of modulating the expression of the target gene sequence is then tested by reference to a change in the detectable reporter molecule.

To enable high-throughput screening with this method, the level of expression of 10 the reporter nucleic acid sequence, and in particular the phenotypic expression of the •reporter molecule, is titrated for each fusion gene-expressing yeast strain following quantitation of the reporter nucleic acid product activity by for example conventional assays such as solution enzyme assay.

According to a first aspect of the present invention there is provided a method for 15 determining whether, or to what degree, a molecule specifically modulates the expression of a target nucleic acid sequence in a eukaryotic cell, which comprises the S. steps of determining for the target sequence the conditions required for the expression of the reporter sequence or for the activity of its expression product in the hSchizosaccharomyces pombe cell of claim 2, contacting the Schizosaccharomyces 20 pombe cell with the molecule under said conditions, and detecting or quantitating a phenotypic change in the Schizosaccharomycespombe cell resulting from the modulation of the expression of the reporter sequence, thereby determining whether, or to what degree, the molecule specifically modulates the expression of the target sequence in a eukaryotic cell.

In a preferred embodiment the phenotypic changes are for example changes in colour, size, shape or texture of the yeast cells of colonies of yeast cells. In a particularly preferred embodiment the phenotypic change is blue colour or size of transformant yeast colonies resulting from the expression of P-galactosidase or orotidine monophosphate (OMP) decarboxylases reporter molecules, respectively.

-8- The molecules capable of modulating target nucleic acid expression are preferably genetic constructs such as antisense, sense nucleic acid sequences or ribozymes relative to a target gene, or other nucleic acid sequences such as oligonucleotides, random nucleic acid sequences or nucleic acid sequences selected for cleavage in vitro. The principles of the present invention, however, extend to nonnucleic acid molecules such as transdominant polypeptides, small peptides and chemical compounds as well as to synthetic nucleic acid molecules or nucleic acid analogue molecules.

The terms "modulating expression" and the like, include in their scope up- 10 regulating and down-regulating expression of a target sequence as well as activity of a product of a target sequence. Thus, molecules which modulate expression or activity may be agonists or antagonists. Furthermore, the present invention extends to identifying agents, such as for example, ribozymes, antisense and sense nucleotide molecules whose activity is regulated by cellular factors including switching mechanisms and intracellular address signals. Additionally, in some circumstances, the level of effect on expression may be enhanced by increasing the level of modulating molecules such as antisense, sense or ribozyme molecules.

The term "gene" is used in its broadest sense to include a genomic gene sequence as well as a genetic sequence comprising only the coding portions of a gene exons), 20 a cDNA sequence corresponding to a mRNA transcript and a synthetic gene sequence.

A "gene" as contemplated herein especially in relation to a target gene includes a naturally occurring gene, a partial gene, a synthetic gene and a fusion between a target gene and another gene or genetic sequence. A "gene", therefore, is considered herein to include any target nucleotide sequence and may be of eukaryotic, prokaryotic or viral origin.

The term "protein" as used in the context of the present invention is intended to encompass polypeptides and peptides.

Preferably, the target nucleic acid sequence is "exogenous" to S. pombe meaning it is a heterologous gene which has been introduced by transformation, conjugation, electroporation or other means to the yeast cell. However, the target gene may alternatively be "endogenous". Particularly preferred endogenous or homologous S.

pombe genes are those which encode cell cycle proteins, modulate cell cycles and/or are involved in programmed cell death. Such genes and in particular antagonists thereof identified in accordance with the present invention may be useful in the treatment of cancers in mammals such as humans. Other important endogenous target genes are S.

pombe homologues of mammalian human) genes. A particularly useful yeast in accordance with this aspect of the present invention has S. pombe genes replaced by mammalian human) homologues or comprise homologous animal, mammalian or plafit target genes or have homologous functions to animal, mammalian or plant target 10 genes.

The target nucleic acid sequences include eukaryotic, prokaryotic and viral genes. Examples of eukaryotic genes include mammalian growth factors and cytokines and their receptors and cancer specified genes and plant genes. Examples ofprokaryotic genes include antimicrobial resistance genes and pathogen-specific genes. Examples of 15 viral genes include retroviral genes such as those of HIV and Hepatitis.

According to a second aspect there is provided a nucleic acid fusion construct capable of being expressed when introduced into Schizosaccharomyces pombe, consisting essentially of a target nucleic acid sequence and a reporter nucleic acid sequence, wherein the target and reporter sequences are situated with respect to each °20 other so that the expression of the reporter sequence is modulated by any molecule which also modulates the expression of the target sequence, and the expression of the reporter sequence confers on Schizosaccharomyces pombe a detectable phenotype.

The reporter nucleic acid sequence product may be any molecule capable of giving an identifiable signal and may be an enzyme, preferably P-galactosidase or orotidine monophosphate (OMP) decarboxylase. Other reporter sequences which tolerate 5' fusions can also be used, for example those which encode luciferase or green fluorescent protein. Examples of other reporter molecules which may be used are chloramphenicol acetyl transferase (CAT), other enzymes confering antibiotic resistance, P-glucuronidase, fluorescent proteins, essential growth factors and cell cycle proteins.

The target-reporter nucleic acid fusion construct may be integrated into the chromosome of S. pombe under the control of an endogenous promoter or exogenous promoter or may exist as an extrachromosomal, replicating element. Any number of exogenous promoters may be used such as an SV40 promoter. A suitable endogenous constitutive promoter is S. pombe adh 1 promoter.

The genetic sequences to be tested an antisense or sense molecule or ribozyme construct) may be introduced to the yeast cell by any number of means including transformation, conjugation, electroporation, amongst others. The genetic sequence may be expressed under an endogenous or exogenous promoter or may be 10 introduced without a promoter.

.The present invention permits the rapid screening of genetic sequences having an effect on expression of a wide range of target genes. The present invention is particularly applicable for the development of a disease model system suitable for S° screening for useful genetic sequences to target viral, cancer and aberrant "self' genes.

15 The method of the present invention may also be useful as a drug screening system. For example, receptor genes are engineered and expressed in a fusion construct with a reporter or a reporter gene sequence. The modified yeast is then used in high throughput assays for agonists or antagonists of the receptors.

In certain preferred embodiments the present invention is concerned with the 20 improvement of the Schizosaccharomycespombe gene suppressor screening system, particularly the phenotypic detection system, to permit the identification of GSCs for any target gene sequence including genes of mammalian origin. Particularly preferred embodiments of the present invention exploit the capacity of a reporter gene sequence to be expressed despite additional sequences at its 5' end and are based on the observation that the nature of the target sequence in the fusion construct governs the level of expression of the reporter gene product, eg -galactosidase or orotidine monophosphate (OMP) decarboxylase, and thus affect the expression of the desired transformant yeast colony phenotype, eg, blue colour or colony size. In particular, the improvement comprising the present invention relies on pre-determining the conditions necessary for 11 expression of reporter gene sequences, eg. P-Galactosidase or orotidine monophosphate (OMP) decarboxylase for each target/reporter gene fusion construct.

In one example of this universal approach based on target-reporter gene fusion (Figure the target sequence is fused to the 5' end of the lacZ reporter gene. The end result is the expression in Schizosaccharomyces pombe of a fusion mRNA containing both target gene and lacZ sequences. Translation of this chimeric mRNA maintains Pgalactosidase enzyme activity and confers upon the cell the blue colour colony phenotype in the presence of X-gal. However, as discussed above, the development of colour and/or its intensity is dependent on the nature of the target sequence attached to 10 the N-terminus of P-galactosidase. Conditions for colour development therefore need to be pre-determined for each such construct before the yeast transformant can be used effectively in a high throughput phenotypic screening system for identifying molecules capable of modulating expression of a target gene sequence. Those conditions include but are not limited to X-gal concentration, detergent content and time of incubation, so 15 that the screening system can be used to detect blue colour changes in transformants expressing low as well as high levels of P-galactosidase.

A random fragment expression library derived from the target gene sequence is then expressed in the fusion gene-containing yeast strain and screened using the lacZ gene-encoded phenotype. Transformants containing effective GSCs may be identified on 20 the-basis of a light blue or white phenotype, however, less advantageously measurement of P-galactosidase enzyme activity may also be used for detection.

A similar approach was taken with S. pombe ura4 gene sequence which encodes orotidine monophosphate (OMP) decarboxylase. This enzyme is involved in the uracil biosynthetic pathway. Yeast strains containing null mutations in this gene are auxotrophic for uracil and able to grow in the presence of the pyrimidine analog fluoro-orotic acid (5-FOA). In contrast, wild type strains are prototrophic for uracil and sensitive to 5-FOA. The association of a negative selection phenotype (resistance to FOA) with the loss of ura4 gene function provided the basis for testing this genetic marker as an alternative reporter gene in gene fusions for use in screening random -12fragment expression libraries to isolate effective GSCs. This approach is discussed in more detail below.

The improved phenotypic cell-based screening methods of the present invention provide for a significant increase in the number of yeast transformants and therefore the number of expression constructs, that can be screened using the desired phenotype, such as change in the colony colour or size, and are high throughput with the capability to screen complex libraries, such as those having more than 105 constructs. It is also amenable to automation and has the potential to develop therapeutic GSCs as well as providing tools for rapid insight into functional genomic studies.

10 Preferred embodiments of the invention will now be described, by way of example only, with reference to the examples and figures.

Example 1 Yeast strains, constructs and expression libraries Yeast media and methods.

All yeast strains and transformants were maintained on standard YES or EMM media (Moreno, S. et al, Meth. Enzymol, 1991. 194:795-823). Yeast cells were transformed with plasmid DNA by electroporation (Prentice, H.L. Nucleic Acids Res., 1992, 20:621). Plasmid DNA was isolated using a glass bead method (Hoffman, C.S.

and Winston, Gene, 1987, 57:267-272) and recovered into DH5a cells using standard electroporation procedures (Sambrook et al, Molecular cloning: a laboratory manual.

1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York) prior to retransformation into yeast cells. b-galactosidase enzyme activity was determined using a cell permeabilization protocol as previously described (Arndt, G.M et al., Mol. Gen.

Genet., 1995, 248:293-300). PCR amplification of plasmid DNA within whole yeast cells involved treating cells with 10 mg/ml zymolyase-20T (ICN Biomedicals, Aurora, Ohio) in 50 mM Tris-HC1 (pH 8.0) for 30 minutes at 30 oC. To these cells was added directly the PCR mixture containing 5 pl 10X reaction buffer [500 mM KCI, 100 mM Tris-HCl (pH 1 pl dNTPs (10 mM each), 1 1l of each primer (0.1 gg/ml), 36pl

H

2 0, and 1.2 units AmpliTaq DNA polymerase (Perkin Elmer, 5 units/pl).

Amplification was performed using a PCR program of 1 cycle at 94 OC for 3 minutes 13and 30 cycles of 94 OC 30 seconds, 55 OC 90 seconds and 72 OC 90 seconds in a Perkin Elmer thermocycler.

(ii) Yeast strains and plasmids.

All routine DNA manipulations were completed using standard protocols (Sambrook et al, Molecular cloning: a laboratory manual. 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York). The Schizosaccharomyces pombe strain expressing higher levels of P-galactosidase activity, designated HS, was constructed by replacing the ura4 wild type gene in strain 972 with the lacZ gene under control of the fission yeast adhl promoter and ura4 3' processing signals. This was SS 10 completed by the utilisation of ura4 flanking sequences contained on plasmid pNEB 195 (Arndt, G.M et al., Mol. Gen. Genet., 1995, 248:293-300) and homologous recombination (Grimm, C.,et al., Mol Gen Genet., 1988, 215:81-86). Plasmid pGT5 was derived from pNEB 195 by deletion of the BamHI site located in the polylinker 5' to the ura4 5' DNA flank. The adhl promoter and ura4 3' processing signal sequences were C 15 derived as a PCR fragment by amplification of plasmid pURAS (Patrikakis, et al., Curr. Genet. 1996, 30:151-158) using the adhl 5' primer TGAAAGCTTGTCGACCTAAGAAAATGGCTATCATGCGG-3' and the ura4 3' primer 5'-GGGAAGCTTGTGATATTGACGAAACTTT-3'. This PCR product was cloned as a HindIII fragment into pGT5, between the ura4 flanking sequences, to 20 produce the plasmid pGT105. The lacZBamHI fragment derived from plasmid pNEB32 (Arndt, G.M et al., Mol. Gen. Genet., 1995, 248:293-300) was then subcloned into the unique BamHI site in the sense orientation between the adhl promoter and ura4 3' processing signal in pGT105 to produce pGT113. A PacI-Pmel fragment from the resulting plasmid, containing the adhl promoter-lacZ-ura4 3' expression cassette flanked at either end by ura4 sequences, was used to transform strain 972. Putative integrants were selected as 5-FOA resistant (5-FOAR) transformants and then characterised for integration of the above expression cassette. The resulting strain, designated 599-2 ura4::adhl-lacZ), was crossed to NCYC 1914 to introduce the leul-32 mutant allele.

The strain HS ura4::adhl-lacZ, leul-32) was identified as an ascospore isolate of this cross and characterised for expression of the lacZ gene.

-14- The construction of the long (pGT2), 5' (pGT59), and 3' (pGT61) lacZantisense gene-containing plasmids and their corresponding sense control plasmids have been described previously (Arndt, G.M et al., Mol. Gen. Genet., 1995, 248:293-300). All of these antisense RNAs and control sense RNAs were expressed under the control of the S.

pombe nmtl promoter in the plasmid pREPI which contains the S. cerevisiae LEU2 selectable marker (Maundrell, J Biol. Chem. 1990, 265:10857-10864).

To construct the SUL strain containing the SV40 promoter-ura4-lacZ-ura4 3' fusion cassette, the SV40 early promoter was PCR amplified from pSVB (Clontech) using the following 5' and 3' primers, respectively: 10 TGGGGATCCAAGCTTAATTCGAGCTCGGTACAGCTTGG and AAGGGATCCCTACTTCTGGAATAGCTCAGAGG The 350 bp fragment was Sdigested with BamHI, gel-purified and subcloned into a pGEM3Zf(-)-based plasmid, containing the adhl promoter and ura4 3' processing signal (Patrikakis, et al., Curr.

Genet. 1996, 30:151-158), in place of the adhl promoter. A HindII fragment containing 15 the.SV40 promoter and ura4 3' sequence was subcloned into plasmid pGT5 between the ura4 5' and 3' flanking sequences to produce plasmid p40F. The 3.1 kb lacZ fragment used in constructing the fusion gene was prepared by PCR amplification, from pGEM- BGal containing the 3.5 kb NotI fragment from pSVB, using the primers GACGGATCCCGGGTCGTTTTACAACGTCGTGAC and 20 TG(GGAATTCGCGAAATACGGGCAGACATGG and was cloned as an EcoRI- BamHI fragment into pSP72 digested with the same enzymes. This fragment was then subcloned as an EcoRI-BamHI fragment into EcoRI-BamHI-digested p40F to create the universal integration vector pINTVEC and the junctions confirmed by DNA sequence analysis. The ura4 test gene sequence was amplified by PCR, from the template pREP4 (Maundrell, Gene 1993, 123:127-130), using the primers GAGAAGCTTGGATCCGCAAAGTTATGGATGCTAGAG and TTGGGATCCAGAATGCTGAGAAAGTCTTTGCTG This fragment was digested with BamHI and cloned into plNTVEC in the sense orientation between the promoter and the lacZ sequences to produce pivURA4s. DNA sequencing confirmed that the.ura4 gene was in-frame with the lacZgene. Plasmid pivURA4s was digested with Pacl-Pmel to release the SV40 promoter-ura4-lacZ-ura4 3' fusion cassette flanked by ura4 chromosomal sequences. Strain HS was transformed with this DNA and integrants selected on EMM medium. Southern analysis confirmed integration of the fusion cassette at the ura4 locus.

To construct the AML 1 strain expressing the c-myc-lacZ fusion under control of the adhl promoter, pINTVEC was partially digested with BamHI and HindIII to remove the SV40 early promoter, and then ligated to the 745 bp adhl promoter fragment •(Patrikakis, et al., Curr. Genet. 1996, 30:151-158) to produce pSIV. The region of the c-myc cDNA used in constructing the c-myc-lacZ fusion gene was amplified from the Quick-Screen human cDNA library (Clontech, Palo Alto, CA) using the following 5' and 3' primers, respectively: CGAGGATCCTTGCAGCTGCTTAGA and TAGGGATCCCGCACAAGAGTTCCGTAG The 1379 bp fragment was digested with BamHI, gel-purified and cloned into the pGEM3Zf(-) plasmid. The complete sequence of this amplified product was determined. The c-myc sequence was subcloned as a BamHI fragment into pSIV in the sense orientation between the adhl promoter and 'o 15 the lacZ gene to produce pAML s. DNA sequencing confirmed that the c-myc gene was in frame with the lacZ gene. Plasmid pAML Is was digested with PacI-PmeI to release the c-myc-lacZ fusion gene flanked by ura4 chromosomal sequences. Strain NCYC 1913 was transformed with this DNA and integrants selected on EMM medium containing 1 mg/ml 5-fluoroorotic acid. The integration was confirmed by Southern analysis.

(iii) Construction of random fragment expression library.

The yeast shuttle plasmid pGT118 was used in constructing all random fragment expression libraries. This vector was created by modifying pREP1 (Maundrell,

J

Bioi. Chem. 1990, 265:10857-10864) to include a single ATG translation start codon in the nmtl promoter region, a unique BglII cloning site between the nmtl promoter and 3' processing signal, and TAA stop codons in three reading frames downstream of the BglII site.

To construct the ura4 gene-specific random fragment library, the promoter-ura4 portion of the fusion expression cassette was PCR amplified from pivURA4s using the primers

TGGAAGCTTAATTCGAGCTCGGTACAGCTTGG

and TTGGGATCCAGAATGCTGAGAAAGTCTTTGCTG This DNA was -16partially digested with DNaseI and fragments greater than 300 bp size-selected using CL-4B Sepharose. Fragments were end-filled with T4 DNA polymerase and Klenow fragment and ligated to BglII linkers (New England Biolabs). After BglII digestion, DNA fragments were ligated to BgllII-digested pBGS, which is a modified version of plasmid pBS8(+) (Spratt, et al., Gene, 1986, 41:337-342) containing a BglII site and the kanamycin resistance marker. A library of 11,000 clones was generated in S.by amplifying single colonies on LB plates containing 50 gg/ml kanamycin. The Bgll inserts from this intermediate library were subcloned into pGT118. Transformation of DH10B cells (BRL) produced the random fragment expression library designated DL6 l10 which contains 20,000 independent clones with 90% of these having inserts.

For the c-myc-specific expression libraries, a region including 20 bp upstream of the-adhl transcription start site and the entire c-myc region was PCR-amplified from pAMLs using the primers TGGCCTTCGCTTTTCTTTAAGCAAGAG and TAGGGATCCCGCACAAGAGTTCCGTAG Following partial digestion with 15 DNaseI, fragments were size-selected to isolate those greater than and less than 300 bp.

In separate reactions, these two pools of fragments were end-filled and ligated to BglII linkers. After BglII digestion, DNA fragments were ligated independently to Bglll- S. digested pBGS, to produce two intermediate libraries in DH10B cells. The BglII inserts contained in each of these libraries were subcloned into pGTl 18. Transformation of DH1OB cells produced two yeast expression libraries designated DL7A and DL7B containing, respectively, large and small inserts derived from the c-myc target sequence.

The DL7A library contained 230,000 clones of which 40% contained inserts. The DL7B library was composed of 1.8x107 clones with 96% of these containing inserts.

(iv) Screening random fragment expression library in yeast.

Following transformation of the fusion gene-expressing yeast strain with 2pg of plasmid DNA from the random fragment expression library, the transformed cell mixture is inoculated into selective liquid EMM medium and these cultures are grown at 30 0 C for approximately 72 hours. This allows recovery of the cells from electroporation, accumulation of the plasmid encoded RNA and turnover of the fusion protein.

Transformations are plated to selective solid EMM medium and grown for approximately 4 to 5 days at 30°C. The plating density is maintained at up to 1000 colonies per plate -17- (diameter of 14cm). These colonies are overlayed with an agarose-Xgal medium and then scored for differences in blue colony colour using the vector alone transformant as a control.

The agarose-Xgal medium is composed of 0.5M sodium phosphate, 0.5% agarose and 2% dimethylformamide. In addition, this medium also contains sodium dodecyl sulfate (SDS) and X-gal. The concentrations of these latter components are pre-determined for each fusion gene-expressing yeast strain following quantitation of the P-galactosidase activity displayed in the standard liquid solution enzyme assay. This is accomplished by S" titrating the SDS and X-gal concentrations in combination with one another and varying the incubation temperature and time. This strategy provides the flexibility to detect blue colony colour for yeast transformants expressing low or high levels of p-galactosidase activity. For example, detection of the blue colony colour for the ura4-lacZ fusion geneexpressing strain requires 0.01% SDS and 200jig/ml X-gal and incubation at 37 0 C for 3 hours. In contrast, strain AML 1 expressing the c-myc-lacZ fusion gene produces 15 detectable blue colour using 0.01% SDS and 1000pg/ml when incubated at 37 C overnight.

All plates were screened as above, in comparison to the parental strain transformed with pGT118, and light blue and white transformants identified, plugged and cells recovered onto selective medium. These transformants were characterised for the presence or absence of a plasmid insert by PCR amplification on whole cells using the pGT118-specific primers TGAAAGCTTTTATAGTCGCTTTGTTAAATCATATG and TGAAAGCTTTACCCGGGGATCCTTAGTTAGTTAAG Also included in each reaction were the actin-specific primers ATGGAAGAAGAAATCGCAGCGTTG and AATACCACGCTTGCTTTGAGCTTC that acted as internal controls.

The transformants containing inserts were then single colony purified and assayed in triplicate for p-galactosidase activity. Select transformants displaying greater than suppression of p-galactosidase activity were re-assayed in triplicate.

To analyse the inserts contained on the library plasmids identified using the yeast screen, each was PCR amplified from whole cells using the pGT118-specific nmtl primers as indicated above. The PCR fragments were purified using Wizard PCR 18columns (Promega) and sequenced in both directions using the following sequencing primers: AAGCTTTTATAGTCGC and AAGCTTTACCCGGGGA All fragments were sequenced using an Applied Biosystems 373 DNA sequencer.

In addition to titration of the SDS and X-gal concentrations contained in the agarose-Xgal medium, the level of the fusion protein expressed in vivo can be controlled using two different promoter systems: the fission yeast adhl promoter (as used with AML1) and the mammalian SV40 early promoter (as used with the ura4-lacZ fusion strain described below).

Example 2 GSCs specific for the fission yeast ura4 gene Colony colour phenotvpe detection As one example of the target reporter gene sequence fusion concept for identifying GSCs, we used the S. pombe ura4 sequence as a test gene. The ura4-lacZ fusion gene-expressing strain SUL was transformed with the DL6 expression library derived from the SV40 early promoter-ura4 sequences contained in the fusion target gene. A total of 47,000 yeast transformants were screened for changes in their blue colony colour and 291 light blue and white transformants were identified. Of these transformants, 170 were assayed for P-galactosidase activity in triplicate and all exhibited suppression of the ura4-lacZ target gene as measured by a reduction in P-galactosidase activity of between 29% and 77%. The specific level of suppression was shown to be reproducible by re-assaying 44 of these transformants for P-galactosidase activity. To further characterise the specificity of the GSCs, 20 light blue transformants were assayed for P-galactosidase activity following growth in the presence of thiamine and, in each case, suppression of P-galactosidase activity was dependent on transcription of the plasmid insert. Plasmids were then isolated from 17 different primary transformants, re-transformed into SUL and three independent transformants for each plasmid assayed for P-galactosidase activity. All plasmid re-transformants showed the same level of suppression of activity as the primary yeast transformant, indicating that the-observed suppression was contingent on the presence of the specific

GSC.

PCR fragments amplified from inserts contained on plasmids of 44 light blue transformants were sequenced and a subset of these are shown aligned with the target 19promoter-ura4 sequence in Figure 2. A total of 29 independent antisense inserts and 2 sense inserts were identified which suppressed ura4-lacZ gene expression.

Interestingly, transformant 1-E6 contained two GSCs each with a ura4-specific insert in the antisense orientation. The top 26 antisense insert-containing GSCs produced a level of suppression equal to or greater than that obtained using the full length antisense RNA complementary to the entire ura4-lacZ fusion transcript (data not shown). All of the inserts indicated in Figure 2 were more effective than the full length ura4 sequence expressed in the antisense orientation, which suppressed the ura4-lacZ fusion gene by 29%. This analysis indicates that using the lacZ gene-encoded cellular phenotype as a 10 genetic screen is an efficient way of identifying highly effective gene constructs capable of suppressing a target gene sequence fused to the 5' end of the lacZ gene, and that this approach yields more effective constructs than a simple full length antisense RNA.

Analysis of the origin of the GSCs shows that 22 out of 29 of these inserts encoded antisense RNAs complementary to the 5' cap site on the ura4-lacZ mRNA. In addition, 90% of the antisense inserts contained a 64 bp ura4-specific region including 13 bp upstream of the ATG codon, the ATG codon and 51 bp downstream of the ATG codon. This suggests that the expressed antisense RNAs from these 25 different inserts do not require complementarity with SV40 5' UTR sequences contained on the ura4lacZ fusion mRNA. The remaining 4 antisense inserts lacking the above 64 bp region were less effective at suppressing fusion gene expression. These observations further illustrate the potential of this genetic screening assay to identify sub-regions of target RNA accessible to regulation by antisense RNA in vivo.

(ii) Colonysize phenotype detection The S. pombe ura4 gene encodes the enzyme orotidine monophosphate

(OMP)

decarboxylase which is involved in the uracil biosynthetic pathway. Yeast strains containing null mutations in this gene are auxotrophic for uracil and able to grow in the presence of the pyrimidine analog 5-fluoro-orotic acid (5-FOA). In contrast, wild type strains are prototrophic for uracil and sensitive to 5-FOA. The association of a negative selection phenotype (resistance to 5-FOA) with the loss of ura4 gene function provided the basis for testing this genetic marker as an alternative reporter gene in gene fusions for use in screening random fragment expression libraries to isolate effective GSCs. As a proof-of-principle, we used the SUL strain containing the ura4-lacZ fusion gene.

Preliminary characteristics of this strain indicated that it was: able to grow in the absence of uracil; sensitive to 1 mg/ml 5-FOA in the presence of 20 mg/L uracil and capable of producing blue coloured colonies following overlay with the Xgal-agarose medium. In addition, Northern blot analysis confirmed expression of the ura4-lacZ mRNA.

In order to test the 5-FOA selection system for its capacity to detect yeast colonies expressing intermediate levels of ura4 gene activity, we expressed an antisense 10 RNA complementary to the full length ura4-lacZ mRNA in SUL. Enzyme assay analysis on three independent transformants revealed that this antisense RNA suppressed -galactosidase activity by 50%. Testing of these transformants (50% ura4 gene activity) and control transformants (100% ura4 gene activity) on standard EMM medium containing 1 mg/ml 5-FOA and 20 mg/L uracil indicated that neither cell type could form colonies. Both of the above transformants were used to identify media conditions (following titration of uracil and 5-FOA) in which colonies expressing 50% ura4 gene activity could be differentiated from those expressing wild type levels on the basis of their size. Plating these transformants on EMM media containing 0.5 mg/ml 5-FOA and mg/L uracil resulted in small background size colonies for the vector control and significantly larger colonies for the yeast transformant expressing 50% less ura4 gene activity. To further test these media conditions for detecting a rare colony with 50% less ura4 gene activity among a larger number of control transformants, we mixed 200 of the former cells with 103, 104, and 105 control cells and plated to the selective medium. The larger colonies were picked and shown by whole cell PCR to contain the antisense ura4lacZ plasmid. Thus, we established conditions in which the ura4 gene in fusion context could be suppressed in trans and determined that yeast transformants expressing intermediate levels of the ura4 gene-encoded enzyme activity could be effectively identified among a greater number of colonies expressing 100% OMP decarboxylase activity.

-21- To demonstrate that the selective conditions established above could be used to select for GSCs from a random expression library, the ura4 gene-specific library (DL6) was used to transform SUL and 4.5x 106 cells plated to EMM media containing mg/ml 5-FOA and 2.5 mg/L uracil. A total of 46 large 5-FOA-resistant colonies were assayed for p-galactosidase activity to quantitate the level of suppression of the fusion gene. This analysis indicated that 75% of these transformants displayed reproducible levels of suppression ranging from 21% to 72%. Plasmids were isolated :from 15 transformants and re-transformed into the ura4-lacZ gene-expressing strain.

Assays indicated that 13 of the 15 re-transformants displayed plasmid-dependent gene suppression. Inserts from GSCs of 32 different 5-FOA-resistant transformants were sequenced. This analysis revealed the presence of 10 different inserts, five of which .were in the antisense orientation and five in the sense direction (Figure 5. All of the antisense inserts identified encoded for antisense RNAs complementary to at least the transcription start site, the ATG codon and 340 bases downstream of the translation start 15 site of the ura4-lacZ fusion mRNA. This indicates that the 5-FOA selection protocol can be used to identify yeast transformants containing GSCs capable of suppressing the ura4-lacZ fusion gene. In this particular case, the ura4 gene was fused to the amino terminus of the lacZ reporter gene. In addition, the newly created ura4-lacZ gene has the potential to be used in the present system as a unique dual reporter for constructing fusion gene targets. GSCs could be identified to the amino terminally fused target sequence by selection for 5-FOA resistant transformants. The level of suppression could then be quantitated using P-galactosidase enzyme assays.

Example 3 GSCs for a human gene To demonstrate the universal nature of the genetic screen and its application to non-yeast target sequences, a screen for GSCs specific for the human c-myc gene was performed. A portion of the human c-myc cDNA was fused 5' to the lacZ gene and the resulting c-myc-lacZ fusion gene expressed from the ura4 locus in strain AML1. The DL7A expression library, containing c-myc-derived inserts greater than 300 bp in size, was used to transform AML1 and from 14,000 transformants screened, 3% exhibited a light blue phenotype. Of these transformants, 28 were assayed for p-galactosidase activity and all displayed suppression ranging from 20% to 72%. The inserts contained -22on the GSCs were PCR amplified, sequenced and aligned with the target gene in Figure 3. Two types of inserts were identified including 19 different antisense and 3 sense inserts. The antisense inserts ranged in size from 593 bp to 1081 bp and transformants containing these GSCs showed reductions in P-galactosidase activity ranging from 43% to 72%. A total of 85% of these inserts contained the ATG codon, suggesting that this region may be an important component of the long antisense RNAs. Four of the 19 antisense GSCs were more effective than a construct expressing c-myc antisense

RNA

complementary to the entire c-myc RNA component of the fusion target mRNA, which reduced p-galactosidase activity by 65%. The level of suppression mediated by the 10 GSCs containing the sense inserts was lower than the antisense insert-containing GSCs, S with the most effective sense construct reducing p-galactosidase activity by 47%. These results demonstrate that the fusion approach can be used to identify GSCs specific for a .°human gene target.

To examine the potential of the genetic screening technology to further define sub-regions of a target RNA sensitive to regulation by trans-acting RNAs, a second c-myc-specific expression library containing inserts ranging in size from 100 bp to 300 bp was constructed. Following transformation of the DL7B library into AML1 and screening of 16,000 transformants, a total of 315 light blue transformants were identified. p-galactosidase assays on 56 of these transformants showed that the GSCs suppressed c-myc-lacZ gene expression by 20% to 81%. Sequencing of the inserts contained on the resident GSCs revealed both single fragment-containing inserts and chifneric inserts (Figure A total of 12 different single antisense inserts were identified and all of these originated from either the 5' end of the c-myc sequence (including part of exon 1 and exon2) or a region midway through exon 2. The implication is that these 77 bp and 45 bp regions of the c-myc target mRNA are favoured domains to target for complementary-dependent suppression.

An unexpected observation from the DL7B library screen was the identification of chimeric inserts composed of two or more fragments in the antisense or sense orientations (Figure The GSCs containing the 15 unique chimeric antisense inserts suppressed P-galactosidase activity by 33% to 78%. All of the fragments contained in these inserts spanned the same regions of the c-myc target sequence as the single long -23antisense inserts, confirming the accessibility of these regions to antisense RNA regulation. The recovery of chimeric inserts demonstrates the power of the genetic screen to identify novel GSCs. It will be of interest to determine the contribution of each component of the chimeric RNAs to the overall level of gene suppression.

Artificially designed antisense RNAs are commonly used tools for suppressing the expression of specific genes or controlling viral replication. The efficacy of this approach is dependent on a number of parameters, one of which is the accessibility of the antisense RNA to the target mRNA in vivo. This accessibility is determined by both the intracellular localisation of the two interacting RNAs as well as the ability of these RNAs to hybridise. A number of in vitro strategies have been developed to identify regions of a target mRNA accessible to hybridisation with antisense sequences (Milner, et al., Nature Biotech., 1997, 15:537-541; Patzel, V. and Sczakiel, Nature Biotech. 1998, 16:64-68). A major limitation of these predictions is that the identified antisense RNAs or oligonucleotides do not always function in vivo. This is not surprising .15 given the fact that the complexity of RNA structures in vivo is determined by long-range secondary and tertiary interactions and RNA binding proteins. The fission yeast genetic screen offers the distinct advantage of identifying effective antisense RNAs within the cellular milieu. By restricting the size of the inserts, it is possible to identify sub-regions of a target mRNA which are accessible in vivo. In the present study, we demonstrated this feature of the genetic screen by identifying two potential accessible regions in the cmyc target RNA.

Using the random fragment expression libraries it was possible to identify novel trans-acting gene suppression molecules. For both the ura4 and c-myc target genes, we identified GSCs encoding sense RNAs. Sense RNA-mediated suppression has been demonstrated to be an effective way of controlling gene expression (Flavell, R.B. Proc.

Natl. Acad. Sci. 1994, 91:3490-3496) and this is the first report of similar activity in yeast. The genetic screen using expression libraries with smaller inserts produced GSCs with chimeric inserts specific for c-myc. Chimeric antisense RNAs have not been commonly used in empirical approaches to designing effective intracellular inhibitors.

The advantages of such antisense RNAs is the ability to hybridise with more than one region of the target mRNA and the potential for one of the antisense sequences to act as -24a facilitator of other antisense sequences contained on the same chimeric RNA (Nesbitt, S. and Goodchild, J. Antisense Res. Develop., 1994, 4:243-249). A biological approach such as the one we have described is the only way to select such effective and novel chimeric gene constructs.

The S. pombe genetic screen described herein provides a method for the identification of GSCs and other molecules capable of modulating target gene expression, for any target sequence in a eukaryotic cell environment. Candidate genes need only to be characterised at the primary nucleic acid sequence level and be amenable to fusion to the N-terminus of the lacZ gene such that P-galactosidase activity is 10 maintained. Clearly the sequence information is not limiting and the lacZ gene, for example, can be incorporated into thousands of functional fusions (Bums, et al, Genes Develop., 1994, 8:1087-1105; Dang, et al., Yeast, 1994, 10:1273-1283).

The screening method of the present invention is both rapid and simple and, as such, is particularly advantageous in identification of hundreds of different GSCs from a pool of hundreds of thousands of gene constructs. Every stage of the genetic screen is amenable to robotic automation and therefore large numbers of target gene sequences can be screened for effective gene-specific GSCs in a high throughput fashion. The populations of GSCs obtained confer a wide range of levels of, for example, suppression and provide the potential to quantitatively modulate the expression of a specific target gene. Furthermore, the more effective GSCs and other molecules capable of modulating target gene expression have the potential to completely suppress gene expression at the mRNA level and produce phenocopies of homozygous null mutants. The universal nature of the fusion approach provides a valuable tool for identifying GSCs and other molecules capable of modulating target gene expression, useful in validating potential novel drug targets and better understanding the biological function of unknown genes.

Example 4 Additional human target genes In addition to the c-myc gene, two additional human gene sequences have been expressed as lacZ fusions in yeast and gene-specific expression libraries screened for GSCs using the lacZ reporter gene-encoded phenotype. The two target genes were Hradl and Hrad27. These genes play potential roles in cell cycle checkpoint functions.

The methods for screening expression libraries using the blue colony colour phenotype, as described in the present patent, were used to screen gene-specific random libraries in yeast for Hradl and Hrad27 GSCs. Figure 6 shows the inserts contained on the GSCs specific for Hradl and their alignment with the target gene sequence. The level of suppression of P-galactosidase activity displayed by these GSCs ranged from 21% to Figures 7 and 8 summarise the arrangement of inserts derived from the Hrad27 GSCs and the level of suppression of the fusion gene target in yeast. In both of these figures, the target gene sequence fused to the amino terminal end of the lacZ reporter is indicated at the top of the diagram. The labels "ATG" and "TGA" refer to the translation initiation and stop codons, respectively. "EVS" indicates the penultimate codon relative to the stop codon. In the screening protocol for these two target genes, the expression library construction process was modified from that indicated in Figure 1.

More specifically, the random fragments derived from each of the target genes were cloned directly into the yeast expression plasmid pGT 18 without first constructing an 15 intermediate library (step 3 in Figure 1).

Although the present invention has been described with reference to specific examples, in particular those describing the use of S. pombe as the model organism and

S.P

3 -galactosidase as the reporter molecule, it will be appreciated by those skilled in the art that the present invention clearly extends to and encompasses other suitable eukaryotic organisms and other reporter molecules, in keeping with the broad concepts and principles of the invention described herein.

Claims (20)

1. 2. A Schizosaccharomycespombe cell comprising the fusion construct according to claim 1.
2. 3. A method for determining whether, or to what degree, a molecule specifically modulates the expression of a target nucleic acid sequence in a eukaryotic cell, which comprises the steps of determining for the target sequence the conditions required for the expression of the reporter sequence or for the activity of its 15 expression product in the Schizosaccharomyces pombe cell of claim 2, contacting the Schizosaccharomyces pombe cell with the molecule under said conditions, and detecting or quantitating a phenotypic change in the Schizosaccharomyces pombe cell resulting from the modulation of the expression of the reporter sequence, thereby determining whether, or to what degree, the molecule specifically modulates the expression of the target sequence in a eukaryotic cell.
3. 4. A method according to claim 3, wherein the molecule is a biological macromolecule or a fragment thereof, selected from the group consisting of a polynucleotide, a protein, a carbohydrate and a lipid. A method according to claim 4, wherein the molecule is selected from the group consisting of an antisense nucleotide sequence, a sense nucleotide sequence, a chimeric nucleotide sequence, a ribozyme, a random nucleotide sequence, a transdominant polypeptide, a small peptide, a nucleotide, and a peptide analogue molecule.
4. 6. A method according to claim 3, wherein the molecule is a simple organic or inorganic chemical substance. -27-
5. 7. A method according to claim 3, wherein the molecule is capable of up- regulating or down-regulating expression of the target sequence, or the activity of its product.
6. 8. A method according to claim 3, wherein the target nucleic acid sequence is selected from the group consisting of a gene encoding a eukaryotic, a prokaryotic or a viral protein, or a fragment or homologue thereof.
7. 9. A method according to claim 8, wherein the target sequence is a mammalian gene sequence.
8. 10. A method according to claim 8, wherein the target sequence encodes all or part of a protein selected from the group consisting of a growth factor, a cytokine, a receptor, a ligand, a cell-cycle regulatory protein, an envelope or surface membrane protein and an enzyme.
9. 11. A method according to claim 8, wherein the target sequence is an oncogene or a proto-oncogene. to. 15 12. A method according to claim 11, wherein the target sequence is of viral origin.
10. 13. A method according to claim 12, wherein the virus is a retrovirus.
11. 14. A method according to claim 3, wherein the product encoded by the reporter sequence is selected from the group consisting of an enzyme, a growth factor, a cytokine, a receptor, a chemiluminescent or fluorescent protein and a cell-cycle regulatory protein. A method according to claim 3, wherein the fusion construct comprises a homologous and a heterologous sequence.
12. 16. A method according to claim 3, wherein both the target sequence and the reporter sequence are heterologous.
13. 17. A method according to claim 16, wherein the fusion construct includes a mammalian target and reporter sequence.
14. 18. A method according to claim 14, wherein the reporter sequence is the lacZ gene sequence or a fragment thereof. 28
15. 19. A method according to claim 14, wherein the reporter sequence is the ura4 gene sequence or a fragment thereof. A method according to claim 3, wherein the phenotypic change is selected from the group consisting of colour, size, shape and texture of the Schizosaccharomyces pombe cell or colony.
16. 21. A method according to claim 20, wherein the phenotypic change is optically detectable. 0*0
17. 22. A method according to claim 20, wherein the phenotypic change is chemically or enzymatically detectable. 10 23. A method according to claim 20, wherein the phenotypic change is a change in the colour of the Schizosaccharomyces pombe cell or colony.
18. 24. A method according to claim 20, wherein the phenotypic change is a change in the size of Schizosaccharomyces pombe cell or colony.
19. 25. A method according to claim 3, wherein the expression of the fusion 15 construct is under the control of an endogenous promoter.
20. 26. A method according to claim 3, wherein the expression of the fusion construct is under the control of an exogenous promoter. DATED this 8th Day of January, 1999 JOHNSON JOHNSON RESEARCH PTY. LIMITED Attorney: PAUL G. HARRISON Fellow Institute of Patent Attorneys of Australia of BALDWIN SHELSTON WATERS