AU7705500A - Methods for site-associated modification of gene activity and nucleic acid structure - Google Patents

Description

WO 01/21781 PCT/USOO/25778 METHODS FOR SITE-ASSOCIATED MODIFICATION OF GENE ACTIVITY AND NUCLEIC ACID STRUCTURE TECHNICAL FIELD This invention relates generally to methods for making transgenic plants with modified gene activity or nucleic acid structure. BACKGROUND OF THE INVENTION Directed manipulation of gene activity affords a valuable approach to generating organisms with phenotypic changes. Altering expression of genes can lead, however, to non-viable organisms. Conventional insertion mutagenesis that relies on gene inactivation has limitations. As many as two-thirds of all such mutated genes in Drosophila, Caenorabditis elegans, yeast, and mouse, may exhibit no apparent loss-of function phenotype. This may be due to any of several reasons: the functional redundancy of the genome, the lethality of many mutations when in a homozygous state, or subtle phenotypes that can only be assessed under competitive conditions in nature, or for which the existing morphological variation is inadequate to distinguish. A strategy has been devised to identify and gain functional insight into such genes. This strategy seeks to identify regions of the genome that are expressed under specific developmental conditions, with the goal of finding genes that mediate specific processes, such as root development or flower formation. It entails an intensive effort to find genes that are differentially expressed. As expression differences that lead to dramatic phenotypic changes may involve subtle pattern differences or as there may be multiple genes necessary for mediating processes, this strategy may fail to identify the key genes. A second strategy seeks to generate gain-of-function mutations, in which a developmental pathway is expressed in the wrong place and/or time, to result in a novel phenotype that provides insight into the function of the mis-expressed gene. In the most common implementation of this strategy, a desired gene is placed under WO 01/21781 PCTIUSOO/25778 2 control of a promoter that is active only in a particular tissue or developmental time. This method, however, requires first that a promoter and desired gene be identified and cloned. In an attempt to find such regulatory elements, e.g., enhancers, a reporter target gene is separated from its transcriptional activator in two distinct transgenic lines (Brand and Perrimon, Development 118: 401, 1993; Guyer et al., Genetics 149: 633, 1998). When the gene encoding the transcriptional activator is introduced under control of a minimal promoter, the activator is only expressed when the element integrates near genomic regulatory sequences. In the second transgenic line, the target gene is linked to a sequence recognized by the transcriptional activator. By screening for expression of the reporter in different tissues and times, lines can be identified in which the transcriptional activator is nearby to a desired regulatory sequence. Thus, lines are produced that can be used to express any gene of interest with the same expression pattern as that exhibited by the transgenic organism containing the transcriptional activator gene. While this scheme allows control of the expression of a single chosen target gene, the number of target genes that can be manipulated is severely limited by the necessity of preparing a separate transgenic construct for each target gene. Furthermore. this scheme limits the number of potential phenotypic changes by the number of target genes available. A system that allows manipulation of a large number of target genes offers many advantages, especially by not requiring cumbersome, time and resource-consuming individual constructs. In addition, the manipulation of a large number of target genes greatly increases the likelihood of generating a desired phenotype. The present invention discloses methods for generating transgenic plants that exhibit novel phenotypes, transgenic plants, and further provides other related advantages.

WO 01/21781 PCT/USOO/25778 3 SUMMARY OF THE INVENTION Within one aspect of the present invention, methods are provided comprising, cross fertilizing two transgenomic plant lines to produce seed, wherein the first transgenomic plant line contains an introduced first nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule under control of a minimal promoter, and the second transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous site-specific nucleic acid effector molecule; and growing the seed to produce a plant; wherein the introduced first nucleic acid molecule is operably linked to an endogenous enhancer sequence; and wherein the binding of the effector molecule to the binding site modifies gene activity in a binding site-associated manner. In one embodiment, the non-native nucleic acid effector molecule is a fusion protein of a plant-derived effector molecule with a heterologous DNA binding domain. In another embodiment, the non-native effector molecule is a plant-derived molecule with an altered DNA binding domain that recognizes a different recognition sequence. In another aspect, the method is used to affect nucleic acid structure. In yet other aspects, transgenic plants and seeds produced by the cross of the two transgenic plants are provided. In another aspect, methods are provided for modifying gene activity in a plant, comprising the steps of: generating a transformed plant line with an introduced nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule under control of a minimal promoter and a transposable element containing a minimal promoter operably linked to at least one binding site for the nucleic acid effector molecule; cross-fertilizing the transformed plant line with a second plant line that expresses a transposase to produce seed; and growing the seed into a plant. In certain embodiments, plants that display a desired phenotype are selected. In particular embodiments, the transposable element is Ds and the transposase is Ac and the introduced nucleic acid molecule further comprises a reporter gene operably linked to at WO 01/21781 PCT/USOO/25778 4 least one binding site for the nucleic acid effector molecule. In preferred embodiments, the site-specific nucleic acid effector molecule is a transactivator. These and other aspects of the present invention will become evident upon reference to the following detailed description and attached drawings. In addition, various references are set forth below which describe in more detail certain procedures or compositions (e.g., plasmids, etc.), and are therefore incorporated by reference in their entirety. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic of an exemplary transactivating vector. Figure 2 is a schematic of an exemplary transactivating vector, pSMR J18R, which contains a GFP reporter gene. Figure 3 is a schematic of an exemplary transactivating vector, DsMutagenvector, which contains a GUS reporter gene. DETAILED DESCRIPTION OF THE INVENTION Prior to setting forth the invention, it may be helpful to an understanding thereof to set forth definitions of certain terms that will be used hereinafter. As used herein, "site-specific nucleic acid effector molecule" refers to a molecule that binds a specific recognition sequence in a nucleic acid molecule and has a function that affects nucleic acid. The effector molecule may be protein, peptide, DNA, RNA, a chimera of these molecules, or the like. As described in more detail herein, functions include, but are not limited to, activating transcription, unwinding DNA, methylating DNA. Generally, the nucleic acid binding and function will reside in separate domains. A "binding site" or "recognition sequence" for a site-specific nucleic acid effector molecule" refers to the specific nucleic acid sequence that the nucleic acid effector molecule binds to.

WO 01/21781 PCT/USOO/25778 5 As used herein, "enhancer" refers to any one of a class of cis-acting DNA sequences that function in a non-directional manner and that increase transcriptional activity of an operably linked promoter. As used herein, a "promoter" means a nucleotide sequence comprising one or more sequences that proteins and other molecules that are involved in the transcription process bind to and initiate transcription from. A "minimal promoter" refers to a promoter sequence that cannot initiate detectable transcription in the absence of additional transcriptional elements. As used herein, "transcriptional elements" refers to nucleotide sequences involved, directly or indirectly, in regulating transcription of cis-linked genes. Such elements include, but are not limited to, enhancers, promoters, transcriptional termination sequences, and the like. The term "operably linked" refers to the association of nucleic acid sequences on a single nucleic acid fragment so that the function of one is affected by the other. For example, a promoter is operably linked to a coding sequence when it is capable of affecting the expression of that coding sequence. As used herein, a "plant line" means a group of individual plants derived from a common ancestry. As used herein, "phenotype" or "trait" refers to a set of observable physical characteristics of an individual organism, which is the observed manifestation of a genotype. A phenotype may be expressed physically, biochemically, or physiologically. A. PARENTAL TRANSGENIC PLANT LINES As described above, in one aspect the parental transgenic plant lines comprise a line containing an introduced nucleic acid molecule that expresses a site specific nucleic acid effector molecule and a second line containing a second introduced nucleic acid molecule having a sequence that the effector molecule specifically binds. In another aspect, there are three parental lines, one containing a gene encoding a WO 01/21781 PCT/USOO/25778 6 transcriptional activator, a second containing a recognition sequence for the transcriptional activator that is operably linked to a gene encoding a different site specific nucleic acid effector molecule, and a third line that contains linked binding sites for the transcriptional activator and the nucleic acid effector molecule. In either aspect, the parental lines are crossed to produce progeny that have modified endogenous gene activity or modified nucleic acid structure. In yet another aspect, the parental line contains an introduced nucleic acid molecule that has a site-specific nucleic acid effector molecule (e.g., a transcriptional activator) operably linked to a minimal promoter and at least one binding site operably linked to a minimal promoter. Depending on which nucleic acid effector molecules are chosen, gene activity may be enhanced, suppressed, or the like and the nucleic acid structure may be modified by unwinding, overwinding, nicked, cut, or the like. 1. Site-specific nucleic acid effector molecules and recognition sequences As noted above, one parental transgenic plant line contains a nucleic acid molecule that expresses a site-specific nucleic acid effector molecule. Within the context of this invention, the nucleic acid effector molecule may be proteinaceous, such as a peptide or protein, a nucleic acid, such as single or double-stranded DNA or RNA or DNA:RNA hybrids, protein-nucleic acid chimeras, such as PNA or conjugates, and the like. Whatever its form, the nucleic acid effector molecule must bind to a nucleotide recognition sequence in a site-specific manner. In the context of the present invention, site-specificity means that that the nucleic acid effector molecule binds in a manner dependent upon the sequence of the recognition site. For example, a restriction enzyme such as EcoRI binds to the sequence 5'-GAATTC-3' when present in a double stranded form. The nucleic acid effector molecule may be any molecule that affects either the structure or function of nucleic acids. Within the context of this invention, the nucleic acid effector molecule should affect nucleic acids that are cis-linked to the recognition sequence. It is preferred that the site of action occurs only or preferentially occurs in a localized area to the recognition sequence. In certain embodiments, the WO 01/21781 PCT/USOO/25778 7 nucleic acid effector molecule comprises at least an active fragment of a transcription factor, a methylase, a repressor, a gyrase, a kinkase, a histone deactylase, a histone acetylase, a topoisomerase, an enhancer, any one of a suite of transcriptional factors, a restriction enzyme, or the like. Methylases, gyrases, histone deactylases and acetylases, and the like are well-known in the art. Sequences that encode these proteins may be found in GenBank for example. In another embodiment the nucleic acid effector molecule has a null function, such that binding of the molecule to the recognition sequence blocks other molecules from binding, but otherwise has no direct effect on the nucleic acid. One means to constructing a null function effector molecule is to use only the nucleic acid binding domain. Transcription factors regulate gene expression and are a diverse group of proteins. Host transcription factors have been grouped into seven well-established classes based upon the structural motif used for recognition. Major families include helix-turn-helix proteins, homeodomains, zinc finger proteins, steroid receptors, leucine zipper proteins, helix-loop-helix proteins, and p-sheets. Other classes or subclasses may eventurally be delineated as more factors are discovered and defined. These families of transcription factors are generally well-known (see GenBank; Pabo and Sauer, Ann. Rev. Biochem. 61: 1052-1095, 1992). Many of these factors are cloned and the precise DNA-binding region delineated in certain instances. When the sequence of the DNA-binding domain is known, a gene encoding it may be synthesized, cloned from host genomic or cDNA libraries, or amplified from cDNA or genomic DNA. Examples of transcription factors useful in the context of this invention include gus repressor (U.S. Patent No: 5,879,906); Gal4 (U.S. Patent No: 5,968,793); HIV rev; cro, lac repressor, glucocorticoid receptor, trp repressor, TFIIIA, Sp-1, GCN4, AP-2, and the like (Pabo and Sauer, Ann. Rev. Biochem. 61: 1052-1095, 1992). In certain embodiments, the nucleic acid effector molecule is a fusion protein comprising a peptide or polypeptide that confers binding to a recognition sequence and a protein that affects the function or structure of nucleic acids. In a specific embodiment, nucleic acid binding transcription factors may be generated as a WO 01/21781 PCT/US00/25778 8 fusion protein of a nucleic acid-binding domain and a transcriptional effector domain. As used herein, "transcriptional effector domain" means a polypeptide that effects transcription, such as activating and inhibiting. The domain may be a portion or fragment of a larger molecule (e.g., the activation domain of Gal4) or a full-length molecule (e.g., VP16 or Gal4). Construction of these fusion proteins is preferably accomplished by amplification of the desired domain regions and ligation of the amplified products. Thus, primers flanking a DNA binding domain, selected from a DNA-binding protein and a effector domain, including activators and repressors, are useful within the context of this invention. Compatible restriction sites are preferably incorporated into the primers, such that the products when joined are in the same reading frame. Amplified products of the two domains are restricted and ligated together and inserted into an appropriate vector. Verification of the resulting clone is readily done by restriction mapping and DNA sequence analysis. DNA sequence analysis is preferable so that an in-frame reading frame can be verified. One of skill in the art recognizes that other routine methods and procedures may be alternatively used. In other embodiments, the nucleic acid effector molecule is a conjugate of a protein that affects the function or structure of nucleic acids and a oligonucleotide or polynucleotide that binds to the recognition sequence. As used herein, an oligonucleotide is a short polynucleotide. Typically, an oligonucleotide is from a few bases to a hundred or so, and more typically from 10 to 100 bases. The oligo- or poly nucleotide preferably comprises a sequence complementary to the recognition sequence. In other embodiments, the nucleic acid effector molecule is selected to bind a defined recognition sequence. In certain embodiments, desirable recognition sequences include sequences found in natural transposable elements, such as Ds from maize. Techniques for selecting such characteristics are well-known in the art and include phage display (see for example, U.S. Patent No: 5,223,409, gene shuffling, site-directed mutagenesis, and the like. In another embodiment, variants are generated by "exon shuffling" (see U.S. Patent No. 5,605,793). Variant sequences may also be generated by "molecular evolution" techniques (see U. S. Patent No. 5,723,323).

WO 01/21781 PCT/USOO/25778 9 In other embodiments, the nucleic acid effector molecule may be a nucleic acid, such as single-stranded or double-stranded DNA or RNA. In such cases, the nucleic acid effector molecule may be complementary to the recognition sequence and effect binding as well as affecting the function or structure of cis-linked nucleic acids. In other cases, the nucleic acid effector molecule may be a conjugate of a nucleic acid and an effector protein. Within the context of this invention, a nucleic acid effector molecule will affect either the function or structure of nucleic acids that are cis-linked to the recognition sequence. Function may be affected either directly or indirectly. For example, but not limited to, direct action includes the enhancement of gene expression through activation of a promoter and the diminishing of gene expression through action of a repressor. Indirect action includes, but is not limited to, alterations of nucleic acid structure, such as by deactylation of histones, association of the nucleic acid effector molecule with a factor necessary for transcription thus inhibiting the action of the factor, and the like. Structure can be affected in a large variety of means, such as by unwinding or increasing the winding of nucleic acids, increasing or decreasing the number or type of chromatin-associated proteins, causing nucleic acid kinking, and the like. As noted above, the recognition sequence for the nucleic acid effector molecule is a specific sequence such that the nucleic acid effector molecule binds in a sequence-specific manner. Generally, the recognition sequence will be at least three nucleotides and preferably at least four nucleotides, at least five, six, seven or eight nucleotides. While there is no theoretical upper limit to the length of the recognition sequence, pragmatically the sequence will not be more than 50 nucleotides and in certain embodiments, not more than 40, 30, 25 or 20 nucleotides. The recognition sequence for many DNA-binding protein is well-known. For example, a widely-used recognition sequence for Gal4 is a consensus sequence of the four Gal4 sites found in the UAS controlling expression of the yeast GAL1 and GAL10 genes and the one Gal4 site found in the GAL7 promoter (Giniger et al., Cell 40:767-774, 1985). These WO 01/21781 PCT/USOO/25778 10 different sequences demonstrate some of the variation that can be made in the Gal4 binding site while maintaining the ability of Gal4 to bind. Other suitable binding sequences include sequences found in natural transposons. In such embodiments, one of the parental lines carries one or more, and preferably many copies of the transposon and is crossed to a line that expresses a site-specific nucleic acid effector molecule under control of a minimal promoter. 2. Populations of transgenic plant lines In one aspect of this invention, parental plant lines containing nucleic acid effector molecules and recognition sequences are generated. Within the context of this invention, multiple plant lines of each parental type are desirable. As each transgenic line will likely carry the introduced nucleic acid molecule in a unique location, a larger number of lines will increase the number of new phenotypes observed. There are a variety of methods that may be used to generate the plant lines. For exemplary purposes only, several methods are described herein. The first method is based on T-DNA integration into a plant genome during Agrobacterium transformation. Examples of suitable vectors for this method are described below. In the initial phase, 10,000 independent primary transgenic lines of rice are generated, resulting in insertions in approximately 10-20% of rice genes (with an average number of T-DNA copies of 1.5-2 per transgenic). In 251 lines analyzed, about 48% of lines have a single T-DNA insertion. The physical location of each T-DNA insertion can be mapped by standard techniques, such as fluorescent-labeled hybridization to chromosomes, similarity searching against DNA sequence data, restriction mapping, and each corresponding line scored for enhancer trap activity and homozygous phenotype. A subset of these lines that carry single copy insertions evenly distributed throughout the genome and exhibit interesting patterns of enhancer trap expression may be chosen for saturation mutagenesis. Saturation mutagenesis or coverage of a complete genome with insertional mutagens may be achieved in one of a variety of methods. Briefly, in one such method, a Ds transposable element, which is part of the Agrobacterium vector is WO 01/21781 PCT/USOO/25778 11 activated to transpose itself from the T-DNA insertion site. Other transposition systems in which transposition can be controlled, such as by expression of a transposase are also suitable within the context of the present invention. Transposition is achieved by crossing transgenic plants carrying a T-DNA insert with plants expressing Ac transposase. Further control of transposition is attained by using different promoters to regulate Ac transposase. Promoters that are active in the germline can be used. Of particular interest are promoters driving transcription during inflorescence meristem development, as such fusions are likely to provide a high frequency of germinal excisions of the Ds element while exhibiting low levels of somatic excisions. Somatic excisions (especially early ones) may contribute unwanted "background" variation, potentially masking/modifying mutant phenotypes resulting from germinal excisions. By incorporating a reporter gene, such as P-glucuronidase or green fluorescent protein, in the construct under control of the same promoter, the activity of the promoter can be readily monitored. For exemplary purposes only, promoter fusions with Ac transposase are introduced into Nipponbare rice plants that are already transformed with a construct comprising the Ds element and the marker gene p-glucuronidase (GUS). These plants do not express GUS unless the Ds element is excised. These tester plants allow quantification of germinal and somatic excision events by simple GUS staining of respective tissues. A selected set of transcriptional activator facilitated enhancer trap/Ds transposon (TA ET/Ds) lines are crossed with an Ac transposase-expressing line. The number of TA ET/Ds lines and the number of crosses per line is determined by the frequency of germinal excisions. With an excision frequency of approximately 10%, between 500-1000 TA ET/Ds lines will yield a number of Ds insertions sufficient for complete genome saturation with a very modest amount of crossing involved. Even with excision frequencies below 1% the number of sexual crosses needed for complete saturation can be easily achieved in a short amount of time. In an exemplary vector system, the integrated T-DNA of the TA-ET/Ds comprises a 'Launching Pad' and contains a non-autonomous Ds transposable element.

WO 01/21781 PCTIUSOO/25778 12 This element (Figure 1) contains not only a dominant herbicide resistance marker that can be used under field conditions, but an 'outward-facing' UAS-minimal promoter at one side. As will be appreciated by those in the art, UAS is but one example of a recognition sequence. This Ds element is positioned between a promoter active in seedlings, and an antibiotic resistance (e.g., hygromycin resistance; kanamycin resistance) gene. Other selectable markers for plants are well known and may alternatively be used. Thus, when the Ds-element is induced to transpose by provision of the appropriate transposase enzyme through a genetic cross to a line engineered to express the transposase, the promoter and the resistance gene are brought together, and result in a seedling resistant to the antibiotic. This is referred to as an 'excision marker'. The biology of the Ac/Ds system ensures that most of the excision events are correlated with an integration event somewhere in the genome, typically at a genetically linked site. The progeny of TA ET/Ds lines and Ac-expressing lines are selected for Ds excision and reinsertion using Selectable Marker I and Selectable Marker II (Figure 1). Selection against the presence of Ac transposase gene is performed at the same time (germination to seedling stage) using a negative selection marker present on the same T-DNA as Ac-construct. Although not required, elimination of Ac-positive plants will stabilize Ds in the new location. Two alternative approaches can be used to increase numbers of inserted recognition sequences. One is transposon (e.g., Ac/Ds) based, the other is not. In the transposon-based approach, a gemini virus is used as a vehicle for generating high copy number Ds element substrates for Ac transposase. The Gemini virus vector contains the Ds element with recognition sequences operably linked to a minimal promoter. The virus may also contain a transposase to initiate transposition. Alternatively, the plant cell infected with the Gemini virus may be super-infected with a vector expressing transposase or the plant cell may express transposase. A second strategy involves a transposon-independent approach to recognition-sequence mutagenesis based on co transformation of DNA molecules that provide selection function in callus and a molar WO 01/21781 PCT/USOO/25778 13 excess of DNA molecules that contain recognition sequence-minimal promoter mutagen. Each insertion has a "unique identifier" - DNA sequence tag, which allows not only easy identification of specific mutagens elements in segregating populations, but also quantification of the transcripts originating from a specific recognition sequence-minimal promoter. 3. Enhancer Traps As described herein, in one aspect the nucleic acid effector molecule is under control of a minimal promoter and requires the presence of a cis-acting enhancer sequence for expression. When the introduced nucleic acid comprising the nucleic acid effector molecule construct integrates near an endogenous enhancer the nucleic acid effector molecule will be expressed. Moreover, the nucleic acid effector molecule will be expressed in different cell types and at various developmental stages depending on when and where the enhancer is activated. By optionally including a reporter gene under control of a minimal promoter in the construct, transcription can be readily monitored. The reporter gene is not expressed by the minimal promoter per se. Instead, insertion of the element within or fairly close to a gene in the target genome may result in reporter gene expression being activated by the regulatory sequences of the tagged gene. Thus, if GUS is used as the reporter gene, the presence of GUS activity in a particular organ or cell at a particular developmental stage will identify sequences expressed at this place and time. Indeed, GUS enhancer traps have proved successful in detecting novel genes in Arabidopsis. An exemplary vector (Figure 1) functions as a transactivator enhancer trap, as it carries the Gal4/VP16 gene under the control of a minimal promoter adjacent to the right T-DNA border. The construct also carries marker genes, either GUSPlusTM (see, WO 99/13085) or both the GUSPlus T M and GFP (green fluorescent protein) reporter genes, under the control of UAS elements to provide sensitive readouts of patterns of expression of the transactivator protein. Other exemplary vectors are shown in Figures 2 and 3. In systems such as Drosophila and Arabidopsis, approximately 50% of enhancer trap lines generate exhibit specific patterns of expression. Furthermore, a WO 01/21781 PCT/USOO/25778 14 remarkably broad range of different expression pattern lines are obtained, from those where the reporter gene is expressed throughout the organism to those where it is only expressed in one or a few cells during a very brief period of development. Therefore, a significant proportion of primary transformants will be useful enhancer-trap pattern lines that display regulated patterns of expression, and a large collection of pattern lines that exhibit a very diverse set of discrete expression patterns will be obtained. The various components of the vector may be interchanged with other components that accomplish similar function. Some of these interchanges have been discussed above (e.g., nucleic acid binding protein). For example, the reporter can be any protein that allows convenient and sensitive measurement or facilitates isolation of the gene product and does not interfere with the function of the telomerase. For reporter function, p-glucuronidase (U.S. Patent No: 5,268,463 and 5,599,670), green fluorescent protein (GFP), GUSPlusTM (WO 99/13085) and p-galactosidase are readily available as DNA sequences. A peptide tag may be additionally be used. A tag is a short sequence, usually derived from a native protein, which is recognized by an antibody or other molecule. Peptide tags include FLAG®, Glu-Glu tag (Chiron Corp., Emeryville, CA) KT3 tag (Chiron Corp.), T7 gene 10 tag (Invitrogen, La Jolla, CA), T7 major capsid protein tag (Novagen, Madison, WI), His 6 (hexa-His), and HSV tag (Novagen). Besides tags, other types of proteins or peptides, such as glutathione-S-transferase may be used. Similarly, selectable markers include, hygromycin resistance (U.S. Patent Nos: 4,727,028; 4,960,704; and 5,668,298) G418 resistance, ampicillin resistance, kanamycin resistance, other antibiotic resistance genes, positive selection genes, use of beta-glucuronidase transgene and application of cytokinin-glucuronide for selection and use of mannophosphatase or phosphmanno-isomerase transgene and application of mannose for selection (U.S. Patent Nos. 5994629; 5767378; and 5,599,670). The present invention also provides an opportunity to use the information and genetic resources developed through many years of plant genetic research. For example, a large number of "classical" mutants are mapped genetically and genes responsible for mutant phenotypes may be targeted for Ds mutagenesis by WO 01/21781 PCTUSOO/25778 15 launching TA ET/Ds element mapping in the vicinity. This is done through crossing respective TA ET/Ds lines with Ac transposase expressing lines. Alternatively, DNA isolated from the primary and secondary mutant populations may be used in a reverse genetics strategy to identify lines that carry insertional mutations within specified genes. The information and genetic resources available are not limited to a single species. For example, one can also exploit the enormous information resources generated for other cereals, due to grass genome synteny. Mutant phenotypes identified and mapped in other cereal species, are likely to be in syntenic location in the rice genome, and their "tagging" may be done in rice. Searching for corresponding genes in rice will not only offer a shortcut, but, once a target gene is isolated through this synteny-with-rice approach, sequence similarity between homologues will provide many clues about regions important for gene regulation and function as a bonus. There is little doubt that this evolutionary approach will be beneficial both for rice and other cereal and grass research and breeding. 4. Construction ofparental lines The transgenic parental lines are constructed by introduction of a nucleic acid molecule, typically a DNA-based vector, containing one or more of the genes noted above. The vectors should be functional in plant cells. Vectors and procedures for cloning and expression in E. coli and animal cells are discussed herein and, for example, in Sambrook et al (supra) and in Ausubel et al. (supra). Vectors that are functional in plants are preferably binary plasmids derived from Agrobacterium plasmids. Such vectors are capable of transforming plant cells. These vectors contain left and right border sequences that are required for integration into the host (plant) chromosome. At minimum, between these border sequences is the gene to be expressed under control of a promoter. In preferred embodiments, a selectable marker and a reporter gene are also included. The vector also preferably contains a bacterial origin of replication. Available vectors include co integrative vectors and variants (U.S. Patent No: 5,731,179, 5,635,381; 4,693,976;), WO 01/21781 PCT/USOO/25778 16 binary vectors (U.S. Patent Nos: 4,940,838; 5,464,763) and variants (U.S. Patent No: 5, 149,645). The promoter should be functional in a plant cell. The promoter may be derived from a host plant gene, but promoters from other plant species and other organisms, such as insects, fungi, viruses, mammals, and the like, may also be suitable, and at times preferred. For a minimal promoter, the promoter region is designed narrowly to comprise a transcription initiation site, but lack sequences and elements necessary for activity. One such promoter is the minimal promoter derived from CaMV 35S promoter, which comprises a TATA box, needed for assembly of active RNA transcription complex, and supports only basal, non-detectable reporter activity. The minimal promoter region typically used encompasses nucleotides -46 to +8 of CaMV 35S promoter (see U.S. Patent 5,097,025). Preferably, the vector contains a selectable marker for identifying transformants. A selectable marker confers a growth advantage under appropriate conditions. Generally, selectable markers are drug resistance genes, such as neomycin phosphotransferase. Other drug resistance genes are known to those in the art and may be readily substituted. For example, hygromycin resistance (U.S. Patent Nos: 4,727,028; 4,960,704; and 5,668,298) G418 resistance, ampicillin resistance, and kanamycin resistance genes are commonly employed. Alternatively, positive selection systems, such as taught in U.S. Patent Nos. 5994629; 5767378; and 5,599,670 may be used. The selectable marker also preferably has a linked constitutive or inducible promoter and a termination sequence, including a polyadenylation signal sequence. Additionally, a bacterial origin of replication and a selectable marker for bacteria are preferably included in the vector. Of the various origins (e.g., colEl, fd phage), a colEl origin of replication is preferred. Most preferred is the origin from the pUC plasmids, which allow high copy number. Selectable markers for bacteria include, ampicillin resistance, tetracycline resistance, kanamycin resistance, chloramphenicol resistance, and the like.

WO 01/21781 PCT/USOO/25778 17 General vectors suitable for use in the present invention are based on the vectors described above and in U.S. Patent No. 4,940,838 and 5,464,763 and pBIl21 (U.S. Patent No. 5,432,081) a derivative of pBIN19. Other vectors have been described (U.S. Patent No. 4,536,475) or may be constructed based on the guidelines presented herein. The plasmid pSMR-J18R, an exemplary plasmid used in the present invention (see Figure 2), contains a left and right border sequence for integration into a plant host chromosome and also contains bacterial origins of replication and ampicillin selectable marker for bacterial growth. The border sequences flank several genes. One is a hygromycin resistance gene driven by a CaMV 35S promoter and using a nopaline synthase polyadenylation site. A second gene is the Streptomyces GUS gene (GUSPlus TM or BoGUS) operably linked to six copies of Gal4 UAS. Furthermore, a gene encoding a fusion protein of a nucleic acid binding domain (Gal4) and a transcriptional activator (VP16) is located near the right border. Other elements may also be included. Plants may be transformed by any of several methods. For example, plasmid DNA may be introduced by Agrobacterium co-cultivation (U.S. Patents 4,940,838 and 5,591,616, or bombardment. Other transformation methods include electroporation (U.S. Patent No. 5,859,327), CaPO,-mediated transfection, gene transfer to protoplasts, microinjection, and the like (see, Gene Transfer to Plants, Ed. Potrykus and Spangenberg, Springer, 1995, for procedures). Preferably, vector DNA is first transfected into Agrobacterium and subsequently introduced into plant cells. Most preferably, the infection is achieved by co-cultivation. In part, the choice of transformation methods depends upon the plant to be transformed. For example, monocots can be transformed by Agrobacterium using the methods claimed in U.S. Patent 5,591,616 and U.S. 6,037,522. (See also, U.S. Patent No 6,074,877 for transformation of cerials; U.S. Patent No. 5,177, 010 and 5, 981,840 for transformation of maize; U.S. Patent No. 5, 187,073 and 6,020,539 for transformation of Gramineae) Agrobacterium transformation by co-cultivation is also appropriate for dicots and for mitotically active tissue. (See U.S. Patent No: 5,188,958; 5,463,174; and 5, 750,871 for WO 01/21781 PCT/USOO/25778 18 transformation of Brassica spp.; U.S. Patent Nos: 5,416,011; 5, 569,834; 5,824,877; 5, 563,055; and 5, 968,830 for transformation of soybean; U.S. patent Nos: 5, 846,797 and 5, 004,863 for transformation of cotton; U.S. Patent No 5, 565,347 for transformation of tomato.) Non-mitotic dicot tissues can be efficiently infected by Agrobacterium when a projectile or bombardment method is utilized (U.S. Patent No: 5,932,782). Projectile methods are also generally used for transforming plants. (See U.S. Patent Nos: 5,976,880; 5,877023; 5,179,022; 5,036,006; 4,945,050; 5,371,015; 5,100,792; for methods and apparatus.) Bombardment is most often used with naked DNA, typically Agrobacterium or pUC-based plasmids, for transformation or transient expression. Briefly, co-cultivation is performed by first transforming Agrobacterium by freeze-thawing (Holsters et al., Mol. Gen. Genet. 163: 181-187, 1978) or by other suitable methods (see, Ausubel, et al. supra; Sambrook et al., supra). A culture of Agrobacterium containing the plasmid is incubated with leaf disks, protoplasts or meristematic tissue to generate transformed plants (Bevan, NucL. Acids. Res. 12:8711, 1984) or for monocots, incubated with callus derived from explants. B. ASSAYS FOR CHANGE IN PHENOTYPE OF HYBRID LINES Transgenic plant lines generated according to the teachings of the present invention will allow gene identification and generation of new phenotypes via a combination of approaches: The Ds element harbored on the primary T-DNA insertion in each of these lines is launched to generate a large second set of insertional mutations. These mutations are then mapped and screened in a variety of ways, including for gain of-function phenotypes in a heterozygous state and for loss-of-function phenotypes in a homozygous state. 1. Loss-of-Function (LoF) mutations Insertional mutagenesis has proved effective for cloning genes in a variety of organisms and causing a wide range of novel phenotypes. T-DNA or transposable element insertion within the coding region of the gene (an exon) will usually completely abolish gene function leading to a null mutation. When insertion WO 01/21781 PCT/US00/25778 19 occurs in the promoter region, 5' or 3' untranslated region, or intron region it may lead to substantial reduction of the amount of gene product being expressed and a hypomorph mutation as a result. While there is almost no limit to the type of phenotypic alterations one can search for in the mutagenized population, several classes of LoF mutations are of especial interest. One of the most interesting and abundant types of mutations to be identified in the T-DNA and Ds-tagged population include those leading to male and/or female sterility. Male sterility mutations may prove very useful in developing genetic systems for hybrid seed production. Female sterility will be an indication of an insertion of the mutagenic element into a gene essential for female reproductive functions, which may result in discovery of a gene important for development of apomictic character. Another class of desirable phenotypes are related to floral structure or other development alterations that can facilitate cross-pollination, such as precocious lodicule expansion, stigma exersion or prolonged flower-opening times. The present invention will also allow access to the genes leading to lethal phenotypes, due to the ability to maintain recessive lethals using the dominant herbicide locus associated with the insertional lesion. Amplifying sequences proximal to the sterility-causing insertion site, together with molecular characterization (high resolution mapping and sequencing) will allow rapid access to candidate genes. 2. Gain-of-Function (GoF) mutations Gain-of-function mutations can result in organisms having novel phenotypes. Many of the most striking examples of such mutations involve the mis expression of genes that control developmental pathways. Two striking gain-of function mutations in plants are Knl in maize (Veit B, et al. Genetics 125:623-31, 1990) and LECI in Arabidopsis, (Lotan T, et al. Cell 26:1195-205. 1998), which respectively display grossly modified leaf morphology, and the ectopic formation of embryo-like structures in adult tissues. Many new and useful traits that have arisen during evolution are also likely the result of the changed expression of specific genes. To this end, certain gain-of-function mutations in plants might be predicted to have WO 01/21781 PCT/USOO/25778 20 agriculturally useful phenotypes. For example, the inappropriate expression of a rice gene that controls panicle initiation in the vegetative meristem of the node could lead to plants having increased panicle number. From the teachings of the present invention, two significant outcomes are the functional characterization of a very large number of rice genes and the generation of lines having novel and useful phenotypes. Many genes do not have any apparent loss-of-function phenotype, and thus can only be identified through gain-of function mutations. Such mutations can also lead to novel traits. Thus, a gain-of function mutagenesis strategy is vital to comprehensively discover new genes. To this end, the Ds element on the TA ET/Ds vector (Figure 1) leads to the forced expression, via Gal4/VP16 transactivation, of genomic sequences into which it has transposed. To generate gain-of-function mutations, this element is launched in specified GAL4/VP16 pattern lines. Transposition of this element upstream of a gene will lead to the forced mis-expression of the gene in all cells that express the transactivator protein. Specifically, the Ds element carries a Gal4-responsive UAS recognition sequence and a minimal promoter adjacent to its rightward inverted repeat and oriented to read outward into the genomic region following transposition (Figure 1). In one aspect of the invention the UAS recognition sequence that is operably linked to a minimal promoter is in a separate parental transgenic plant line. An additional consideration that can be made is the selection of the pattern line in which the Ds element will be launched. Selection criteria will depend on the experimental goals; for example, if novel root phenotypes are sought, then a root pattern line would be used. While judicious choice of the pattern line may facilitate experimental success, the novel phenotypes generated with this approach can not be conceived a priori. Therefore, this strategy has the potential to generate useful alleles/phenotypes that are not accessible via any other approach. 3. Reverse genetics approach As in most genome sequence efforts, a large number of rice genes have been partially sequenced by the Rice Genome Research Project (over 12 million base WO 01/21781 PCT/USOO/25778 21 pairs have been sequenced) and the International Consortium, but only a small number of them have been assigned biological function. For those genes with significant homology to well-characterized genes, possible function may be inferred, but the majority either do not exhibit significant sequence identity or the similarities do not impart functional clues. Because gene-targeting methods that rely on homologous recombination are not well established for plants, targeted gene disruption is not an option to relate DNA sequence information to biological function. The mutagenized population generated within the present invention, however, can be used to identify lines that carry insertions within specific desired sequences. The strategy is amplification-based and relies on the availability of a mutagenized population for which any given gene has a high probability of carrying at least one transposon insertion. In this method, a DNA sequence is first selected for which the identification of the loss-of function phenotype is desired, and corresponding PCR primers are generated. Then DNA from all the transposon-tagged lines is pooled and used as a target for PCR reactions employing a combination of primers to the ends of the insertion mutagen (in our invention either T-DNA or Ds terminal sequences) and the gene of interest. Amplification of a product indicates that the population contains a line with an insertion in the selected sequence. By subdividing pools in successive rounds of PCR, a single line that carries an insertion within the sequence of interest is identified. Plants of this line are then scrutinized for mutant phenotype. Since the present mutations will consist of single copy, transposon elements and single copy T-DNA inserts, phenotype can be directly related to the mutated DNA sequence The following examples are offered by way of illustration, and not by way of limitation.

WO 01/21781 PCT/USOO/25778 22 EXAMPLES EXAMPLE 1 CONSTRUCTION OF TRANSFORMATION VECTORS In this example, two independent plant transformation vectors are described: one is an enhancer trap vector and the other one is UAS mutagenic vector. A number of enhancer trapping vectors are available using various reporter genes. The main component of the vector is transcriptional activator (transactivator) under the control of minimal promoter (supporting only basal transcription level). One transactivator is a GAL4/VP16 fusion protein with the DNA binding domain derived from the yeast GAL4 protein and the activation domain derived from the VP16 protein of Herpes Simplex Virus. The transactivator activates the reporter gene by binding to the UAS element containing the recognition sequence for the DNA binding domain of the transactivator. Among the reporter genes used, GUS, GFP and BoGUS, BoGUS is the preferred reporter molecule. Enhancer trapping vectors that are available are: pSK66.1 (GUS reporter); pFX (EGFP reporter); pFX (GUS/EGFP fusion reporter); pFX (BoGUS/EGFP fusion reporter); and pSR (BoGUS reporter). The second vector comprises the DNA binding domain recognition sequence (UAS) of the transactivator. Preferably UAS sequences are included on the transposable element (e.g. Ds element of maize) or in the vicinity of the Right Border of T-DNA when UAS mutagenesis is performed using Agrobacterium transformation. When transposon mutagenesis is used, plants with UAS elements on the transposable element are crossed with line expressing transposase, to allow transposition of UAS element into new location in the genome. In this type of mutagenesis, an excision marker is often used to identify (usually in the F2 generation of the cross) the plants with transposon excision from its original location. Chloramphenicol acetyl transferase can be used as an excision marker in plants.

WO 01/21781 PCT/USOO/25778 23 To generate novel phenotypes through gene activation in the native location of a gene, plants with tissue-specific expression of transactivator are crossed with plants mutagenized with UAS element. Appropriate screen is carried in the progeny of these crosses to identify plants with useful characters. An alternative method uses a single vector for enhancer trapping and UAS mutagenesis. This vector has a composition similar to the one exhibited in the Figure 1. This vector is used for Agrobacterium transformation and upon insertion of T-DNA into plant genomic DNA minimal promoter of transactivator is up-regulated through the activity of genomic enhancers that are in the vicinity of insertion site. After selecting enhancer trap lines with interesting pattern of transactivator activity, this line is crossed with transposase expressing line and the plants with UAS element transposed into new location identified through the use of excision marker selection and/or screening. EXAMPLE 2 AGROBACTERIUM-MEDIATED TRANSFORMATION OF RICE Plant material: Surface sterilized rice seeds are grown on 2N6 medium containing auxin (2,4-D) in darkness at 26"C for three weeks (21 d) to form calluses. Scutellum-derived calli obtained from these seeds are used for transformation. Agrobacterium strains: Agrobacterium vir helper strains LBA4404, EHA105, and AGL-1 that harbor pCAMBIA vectors are used for transformation. LBA4404 carries a vir helper Ti plasmid, pAL4404, derived from the octopine Ti plasmid pTiAch5 (Ooms et al. Plasmid 7, 15-19, 1982). The vir helper Ti plasmids in strains EHA105 (Hood et al., Transgenic Res. 2, 208-218, 1993) and AGL-l (Lazo et al., Bio/Technology 9, 963-967, 1991) are derived from leucinopine type supervirulent Ti plasmid pTiBo542. Protocol: Day 1: After three weeks of callusing, scutellum-derived calli are subdivided into 4 to 8 mm diameter pieces and placed on plates containing 2N6 medium and incubated at 26*C in the dark for four to seven days.

WO 01/21781 PCT/US00/25778 24 Day 3: Agrobacterium strains are streaked on AB medium with appropriate antibiotics and incubated at 28-29"C for two days. At this time, Agrobacterium forms a lawn on the plates. Day 5: Agrobacterium strains are resuspended in AAM medium containing 100 ptM acetosyringone by scraping the bacteria from plates with an inoculation loop, shaking vigorously for a minute, and incubated for 3 h. The OD of the bacterial suspension is measured at 600 nm, and approximately 1.0 OD of bacteria are used for transformation. 20 mL of the bacterial suspension is transferred into a petri dish or other suitable sterile container. Four to seven-day incubated calli are added to the bacterial suspension, swirled and left for 30 min. The calli are then blotted dry on sterile Whatman No. 1 filter papers and transferred to 2N6-AS plates. These calli and are then co-cultivated for two days in the dark 26*C. Day 7: Calli co-cultivated with Agrobacterium for two days are washed with water containing 250 mg/L cefotaxime to remove the bacteria by transferring the calli to plates containing 25 ml of water supplemented with 250 mg/L cefotaxime, swirling, and incubating for 1 h. During this period most of the bacteria are released from the calli. The calli are blotted dry on sterile Whatman No. 1 filter paper and then transferred to 2N6-CH plates containing cefotaxime at 250 mg/L (to kill Agrobacterium left attached to the calli) and hygromycin at 50 mg/L (to select transgenic calli). Calli are incubated in the dark at 26'C. Transient GUS expression is tested by staining a few washed calli with X-gluc (5-Bromo-4-chloro-3-indolyl p-D glucuronide). The calli are transferred to fresh selection medium once every two weeks. Small, transgenic hygromycin resistant calli start proliferating after four weeks of selection on hygromycin. Such proliferated calli are sub-cultured and independent proliferating lines are made. These sub-cultured calli further proliferate within two weeks and are transferred onto regeneration medium and cultured in the dark for one week. After a week, the calli are transferred to light. 5-10 days later calli start turning green and in 2-3 weeks time shoots start differentiating. These shoots are then WO 01/21781 PCT/USOO/25778 25 transferred onto rooting medium, and once roots are formed, plants are hardened and transferred to the glass house. Efficiency of rice transformation by A. tumefaciens EHA 105 (average of 5 experiments) Cultivar Inoculated calli % HygR callus % HygR plants lines Million 224 96.7 81.2 Nipponbare 231 98.6 j 76.3 WO 01/21781 PCT/USOO/25778 26 2N6 medium (1 L) pH 5.8 Na 2 MoO 4 . H 2 0 0.025g N6 salts (lOX) lOOmL H 3 BO 3 0.30g Sucrose 30g ZnSO 4 .7 H 1 0 0.20g 2,4-D (1mg/mL) 2mL CuSO 4 .5 H20 0.0039g Chu's vitamins (1OOX) lOmL CoCl,.6 H,0 0.0025g Casamino acids lg KI 0.075g Glutamine 0.5g Proline 0.5g AAM -MS vitamins (IOOX) (100 mL) Phytagel 2.5g Nicotinic acid 0.005g Thiamine.HCl O.OOlg AB Medium (500 mL) pH 7 Pyroxidine.HCI 0.005g AB salts 25mL Myoinositol 1.OOg 2.5 g glucose 2.5 g AB buffer 25mL AAM media. (1 L) pH 5.2 agar 7.5g AA macro (lOX) 1OOmL H20 450mL AA micro (1000X) 1mL AA iron (1 OOX) 1OmL 20xAB salts (200mL) AA amino acids (1 OOX) 1OmL NH4CZ 4 g AAM-MS vitamins (!OOX) 1OmL KCl 0.6g Casamino acid 0.5g CaCl 2 .2H 2 0 0.6g Sucrose 68.5g MgSO 4 .7H,O 5.Og Glucose 35.Og FeSO4.7H20 10mg 2N6-AS (1 L) pH 5.2 20XAB Buffer (200 mL) pH 7 N6 salts (1OX) 1OOmL K2HPO4.3H20 12 g Sucrose 30g NaH2PO4 4 g 2,4-D (lmg/mL) 2mL Chu's vitamins 1OmL A A macro (10X) (1 L) (100X ) In order of addition: Casamino acids lg KCl 29.5g Glucose log CaCl,.2H,0 1.5g Phytagel 2.5g MgSO 4 .7HO 5.Og Acetosyringone (100 1mL NaHPO 4 .2H 2 0 1.7g pM) pH to about 6 to dissolve AA amino acids (100X) (100 mL) Glutamine 8.76g Arginine 1.74g Glycine 0.075g Autoclave to dissolve. AA iron (1 00X) Protect from light. 100 mL EDTA-ferric sodium 0.25g AA micro (IOOOX) (100 mL) MnSO 4 .4 H.0 1.og WO 01/21781 PCT/USOO/25778 27 EXAMPLE 3 PHYSICAL MAPPING OF T-DNA AND Ds ELEMENT INSERTION SITES Three methods are used to isolate DNA fragments adjacent to DNA tag (T-DNA and Ds transposon) insertions into the plant (e.g., rice) genome: TAIL-PCR, ST-PCR and plasmid rescue. TAIL-PCR: This strategy, called thermal asymmetric interlaced (TAIL) PCR, utilizes nested sequence-specific primers together with a shorter arbitrary degenerate primer so that the relative amplification efficiencies of specific and nonspecific products can be thermally controlled. One low-stringency amplification cycle is carried out to create annealing site(s) adapted for the arbitrary primer within the unknown target sequence bordering the known segment. This sequence is then preferentially and geometrically amplified over non-target ones by interspersion of high-stringency amplification cycles with reduced-stringency cycles. The procedure is described in Liu et al. (Plant J 8: 457-463, 1995). The three degenerate TAIL-PCR oligos that are described in Liu et al (1995) are used in combination with DNA-tag specific primers. Degenerate primers are as follows: AD1 : 5' NTCGA(G/C)T(A/T)T(G/C)G(A/T)GTT 3' (SEQ ID No: 1) AD2: 5' NGTCGA(G/C)(A/T)GANA(A/T)GAA 3' (SEQ ID No: 2) AD3 : 5' (A/T)GTGNAG(A/T)ANCANAGA 3' (SEQ ID No: 3) T-DNA primers and Ds primers are selected from T-DNA right and left border sequences or Ds terminal repeat sequences. Three rounds of amplification are performed essentially as described. Positive amplification results from the tertiary reaction are checked by repeating the tertiary TAIL PCR with each oligonucleotide individually as well as with each DNA-tag/ AD oligonucleotide pairs. This ensures that the amplified products are specific for both oligonucleotides. Amplified products from tertiary amplifications are then cloned or directly subjected to DNA sequence analysis. Cloned inserts/amplification products are used as DNA probes against arrayed rice BAC libraries or YAC libraries to identify the physical location of the DNA insertion.

WO 01/21781 PCT/USOO/25778 28 EXAMPLE 4 SCREENING OF MUTANT POPULATIONS Generation of rice plants with increased cold tolerance at the flowering stage. A line with transactivator expression in the panicle during male meiosis stage is identified in the pattern line population. After crossing with transposase expressing line, progeny are selected (usually in the F2 generation of the cross) that has UAS element transposed into rice genome. These lines are identified through the activity of the excision marker (resistance to chloramphenicol treatment, for example). Chloramphenicol resistant lines are grown in the glasshouse and, during male meiosis stage, are subjected to cold stress (14-15 C). Plants with seed set significantly better than control (untransformed) plants are selected and further tested under the field conditions for increased resistance to cold treatment. Generation of rice plants with improved root system development. A line with expression of the transactivator in the root is identified in the pattern line population. After crossing with a transposase expressing line, progeny are selected (usually in the F2 generation of the cross) that have the UAS element transposed into rice genome. These lines are identified through the activity or the excision marker (resistance to chloramphenicol treatment, for example). Chloramphenicol resistant lines are grown in agar medium and development of the root system is compared by visual inspection with control plants. Plants with seed set significantly larger root system than control (untransformed) plants are selected and further tested under the field conditions for increased yielding potential. Generation of rice plants with additional panicles developing from lower nodes. A line with expression of the transactivator in the node tissue (especially nodes III and IV) is identified in the pattern line population. After crossing with transposase expressing line, a progeny is selected (usually in the F2 generation of the cross) that has UAS element transposed into rice genome. These lines are identified through activity ot the excision marker (resistance to chloramphenicol treatment, for example). Chloramphenicol resistant lines are grown in the glasshouse, and plants with panicles WO 01/21781 PCT/USOO/25778 29 initiating from nodes III and IV are identified. These plants are further analyzed under field conditions for increased yield as compared to control (untransformed) plants. EXAMPLE 5 TESTING OF UAS-REPORTER GENE TRANSACTIVATION IN RICE CALLUS Transactivation in rice is tested using 3 different GAL4/VP 16 constructs in an enhancer trap vector: (1) a full length GAL4NP16 transactivator; (2) a full length GAL4/VP16 transactivator with an intron of castor bean catalase (CAT1) inserted into the 5' untranslated region; and (3) a GAL4NP16 transactivator with a deletion of 36 amino acids from the DNA binding domain (DBDdel). An enhancer trap vector with a UAS-GUS reporter is used in Agrobacterium transformation of rice calli. After two weeks of co-cultivation, expression of the reporter is analyzed through histochemical staining with X-Gluc (5 Bromo-4-chloro-3-indolyl p-D glucuronide). Foci with GUS expression are counted and results compared among three constructs. In this test method, reporter gene expression in UAS-reporter gene transgenomic plants is demonstrated as dependent upon expression of the transactivator. When deletion variants of the transactivator DNA binding domain (DBDdel) are used, reporter gene expression is absent, whereas use of a full-length transactivator was very effective in inducing expression of the reporter. Exp. 1 Exp. 2 GAL4NP16 1020* 470 GAL4NP16 + Cat intron 220 235 GAL4/VP16 DBDdel 0 0 * number of GUS+ foci in at least 50 calli stained Similarly, an enhancer trap vector with UAS-EGFP reporter is used in Agrobacterium transformation of rice calli. After two weeks of co-cultivation, expression of the reporter is analyzed through fluorescent microscopy. Calli with EGFP expression are counted and results compared among three constructs.

WO 01/21781 PCT/USOO/25778 30 UAS-EGFP expression in enhancer trap constructs Exp. 1 Exp. 2 GAL4/VP16 10 57 GAL4NP16+Cat intron 6 33 GAL4NP16 DBDdel 0 2 (% of calli with EGFP expression in a sample of at least 100 calli) In addition, reporter gene expression results from co-transformation of a transactivator construct and a UAS-reporter gene construct. Finally, re-transformation of a rice line without expression of a reporter gene from UAS-promoter with 35S CaMV transactivator construct resulted in high level reporter gene expression. The number of T-DNA insertions is analyzed by Southern blot hybridization using nucleic acid encoding GAL4/VP16 as a probe. DNA is extracted from young leaves (1-2 weeks after transferred to the glass house). Some 251 Trans activator Facilitated Enhancer Trap (TAFET) lines with GUS and GUSPlus T M reporter genes are analyzed for T-DNA insertion. The number of T-DNA insertions varies from 1 to 7 but about 48.6% of lines (122) have a single insertion. The data on the population with GUS as a reporter gene are summarized to provide a sense of the high efficiency of the system to trap specific patterns of transactivator/reporter gene expression. Out of first 300 lines analyzed, we found 36 lines with root expression falling into 12 distinct pattern classes with expression in pericycle, vascular bundles, apical meristem, cap or root hair or a specific combination of the above. In shoot/leaf tissues we identified 19 patterns with expression in shoot base, node, coleoptile, leaf blade, collar, ligule or auricle, or a combination of the above. Analysis of floral tissues yielded 55 GUS expressing lines that could be grouped into 32 "pattern" classes with expression in palea, lemma, lodicule, anther filament, anther sac, pollen style or stigma, or a combination of the above. In general, we are getting a wide range of interesting expression patterns in practically all tissues/organs of TAFET lines. Over three thousand TAFET lines containing one of five different reporter genes: GUS, BoGUS, EGFP, GUS/EGFP fusion and BoGUS/EGFP fusion are constructed. In addition, over 1000 transgenic lines are developed for a variety of component testing experiments.

WO 01/21781 PCT/USOO/25778 31 EXAMPLE 6 TESTING OF UAS-REPORTER GENE TRANSACTIVATION IN ARABIDOPSIS Two Arabidopsis lines, which have different Gal4-VP16 expression patterns as determined by GFP expression, have been re-transformed with an UAS-GUS construct. All transformants that showed GUS expression have an identical expression pattern to GFP expression. Similarly, transformants with only UAS-GUS are crossed with a Gal4-VP16 pattern line and show an identical pattern of GUS expression. Currently, our Arabidopsis collection contains several hundred TAFET lines and over 1000 UAS-insertion lines. Furthermore, one Arabidopsis line, which has broad floral expression of Gal4-VP16, has been re-transformed with a double UAS structure designed to express neighboring genomic DNA when in the presence of Gal4-VP16. About 5% of the transformants show reduced fertility and/or changed floral morphology (size, shape). These mutants are being further analyzed to verify the linkage with Gal-VP16 transactivation. From the foregoing it will be appreciated that, although specific embodiments of the invention have been described herein for purposes of illustration, various modifications may be made without deviating from the spirit and scope of the invention. Accordingly, the invention is not limited except as by the appended claims.

Claims (23)

1. A method for generating a transgenic plant, comprising the steps of: (a) cross fertilizing two transgenomic plant lines to produce seed, wherein the first transgenomic plant line contains an introduced first nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule under control of a minimal promoter, and the second transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous site-specific nucleic acid effector molecule; and (b) growing the seed to produce a plant; wherein the introduced first nucleic acid molecule is operably linked to an endogenous enhancer sequence.

2. A method for generating seed of a transgenic plant, comprising the step of cross fertilizing two transgenomic plant lines to produce seed, wherein the first transgenomic plant line contains an introduced first nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule under control of a minimal promoter, and the second transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous site-specific nucleic acid effector molecule; wherein the introduced first nucleic acid molecule is operably linked to an endogenous enhancer sequence.

3. A method for generating a transgenic plant, comprising the steps of: (a) crossing three transgenomic plant lines to produce seed, wherein the first transgenomic plant line expresses a non-native transcriptional activator under control of a minimal promoter from a first introduced nucleic acid molecule, the second WO 01/21781 PCT/USOO/25778 33 transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous transcriptional activator and expresses a site-specific nucleic acid effector molecule; and the third transgenic plant line contains a third introduced nucleic acid molecule containing a binding site for the heterologous transcriptional activator and a binding site for the site-specific nucleic acid effector molecule; and (b) growing the seed into a plant; wherein the first introduced nucleic acid molecule is located [functionally close] to an endogenous enhancer sequence.

4. The method of claims 1-3, wherein the nucleic acid effector molecule is a protein or a nucleic acid.

5. The method of claims 1-3, wherein the nucleic acid effector molecule is selected from the group consisting of transcriptional activator, methylation enzyme, repressor, gyrase, kinkase, topoisomerase, class I restriction enzyme, RNA molecule and DNA molecule.

6. The method of claims 1-3, wherein the plant is a Gramineae.

7. The method of claim 6, wherein the Gramineae is selected from the group consisting of wheat, maize, rice, barley, rye, oats and sugar cane.

8. The method of claims 1-3, wherein the plant is a Solanaceae.

9. The method of claims 1-3, wherein the plant is a Leguminosae.

10. The method of claim 9, wherein the Leguminosae is selected from the group consisting of beans, soybean, lentil, chickpea and peanut.

11. The method of claims 1-3, wherein the binding of the effector molecule to the binding site modifies gene activity. WO 01/21781 PCT/USOO/25778 34

12. The method of claim 11, wherein the modified gene activity is an increase or a decrease in gene expression, a change in timing of gene expression or a change in cell type gene expression.

13. The method of claims 1-3, wherein the binding of the effector molecule to the binding site modifies nucleic acid structure.

14. The method of claims 1-3, wherein the plant exhibits an altered phenotype.

15. The method of claims 1-3, wherein the second nucleic acid molecule is a natural transposon.

16. A transgenic plant, wherein the plant is grown from seed produced by cross fertilizing two transgenomic plant lines, wherein the first transgenomic plant line contains an introduced first nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule operably linked to a minimal promoter, and the second transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous site-specific nucleic acid effector molecule; and wherein the binding of the effector molecule to the binding site modifiies gene expression or nucleic acid structure in a binding site-associated manner.

17. Seed from a transgenic plant produced by cross fertilizing two transgenomic plant lines, wherein the first transgenomic plant line contains an introduced first nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule operably linked to a minimal promoter, and the second transgenomic plant line contains a second introduced nucleic acid molecule containing a binding site for the heterologous site-specific nucleic acid effector molecule; and WO 01/21781 PCTIUSOO/25778 35 wherein the binding of the effector molecule to the binding site modifiies gene expression or nucleic acid structure in a binding site-associated manner.

18. A method for modifying gene activity in a plant, comprising the steps of: (a) generating a transformed plant line with an introduced nucleic acid molecule that expresses a non-native site-specific nucleic acid effector molecule under control of a minimal promoter and a transposable element containing a minimal promoter operably linked to at least one binding site for the nucleic acid effector molecule; (b) cross-fertilizing the transformed plant line with a second plant line that expresses a transposase to produce seed; and (c) growing the seed into a plant.

19. The method of claim 18, further comprising selecting a plant that displays a desired phenotype.

20. The method of claim 18, wherein the transposable element is Ds and the transposase is Ac.

21. The method of claim 18, wherein the introduced nucleic acid molecule further comprises a reporter gene operably linked to at least one binding site for the nucleic acid effector molecule.

22. The method of claim 18, wherein the site-specific nucleic acid effector molecule is a transactivator.

23. The method of claim 22, wherein the transactivator is a fusion protein of a DNA-binding domain of a beta-glucuronidase repressor or Gal4 and VP 16.