EP1673462A2 - Plant transcriptional regulators of abiotic stress - Google Patents

Abstract

The invention relates to transgenic plants and methods of using polynucleotides and polypeptides to produce transgenic plants having advantageous properties, including improved cold, heat and osmotic stress tolerance, as compared to wild-type plants. The invention also pertains to expression systems that regulate these transcription factor polynucleotides, providing constitutive, transient, inducible and tissue-specific regulation.

Description

PLANT TRANSCRIPTIONAL REGULATORS OF ABIOTIC STRESS

RELATIONSHIP TO COPENDING APPLICATIONS This application claims priority from commonly assigned U.S. Provisional Application No. 60/565,948, filed April 26, 2004; U.S. non-provisional Patent Application No. 10/714,887, filed

November 13, 2003; U.S. non-provisional Patent Application No. 10/685,922, filed October 14, 2003; and U.S. non-provisional Patent Application No. 10/838,616, filed May 4, 2004. The entire contents of these applications are hereby incorporated by reference. FIELD OF THE INVENTION The present invention relates to compositions and methods for modifying a plant phenotypically, said plant having an altered response to abiotic stresses such as cold, heat and osmotic stresses.

BACKGROUND OF THE INVENTION A plant's traits may be controlled through a number of cellular processes. One important way to manipulate that control is through transcription factors - proteins that influence the expression of a particular gene or sets of genes. Because transcription factors are key controlling elements of biological pathways, altering the expression levels of one or more transcription factors can change entire biological pathways in an organism. This altered expression can result in plants and crops with new and/or improved commercially valuable properties, including improved yield under non-stressed conditions, or survival and yield during periods of abiotic stress. Problems associated with drought. A drought is a period of abnormally dry weather that persists long enough to produce a serious hydrologic imbalance (for example crop damage, water supply shortage, etc.). Drought is the primary weather-related problem in agriculture and also ranks as one of the major natural disasters of all time, causing not only economic damage, but also loss of human lives. Problems for plants caused by low water availability include mechanical stresses caused by the withdrawal of cellular water. Drought also causes plants to become more susceptible to various diseases (Simpson (1981) in Water Stress on Plants. (Simpson, G. M., ed.), Praeger, NY, pp. 235-265). The most important factor in drought resistance is the ability of the plant to maintain high water status and turgidity while maintaining carbon fixation. Various adaptive mechanisms influence this ability, including increasing root surface area or depth, osmotic adjustment, and the accumulation of hydrophilic proteins. Abscisic acid (ABA) is also an essential regulatory component of many of these protective features. In addition to the many land regions of the world that are too arid for most, if not all, crop plants, overuse and over-utilization of available water is resulting in an increasing loss of agriculturally-usable land, a process which, in the extreme, results in desertification. The problem is further compounded by increasing salt accumulation in soils, which adds to the loss of available water in soils. Problems associated with high salt levels. One in five hectares of irrigated land is damaged by salt, an important historical factor in the decline of ancient agrarian societies. This condition is expected to worsen, further reducing the availability of arable land and crop production, since none of the top five food crops - wheat, corn, rice, potatoes, and soybean - can tolerate excessive salt. Detrimental effects of salt on plants are a consequence of both water deficit resulting in osmotic stress (similar to drought stress) and the effects of excess sodium ions on critical biochemical processes. As with freezing and drought, high saline causes water deficit. The presence of high salt makes it difficult for plant roots to extract water from their environment (Buchanan et al. (2000) in Biochemistry and Molecular Biology of Plants, American Society of Plant Physiologists, Rockville, MD). Soil salinity is thus one of the more important variables that determines where a plant may thrive. Salt tolerance is of particular importance early in a plant's lifecycle, since evaporation from the soil surface causes upward water movement, and salt accumulates in the upper soil layer where the seeds are placed. Thus, germination normally takes place at a salt concentration much higher than the mean salt level in the whole soil profile. Problems associated with excessive heat. Germination of many crops is highly temperature- sensitive. Seedlings and mature plants that are exposed to excess heat may experience heat shock, which may arise in various organs including leaves and particularly fruit, when transpiration is insufficient to overcome heat stress. Heat also damages cellular structures, including organelles and cytoskeleton, and impairs membrane function (Buchanan et al. (2000) supra). Heat shock may produce a decrease in overall protein synthesis, accompanied by expression of heat shock proteins. Heat shock proteins function as chaperones and are involved in refolding proteins denatured by heat. Heat stress often accompanies conditions of low water availability. Heat itself is seen as an interacting stress and adds to the detrimental effects caused by water deficit conditions. Evaporative demand exhibits near exponential increases with increases in daytime temperatures, and can result in high transpiration rates and low plant water potentials (Hall et al. (2000) Plant Physiol. 123 : 1449- 1458). High-temperature damage to pollen almost always occurs in conjunction with drought stress, and rarely occurs under well-watered conditions. A transcription factor that would enhance germination in hot conditions would also be useful for crops that are planted late in the season or in hot climates. Problems associated with excessive cold or chilling conditions. The term "chilling sensitivity" has been used to describe many types of physiological damage produced at low, but above freezing, temperatures. Most crops of tropical origins such as soybean, rice, maize and cotton are easily damaged by chilling. Typical cold damage includes wilting, necrosis, chlorosis or leakage of ions from cell membranes. The underlying mechanisms of chilling sensitivity probably involve the level of membrane saturation and other physiological deficiencies. For example, photoinhibition of photosynthesis (disruption of photosynthesis due to high light intensities) often occurs under clear atmospheric conditions subsequent to cold late summer/autumn nights. Chilling may lead to yield losses and lower product quality through delayed ripening. Another consequence of poor growth is the rather poor ground cover in spring, often resulting in soil erosion, increased occurrence of weeds, and reduced uptake of nutrients. A retarded uptake of mineral nitrogen could also lead to increased losses of nitrate into ground water. By some estimates, chilling accounts for monetary losses behind only to drought and flooding. Desirability of altered sugar sensing. Sugars are key regulatory molecules that affect diverse processes in higher plants including germination, growth, flowering, senescence, sugar metabolism and photosynthesis. Sucrose, for example, is the major transport form of photosynthate and its flux through cells has been shown to affect gene expression and alter storage compound accumulation in seeds (source-sink relationships). Glucose-specific hexose-sensing has also been described in plants and is implicated in cell division and repression of "famine" genes (photosynthetic or glyoxylate cycles). Water deficit is a common component of many plant stresses. Water deficit occurs in plant cells when the whole plant transpiration rate exceeds the water uptake. In addition to drought, other stresses such as salinity and low temperature produce cellular dehydration (McCue and Hanson (1990) Trends Biotechnol. 8: 358-362). Salt and drought stress signal transduction consist of ionic and osmotic homeostasis signaling pathways. The ionic aspect of salt stress is signaled via the SOS pathway where a calcium-responsive SOS3-SOS2 protein kinase complex controls the expression and activity of ion transporters such as SOS1. The pathway regulating ion homeostasis in response to salt stress has been reviewed recently by Xiong and Zhu (Xiong and Zhu (2002) Plant Cell Environ. 25: 131-139). The osmotic component of salt stress involves complex plant reactions that overlap with drought and/or cold stress responses. Common aspects of drought, cold and salt stress response have been reviewed recently by Xiong and Zhu (2002) supra. Those include: (a) transient changes in the cytoplasmic calcium levels very early in the signaling event (Knight, (2000) Int. Rev. Cytol. 195: 269-324; Sanders et al. (1999) Plant Cell 11 : 691 -706); (b) signal transduction via mitogen-activated and/or calcium dependent protein kinases (CDPKs; Xiong and Zhu (2002) supra) and protein phosphatases (Merlot et al. (2001) Plant J. 25: 295- 303; Tahtiharju and Palva (2001) Plant J. 26: 461-470); (c) increases in ABA levels in response to stress triggering a subset of responses (Xiong and Zhu (2002) supra, and references therein); (d) inositol phosphates as signal molecules (at least for a subset of the stress responsive transcriptional changes (Xiong et al. (2001) Genes Dev. 15: 1971-1984)); (e) activation of phospholipases which in turn generate a diverse array of second messenger molecules, some of which might regulate the activity of stress responsive kinases (phospholipase D functions in an ABA independent pathway, Frank et al. (2000) Plant Cell 12: 111-124); (f) induction of late embryogenesis abundant (LEA) type genes including the CRT/DRE-containing COR/RD genes (Xiong and Zhu (2002) supra); (g) increased levels of antioxidants and compatible osmolytes such as proline and soluble sugars (Hasegawa et al. (2000) Annu. Rev. Plant Mol. Plant Physiol. 51: 463-499); (h) accumulation of reactive oxygen species such as superoxide, hydrogen peroxide, and hydroxyl radicals (Hasegawa et al. (2000) supra).

ABA biosynthesis is regulated by osmotic stress at multiple steps. Both ABA-dependent and ABA-independent osmotic stress signaling first modify constitutively expressed transcription factors, leading to the expression of early response transcriptional activators, which then activate downstream stress tolerance effector genes. Based on the commonality of many aspects of cold, drought and salt stress responses, it can be concluded that genes that increase tolerance to cold or salt stress can also improve drought stress protection. In fact, this has already been demonstrated for transcription factors (in the case of AtCBF/DREBl) and for other genes such as OsCDPK7 (Saijo et al. (2000) Plant J. 23: 319-327), or AVP1 (a vacuolar pyrophosphatase-proton-pump; Gaxiola et al. (2001) Proc. Natl. Acad. Sci. USA 98: 11444-11449). We have identified polynucleotides encoding transcription factors in the G867 clade of transcription factor polypeptides, developed transgenic plants using these polynucleotides, and have identified plants tolerant to abiotic stresses. In so doing, we have identified important polynucleotide and polypeptide sequences for producing commercially valuable plants and crops as well as the methods for making them and using them. Other aspects and embodiments of the invention are described below and can be derived from the teachings of this disclosure as a whole.

SUMMARY OF THE INVENTION The invention pertains to a method for increasing a plant's tolerance to abiotic stress. This is accomplished by providing a vector, plasmid or other nucleic acid construct that contains a transcription factor polynucleotide and regulatory elements for transcriptional regulation of the polynucleotide. The polynucleotide is a sequence that encodes a member of the G867 clade of transcription factor polypeptides, which are derived from a common polypeptide ancestor (Figure 3) and which generally contain three subsequences the presence of which identifies the clade members. The G867 clade member sequences that have been successfully used to confer increased tolerance to abiotic stress derive from a number of diverse species, including dicots such as Arabidopsis and soy, and monocots including rice and corn. The subsequences found within the clade member polypeptides are an AP2 domain, a DML motif (exemplified by SEQ ID NO: 54, and having a sequence similar to the consensus sequence of SEQ ID NO: 55), and a B3 domain, each of which are further characterized in the specification below. These subsequences have been found to function in this order from N-terminal to C-terminal, respectively, although a different order may also be considered part of the invention. As noted, the vector, plasmid or nucleic acid construct also contains a regulatory element. This may be a constitutive, inducible or tissue- specific promoter that controls expression of the polynucleotide sequence. The vector, plasmid or nucleic acid construct is then introduced into a target plant, thus transforming the plant into one that has increased tolerance to abiotic stress relative to control plants. Inducible promoters may include, for example, the DREB2A and RD29A promoters. The RD29A promoter has been used to regulate expression of the G867 polynucleotide and confer increased abiotic stress tolerance. Examples of tissue- specific promoters that have been used in this fashion include the ARSK1 (root specific) promoter, the RBSC3 (leaf specific) promoter, the CUT1 (epidermal tissue-specific) promoter, and the SUC2 (vascular specific) promoter. The method also pertains to increasing a plant's tolerance to abiotic stress with a multiple vector approach. In this case, a first vector that comprises a promoter cloned in front of a LexA DNA binding domain fused to a GAL4 activation domain is introduced into the plant. A second vector is then introduced into the same plant; this second vector comprises a polynucleotide sequence encoding a G867 polypeptide clade member. The plant is then allowed to overexpress the G867 member polypeptide, which increases the plant's tolerance to abiotic stress. The promoter cloned in front of a LexA DNA binding domain may be, for example, the RD29A promoter, although other promoters that function in a similar capacity and which may be expressed in an inducible or tissue-specific manner are readily envisioned and also encompassed by the present invention. The methods encompassed by the invention may also be extended to propagation techniques used to generate plants. For example, a target plant that has been transformed with a polynucleotide encoding a G867 polypeptide clade member and which has increased abiotic stress tolerance (relative to a wild-type or non-transformed control) may be "selfed" (i.e., self-pollinated) or crossed with another plant to produce seed. Progeny plants may be grown from this seed, thus generating transformed progeny plants with increased tolerance to abiotic stress, as compared to non-transformed plants of the same species that do not overexpress the member of the G867 polypeptide clade member. Transgenic plants and seed from these plants produced by the methods of the invention are also encompassed by the invention. BRIEF DESCRIPTION OF THE SEQUENCE LISTING AND FIGURES The Sequence Listing provides exemplary polynucleotide and polypeptide sequences of the invention. The traits associated with the use of the sequences are included in the Examples. CD Copy 1 is a read-only memory computer-readable compact disc and contains a copy of the Sequence Listing in ASCII text format. The Sequence Listing is named "MBI0069PCT.ST25.txt" and is 131 kilobytes in size. The copies of the Sequence Listing on the CD-ROM disc are hereby incorporated by reference in their entirety. Figure 1 shows a conservative estimate of phylogenetic relationships among the orders of flowering plants (modified from Angiosperm Phylogeny Group (1998) Ann. Missouri Bot. Gard. 84: 1- 49). Those plants with a single cotyledon (monocots) are a monophyletic clade nested within at least two major lineages of dicots; the eudicots are further divided into rosids and asterids. Arabidopsis is a rosid eudicot classified within the order Brassicales; rice is a member of the monocot order Poales. Figure 1 was adapted from Daly et al. (2001) Plant Physiol. 127: 1328-1333. Figure 2 shows a phylogenic dendogram depicting phylogenetic relationships of higher plant taxa, including clades containing tomato and Arabidopsis; adapted from Ku et al. (2000) Proc. Natl. Acad. Sci. 97: 9121-9126; and Chase et al. (1993) Ann. Missouri Bot. Gard. 80: 528-580. Figure 3 depicts a phylogenetic tree of several members of the RAV family, identified through

BLAST analysis of proprietary (using corn, soy and rice genes) and public data sources (all plant species). This tree was generated as a Clustal X 1.81 alignment: MEGA2 tree, Maximum Parsimony, bootstrap consensus. Examples of clade members from a number of species are found within the box. Figures 4A-4J show an alignment of AP2 transcription factors from Arabidopsis, soybean, rice and corn, showing identical or similar residues (boxes) and AP2 domains, DML motifs, and B3 domains.

DESCRIPTION OF THE SPECIFIC EMBODIMENTS In an important aspect, the present invention relates to polynucleotides and polypeptides, for example, for modifying phenotypes of plants, particularly those associated with abiotic stress tolerance. Throughout this disclosure, various information sources are referred to and/or are specifically incorporated. The information sources include scientific journal articles, patent documents, textbooks, and World Wide Web browser-inactive page addresses. While the reference to these information sources clearly indicates that they can be used by one of skill in the art, each and every one of the information sources cited herein are specifically incorporated in their entirety, whether or not a specific mention of "incorporation by reference" is noted. The contents and teachings of each and every one of the information sources can be relied on and used to make and use embodiments of the invention. As used herein and in the appended claims, the singular forms "a," "an," and "the" include plural reference unless the context clearly dictates otherwise. Thus, for example, a reference to "a plant" includes a plurality of such plants, and a reference to "a stress" is a reference to one or more stresses and equivalents thereof known to those skilled in the art, and so forth.

DEFINITIONS "Nucleic acid molecule" refers to an oligonucleotide, polynucleotide or any fragment thereof. It may be DNA or RNA of genomic or synthetic origin, double-stranded or single-stranded, and combined with carbohydrate, lipids, protein, or other materials to perform a particular activity such as transformation or form a useful composition such as a peptide nucleic acid (PNA). "Polynucleotide" is a nucleic acid molecule comprising a plurality of polymerized nucleotides, e.g., at least about 15 consecutive polymerized nucleotides. A polynucleotide may be a nucleic acid, oligonucleotide, nucleotide, or any fragment thereof. In many instances, a polynucleotide comprises a nucleotide sequence encoding a polypeptide (or protein) or a domain or fragment thereof. Additionally, the polynucleotide may comprise a promoter, an intron, an enhancer region, a polyadenylation site, a translation initiation site, 5' or 3' untranslated regions, a reporter gene, a selectable marker, or the like. The polynucleotide can be single-stranded or double-stranded DNA or RNA. The polynucleotide optionally comprises modified bases or a modified backbone. The polynucleotide can be, e.g., genomic DNA or RNA, a transcript (such as an mRNA), a cDNA, a PCR product, a cloned DNA, a synthetic DNA or RNA, or the like. The polynucleotide can be combined with carbohydrate, lipids, protein, or other materials to perform a particular activity such as transformation or form a useful composition such as a peptide nucleic acid (PNA). The polynucleotide can comprise a sequence in either sense or antisense orientations. "Oligonucleotide" is substantially equivalent to the terms amplimer, primer, oligomer, element, target, and probe and is preferably single-stranded. "Gene" or "gene sequence" refers to the partial or complete coding sequence of a gene, its complement, and its 5' or 3' untranslated regions. A gene is also a functional unit of inheritance, and in physical terms is a particular segment or sequence of nucleotides along a molecule of DNA (or RNA, in the case of RNA viruses) involved in producing a polypeptide chain. The latter may be subjected to subsequent processing such as chemical modification and folding to obtain a functional protein or polypeptide. A gene may be isolated, partially isolated, or be found with an organism's genome. By way of example, a transcription factor gene encodes a transcription factor polypeptide, which may be functional or require processing to function as an initiator of transcription. Operationally, genes may be defined by the cis-trans test, a genetic test that determines whether two mutations occur in the same gene and that may be used to determine the limits of the genetically active unit (Rieger et al. (1976) Glossary of Genetics and Cytogenetics: Classical and Molecular. 4th ed., Springer Verlag. Berlin). A gene generally includes regions preceding ("leaders"; upstream) and following ("trailers"; downstream) the coding region. A gene may also include intervening, non-coding sequences, referred to as "introns", located between individual coding segments, referred to as "exons". Most genes have an associated promoter region, a regulatory sequence 5' of the transcription initiation codon (there are some genes that do not have an identifiable promoter). The function of a gene may also be regulated by enhancers, operators, and other regulatory elements. A "recombinant polynucleotide" is a polynucleotide that is not in its native state, e.g., the polynucleotide comprises a nucleotide sequence not found in nature, or the polynucleotide is in a context other than that in which it is naturally found, e.g., separated from nucleotide sequences with which it typically is in proximity in nature, or adjacent (or contiguous with) nucleotide sequences with which it typically is not in proximity. For example, the sequence at issue can be cloned into a vector, or otherwise recombined with one or more additional nucleic acid. An "isolated polynucleotide" is a polynucleotide, whether naturally occurring or recombinant, that is present outside the cell in which it is typically found in nature, whether purified or not. Optionally, an isolated polynucleotide is subject to one or more enrichment or purification procedures, e.g., cell lysis, extraction, centrifugation, precipitation, or the like. "Fragment", with respect to a polynucleotide, refers to a clone or any part of a polynucleotide molecule that retains a usable, functional characteristic. Useful fragments include oligonucleotides and polynucleotides that may be used in hybridization or amplification technologies or in the regulation of replication, transcription or translation. A "polynucleotide fragment" refers to any subsequence of a polynucleotide, typically at least about nine or more consecutive nucleotides of any of the sequences provided herein. Exemplary polynucleotide fragments are the first sixty consecutive nucleotides of the transcription factor polynucleotides listed in the Sequence Listing. Exemplary fragments also include fragments that comprise a region that encodes an AP2 domain, a DML motif or a B3 domain of a transcription factor. Fragments may also include subsequences of polypeptides and protein molecules, or a subsequence of the polypeptide. Fragments may have uses in that they may have antigenic potential. In some cases, the fragment or domain is a subsequence of the polypeptide which performs at least one biological function of the intact polypeptide in substantially the same manner, or to a similar extent, as does the intact polypeptide. For example, a polypeptide fragment can comprise a recognizable structural motif or functional domain such as a DNA-binding site or domain that binds to a DNA promoter region, an activation domain, or a domain for protein-protein interactions, and may initiate transcription. Fragments can vary in size from as few as 3 amino acid residues to the full length of the intact polypeptide, but are preferably at least about 30 amino acid residues in length. Exemplary polypeptide fragments are the first twenty consecutive amino acids of a mammalian protein encoded by are the first twenty consecutive amino acids of the transcription factor polypeptides listed in the Sequence Listing. Exemplary fragments also include fragments that comprise an AP2 binding or a B3 domain of a transcription factor, for example, amino acid residues 59-124 or amino acid residues 187-272 of G867 (SEQ ID NO: 2), as noted in Table 1. The invention also encompasses production of DNA sequences that encode transcription factors and transcription factor derivatives, or fragments thereof, entirely by synthetic chemistry. After production, the synthetic sequence may be inserted into any of the many available expression vectors and cell systems using reagents well known in the art. Moreover, synthetic chemistry may be used to introduce mutations into a sequence encoding transcription factors or any fragment thereof. A "polypeptide" is an amino acid sequence comprising a plurality of consecutive polymerized amino acid residues e.g., at least about 15 consecutive polymerized amino acid residues. In many instances, a polypeptide comprises a polymerized amino acid residue sequence that is a transcription factor or a domain or portion or fragment thereof. Additionally, the polypeptide may comprise (i) a localization domain; (ii) an activation domain; (iii) a repression domain; (iv) an oligomerization domain; or (v) a DNA-binding domain, or the like. The polypeptide optionally comprises modified amino acid residues, naturally occurring amino acid residues not encoded by a codon, or non-naturally occurring amino acid residues. "Protein" refers to an amino acid sequence, oligopeptide, peptide, polypeptide or portions thereof whether naturally occurring or synthetic. With respect to a polypeptide, "portion", as used herein refers to any part of a polypeptide used for any purpose, including the screening of a library of molecules that specifically bind to that portion or for the production of antibodies. A "recombinant polypeptide" is a polypeptide produced by translation of a recombinant polynucleotide. A "synthetic polypeptide" is a polypeptide created by consecutive polymerization of isolated amino acid residues using methods well known in the art. An "isolated polypeptide," whether a naturally occurring or a recombinant polypeptide, is more enriched in (or out of) a cell than the polypeptide in its natural state in a wild-type cell, e.g., more than about 5% enriched relative to wild type standardized at 100%. Such an enrichment is not the result of a natural response of a wild-type plant. Alternatively, or additionally, the isolated polypeptide is separated from other cellular components with which it is typically associated, e.g., by any of the various protein purification methods herein. "Homology" refers to sequence similarity between a reference sequence and at least a fragment of a newly sequenced clone insert or its encoded amino acid sequence. "Identity" or "similarity" refers to sequence similarity between two polynucleotide sequences or between two polypeptide sequences, with identity being a more strict comparison. The phrases "percent identity" and "% identity" refer to the percentage of sequence similarity found in a comparison of two or more polynucleotide sequences or two or more polypeptide sequences. "Sequence similarity" refers to the percent similarity in base pair sequence (as determined by any suitable method) between two or more polynucleotide sequences. Two or more sequences can be anywhere from 0-100% similar, or any integer value therebetween. Identity or similarity can be determined by comparing a position in each sequence that may be aligned for purposes of comparison. When a position in the compared sequence is occupied by the same nucleotide base or amino acid, then the molecules are identical at that position. A degree of similarity or identity between polynucleotide sequences is a function of the number of identical similar or matching nucleotides at positions shared by the polynucleotide sequences. A degree of identity of polypeptide sequences is a function of the number of identical amino acids at positions shared by the polypeptide sequences. A degree of homology or similarity of polypeptide sequences is a function of the number of amino acids at positions shared by the polypeptide sequences. With regard to polypeptides, the terms "substantial identity" or "substantially identical" may refer to sequences of sufficient similarity and structure to the transcription factors in the Sequence Listing to produce similar function when expressed or overexpressed in a plant; in the present invention, this function is increased tolerance to abiotic stress. Sequences that are at least about 80% identical, to the instant polypeptide sequences, including AP2 and B3 domain sequences, are considered to have "substantial identity" with the latter. Sequences having lesser degrees of identity but comparable biological activity are considered to be equivalents. The structure required to maintain proper functionality is related to the tertiary structure of the polypeptide. There are discreet domains and motifs within a transcription factor that must be present within the polypeptide to confer function and specificity. These specific structures are required so that interactive sequences will be properly oriented to retain the desired activity. "Substantial identity" may thus also be used with regard to subsequences, for example, motifs, that are of sufficient structure and similarity, being at least about 80% identical to similar motifs in other related sequences so that each confers or is required for increased tolerance to abiotic stress. "Alignment" refers to a number of nucleotide or amino acid residue sequences aligned by lengthwise comparison so that components in common (i.e., nucleotide bases or amino acid residues at corresponding positions) may be visually and readily identified. The fraction or percentage of components in common is related to the homology or identity between the sequences. Alignments such as those found in Figures 4A-4J may be used to identify AP2, DML and B3 domains and relatedness within these domains. An alignment may suitably be determined by means of computer programs known in the art, such as Mac Vector (1999) (Accelrys, Inc., San Diego, CA). A "conserved domain" or "conserved region" as used herein refers to a region in heterologous polynucleotide or polypeptide sequences where there is a relatively high degree of sequence identity between the distinct sequences. AP2 binding domains and B3 domains are examples of conserved domains. With respect to polynucleotides encoding presently disclosed transcription factors, a conserved domain is preferably at least 10 base pairs (bp) in length. A "conserved domain", with respect to presently disclosed polypeptides refers to a domain within a transcription factor family that exhibits a higher degree of sequence homology, such as at least 70% sequence similarity, including conservative substitutions, greater than about 70% identity, or at least about 79%, 81%, 86%, 87%, 89%, 91%, 95%, or 98% amino acid residue sequence identity of a polypeptide of consecutive amino acid residues. Sequences that encode for or possess conserved domains that recognizably fall within the G867 clade of transcription factor polypeptides and that have comparable biological activity to the present sequences are encompassed by the invention. A fragment or domain can be referred to as outside a conserved domain, outside a consensus sequence, or outside a consensus DNA- binding site that is known to exist or that exists for a particular transcription factor class, family, or subfamily. In this case, the fragment or domain will not include the exact amino acids of a consensus sequence or consensus DNA-binding site of a transcription factor class, family or sub-family, or the exact amino acids of a particular transcription factor consensus sequence or consensus DNA-binding site. Furthermore, a particular fragment, region, or domain of a polypeptide, or a polynucleotide encoding a polypeptide, can be "outside a conserved domain" if all the amino acids of the fragment, region, or domain fall outside of a defined conserved domain(s) for a polypeptide or protein. Sequences having lesser degrees of identity but comparable biological activity are considered to be equivalents. As one of ordinary skill in the art recognizes, conserved domains may be identified as regions or domains of identity to a specific consensus sequence ( for example, Riechmann et al. (2000) Science 290: 2105-2110). Thus, by using alignment methods well known in the art, the conserved domains (i.e., the AP2 or B3 domains or DML motifs) of the AP2 plant transcription factors (Riechmann and Meyerowitz (1998) Biol. Chem. 379:633-646) may be determined. The AP2 -binding and B3 (or conserved) domains for SEQ ID NO: 2, 4, 6, and 8 and orthologs from diverse species are listed in Table 1. Also, the polypeptides of Table 1 have AP2-binding and B3 domains specifically indicated by amino acid coordinates. A comparison of the regions of the polypeptides in Table 1 allows one of skill in the art to identify AP2 -binding and B3 domains for any of the polypeptides listed or referred to in this disclosure. "Complementary" refers to the natural hydrogen bonding by base pairing between purines and pyrimidines. For example, the sequence A-C-G-T (5' -> 3') forms hydrogen bonds with its complements A-C-G-T (5' -> 3') or A-C-G-U (5' -> 3'). Two single-stranded molecules may be considered partially complementary, if only some of the nucleotides bond, or "completely complementary" if all of the nucleotides bond. The degree of complementarity between nucleic acid strands affects the efficiency and strength of hybridization and amplification reactions. "Fully complementary" refers to the case where bonding occurs between every base pair and its complement in a pair of sequences, and the two sequences have the same number of nucleotides. The terms "highly stringent" or "highly stringent condition" refer to conditions that permit hybridization of DNA strands whose sequences are highly complementary, wherein these same conditions exclude hybridization of significantly mismatched DNAs. Polynucleotide sequences capable of hybridizing under stringent conditions with the polynucleotides of the present invention may be, for example, variants of the disclosed polynucleotide sequences, including allelic or splice variants, or sequences that encode orthologs or paralogs of presently disclosed polypeptides. Nucleic acid hybridization methods are disclosed in detail by Kashima et al. (1985) Nature 313:402-404; Sambrook et al. (1989) Molecular Cloning: A Laboratory Manual. 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. ("Sambrook"); and by Haymes et al., Nucleic Acid Hybridization: A Practical Approach. IRL Press, Washington, D.C. (1985), which references are incorporated herein by reference. In general, stringency is determined by the temperature, ionic strength, and concentration of denaturing agents (e.g., formamide) used in a hybridization and washing procedure (a more detailed description of establishing and determining stringency is presented below). The degree to which two nucleic acids hybridize under various conditions of stringency is correlated with the extent of their similarity. Thus, similar nucleic acid sequences from a variety of sources, such as within a plant's genome (as in the case of paralogs) or from another plant (as in the case of orthologs) that may perform similar functions can be isolated on the basis of their ability to hybridize with known transcription factor sequences. Numerous variations are possible in the conditions and means by which nucleic acid hybridization can be performed to isolate transcription factor sequences having similarity to transcription factor sequences known in the art and are not limited to those explicitly disclosed herein. Such an approach may be used to isolate polynucleotide sequences having various degrees of similarity with disclosed transcription factor sequences, such as, for example, transcription factors having 60% or more identity with disclosed transcription factors. Regarding the terms "paralog" and "ortholog", homologous polynucleotide sequences and homologous polypeptide sequences may be paralogs or orthologs of the claimed polynucleotide or polypeptide sequences. Orthologs and paralogs are evolutionarily-related genes that have similar sequence and similar functions. Orthologs are structurally-related genes in different species that are derived by a speciation event. Paralogs are structurally related genes within a single species that are derived by a duplication event. Sequences that are sufficiently similar to one another will be appreciated by those of skill in the art and may be based upon percentage identity of the complete sequences, percentage identity of a conserved domain or sequence within the complete sequence, percentage similarity to the complete sequence, percentage similarity to a conserved domain or sequence within the complete sequence, and/or an arrangement of contiguous nucleotides or peptides particular to a conserved domain or complete sequence. Sequences that are sufficiently similar to one another will also bind in a similar manner to the same DNA binding sites of transcriptional regulatory elements using methods well- known to those of skill in the art. The term "equivalog" describes members of a set of homologous proteins that are conserved with respect to function since their last common ancestor. Related proteins are grouped into equivalog families, and otherwise into protein families with other hierarchically defined homology types. This definition is provided at the Institute for Genomic Research (TIGR) world wide web (www) website, " tigr.org " under the heading "Terms associated with TIGRFAMs". The term "variant", as used herein, may refer to polynucleotides or polypeptides that differ from the presently disclosed polynucleotides or polypeptides, respectively, in sequence from each other, and as set forth below. With regard to polynucleotide variants, differences between presently disclosed polynucleotides and polynucleotide variants are limited so that the nucleotide sequences of the former and the latter are closely similar overall and, in many regions, identical. Due to the degeneracy of the genetic code, differences between the former and latter nucleotide sequences may be silent (i.e., the amino acids encoded by the polynucleotide are the same, and the variant polynucleotide sequence encodes the same amino acid sequence as the presently disclosed polynucleotide). Variant nucleotide sequences may encode different amino acid sequences, in which case such nucleotide differences will result in amino acid substitutions, additions, deletions, insertions, truncations or fusions with respect to the similar disclosed polynucleotide sequences. These variations may result in polynucleotide variants encoding polypeptides that share at least one functional characteristic. The degeneracy of the genetic code also dictates that many different variant polynucleotides can encode identical and/or substantially similar polypeptides in addition to those sequences illustrated in the Sequence Listing. Also within the scope of the invention is a variant of a transcription factor nucleic acid listed in the Sequence Listing, that is, one having a sequence that differs from the one of the polynucleotide sequences in the Sequence Listing, or a complementary sequence, that encodes a functionally equivalent polypeptide (i.e., a polypeptide having some degree of equivalent or similar biological activity) but differs in sequence from the sequence in the Sequence Listing, due to degeneracy in the genetic code. Included within this definition are polymorphisms that may or may not be readily detectable using a particular oligonucleotide probe of the polynucleotide encoding polypeptide, and improper or unexpected hybridization to allelic variants, with a locus other than the normal chromosomal locus for the polynucleotide sequence encoding polypeptide. "Allelic variant" or "polynucleotide allelic variant" refers to any of two or more alternative forms of a gene occupying the same chromosomal locus. Allelic variation arises naturally through mutation, and may result in phenotypic polymorphism within populations. Gene mutations may be "silent" or may encode polypeptides having altered amino acid sequence. "Allelic variant" and "polypeptide allelic variant" may also be used with respect to polypeptides, and in this case the terms refer to a polypeptide encoded by an allelic variant of a gene. "Splice variant" or "polynucleotide splice variant" as used herein refers to alternative forms of RNA transcribed from a gene. Splice variation naturally occurs as a result of alternative sites being spliced within a single transcribed RNA molecule or between separately transcribed RNA molecules, and may result in several different foπns of mRNA transcribed from the same gene. Thus, splice variants may encode polypeptides having different amino acid sequences, which may or may not have similar functions in the organism. "Splice variant" or "polypeptide splice variant" may also refer to a polypeptide encoded by a splice variant of a transcribed mRNA. As used herein, "polynucleotide variants" may also refer to polynucleotide sequences that encode paralogs and orthologs of the presently disclosed polypeptide sequences. "Polypeptide variants" may refer to polypeptide sequences that are paralogs and orthologs of the presently disclosed polypeptide sequences. Differences between presently disclosed polypeptides and polypeptide variants are limited so that the sequences of the former and the latter are closely similar overall and, in many regions, identical. Presently disclosed polypeptide sequences and similar polypeptide variants may differ in amino acid sequence by one or more substitutions, additions, deletions, fusions and truncations, which may be present in any combination. These differences may produce silent changes and result in a functionally equivalent transcription factor. Thus, it will be readily appreciated by those of skill in the art, that any of a variety of polynucleotide sequences is capable of encoding the transcription factors and transcription factor homolog polypeptides of the invention. A polypeptide sequence variant may have "conservative" changes, wherein a substituted amino acid has similar structural or chemical properties. Deliberate amino acid substitutions may thus be made on the basis of similarity in polarity, charge, solubility, hydrophobicity, hydrophilicity, and/or the amphipathic nature of the residues, as long as a substantial amount of the functional or biological activity of the transcription factor is retained. For example, negatively charged amino acids may include aspartic acid and glutamic acid, positively charged amino acids may include lysine and arginine, and amino acids with uncharged polar head groups having similar hydrophilicity values may include leucine, isoleucine, and valine; glycine and alanine; asparagine and glutamine; serine and threonine; and phenylalanine and tyrosine (more detail on conservative substitutions is presented in Table 4). More rarely, a variant may have "non-conservative" changes, for example, replacement of a glycine with a tryptophan. Similar minor variations may also include amino acid deletions or insertions, or both. Related polypeptides may comprise, for example, additions and/or deletions of one or more N-linked or O-linked glycosylation sites, or an addition and/or a deletion of one or more cysteine residues. Guidance in determining which and how many amino acid residues may be substituted, inserted or deleted without abolishing functional or biological activity may be found using computer programs well known in the art, for example, DNASTAR software (for example, in USPN 5,840,544). "Modulates" refers to a change in activity (biological, chemical, or immunological) or lifespan resulting from specific binding between a molecule and either a nucleic acid molecule or a protein. The term "plant" includes whole plants, shoot vegetative organs/structures (for example, leaves, stems and tubers), roots, flowers and floral organs/structures (for example, bracts, sepals, petals, stamens, carpels, anthers and ovules), seed (including embryo, endosperm, and seed coat) and fruit (the mature ovary), plant tissue (for example, vascular tissue, ground tissue, and the like) and cells (for example, guard cells, egg cells, and the like), and progeny of same. The class of plants that can be used in the method of the invention is generally as broad as the class of higher and lower plants amenable to transformation techniques, including angiosperms (monocotyledonous and dicotyledonous plants), gymnosperms, ferns, horsetails, psilophytes, lycophytes, bryophytes, and multicellular algae (for example, Figure 1, adapted from Daly et al. (2001) Plant Physiol. 127: 1328-1333; Figure 2, adapted from Ku et al. (2000) Proc. Natl. Acad. Sci. 97: 9121-9126; and also Tudge in The Variety of Life. Oxford University Press, New York, NY (2000) pp. 547-606). A "transgenic plant" refers to a plant that contains genetic material not found in a wild-type plant of the same species, variety or cultivar. The genetic material may include a transgene, an insertional mutagenesis event (such as by transposon or T-DNA insertional mutagenesis), an activation tagging sequence, a mutated sequence, a homologous recombination event or a sequence modified by chimeraplasty. Typically, the foreign genetic material has been introduced into the plant by human manipulation, but any method can be used as one of skill in the art recognizes. A transgenic plant may contain an expression vector or cassette. The expression vector or cassette typically comprises a polypeptide-encoding sequence operably linked (i.e., under regulatory control of) to appropriate inducible or constitutive regulatory sequences that allow for the expression of polypeptide. The expression vector or cassette can be introduced into a plant by transformation or by breeding after transformation of a parent plant. A plant refers to a whole plant as well as to a plant part, such as seed, fruit, leaf, or root, plant tissue, plant cells or any other plant material, for example, a plant explant, as well as to progeny thereof, and to in vitro systems that mimic biochemical or cellular components or processes in a cell. "Wild type" or "wild-type", as used herein, refers to a plant cell, seed, plant component, plant tissue, plant organ or whole plant that has not been genetically modified or treated in an experimental sense. Wild-type cells, seed, components, tissue, organs or whole plants may be used as controls to compare levels of expression and the extent and nature of trait modification with cells, tissue or plants of the same species in which a transcription factor expression is altered, e.g., in that it has been knocked out, overexpressed, or ectopically expressed. A "control plant" as used in the present invention refers to a plant cell, seed, plant component, plant tissue, plant organ or whole plant used to compare against transgenic or genetically modified plant for the purpose of identifying an enhanced phenotype in the transgenic or genetically modified plant. A control plant may in some cases be a transgenic plant line that comprises an empty vector or marker gene, but does not contain the recombinant polynucleotide of the present invention that is expressed in the transgenic or genetically modified plant being evaluated. In general, a control plant is a plant of the same line or variety as the transgenic or genetically modified plant being tested. A suitable control plant would include a genetically unaltered or non-transgenic plant of the parental line used to generate a transgenic plant herein. "Derivative" refers to the chemical modification of a nucleic acid molecule or amino acid sequence. Chemical modifications can include replacement of hydrogen by an alkyl, acyl, or amino group or glycosylation, pegylation, or any similar process that retains or enhances biological activity or lifespan of the molecule or sequence. A "trait" refers to a physiological, morphological, biochemical, or physical characteristic of a plant or particular plant material or cell. In some instances, this characteristic is visible to the human eye, such as seed or plant size, or can be measured by biochemical techniques, such as detecting the protein, starch, or oil content of seed or leaves, or by observation of a metabolic or physiological process, e.g. by measuring tolerance to water deprivation or particular salt or sugar concentrations, or by the observation of the expression level of a gene or genes, for example, by employing Northern analysis, RT-PCR, microarray gene expression assays, or reporter gene expression systems, or by agricultural observations such as abiotic stress tolerance or yield. Any technique can be used to measure the amount of, comparative level of, or difference in any selected chemical compound or macromolecule in the transgenic plants, however. "Trait modification" refers to a detectable difference in a characteristic in a plant ectopically expressing a polynucleotide or polypeptide of the present invention relative to a plant not doing so, such as a wild-type plant. In some cases, the trait modification can be evaluated quantitatively. For example, the trait modification can entail at least about a 2% increase or decrease in an observed trait or an even greater difference, compared with a wild-type plant. It is known that there can be a natural variation in the modified trait. Therefore, the trait modification observed entails a change of the normal distribution and magnitude of the trait in the plants compared with the distribution and magnitude observed in wild-type plants. When two or more plants have "similar morphologies", "substantially similar morphologies" or "a morphology that is substantially similar", the plants have comparable forms or appearances, including analogous features such as overall dimensions, height, width, mass, root mass, shape, glossiness, color, stem diameter, leaf size, leaf dimension, leaf density, internode distance, branching, root branching, number and form of inflorescences, and other macroscopic characteristics, and the genotypes of the plants with similar morphologies are not readily distinguishable based on morphological characteristics alone. The term "transcript profile" refers to the expression levels of a set of genes in a cell in a particular state, particularly by comparison with the expression levels of that same set of genes in a cell of the same type in a reference state. For example, the transcript profile of a particular transcription factor in a suspension cell is the expression levels of a set of genes in a cell repressing or overexpressing that transcription factor compared with the expression levels of that same set of genes in a suspension cell that has normal levels of that transcription factor. The transcript profile can be presented as a list of those genes whose expression level is significantly different between the two treatments, and the difference ratios. Differences and similarities between expression levels may also be evaluated and calculated using statistical and clustering methods. "Ectopic expression or altered expression" in reference to a polynucleotide indicates that the pattern of expression in, for example, a transgenic plant or plant tissue, is different from the expression pattern in a wild-type plant or a reference plant of the same species. The pattern of expression may also be compared with a reference expression pattern in a wild-type plant of the same species. For example, the polynucleotide or polypeptide is expressed in a cell or tissue type other than a cell or tissue type in which the sequence is expressed in the wild-type plant, or by expression at a time other than at the time the sequence is expressed in the wild-type plant, or by a response to different inducible agents, such as hormones or environmental signals, or at different expression levels (either higher or lower) compared with those found in a wild-type plant. The term also refers to altered expression patterns that are produced by lowering the levels of expression to below the detection level or completely abolishing expression. The resulting expression pattern can be transient or stable, constitutive or inducible. In reference to a polypeptide, the term "ectopic expression or altered expression" further may relate to altered activity levels resulting from the interactions of the polypeptides with exogenous or endogenous modulators or from interactions with factors or as a result of the chemical modification of the polypeptides. The term "overexpression" as used herein refers to a greater expression level of a gene in a plant, plant cell or plant tissue, compared to expression in a wild-type plant, cell or tissue, at any developmental or temporal stage for the gene. Overexpression can occur when, for example, the genes encoding one or more transcription factors are under the control of a strong promoter described herein (for example, the cauliflower mosaic virus 35 S transcription initiation region) or be induced when an appropriate environmental signal is present. Overexpression may occur throughout a plant or in specific tissues of the plant, depending on the promoter used, as described below. Overexpression may take place in plant cells normally lacking expression of polypeptides functionally equivalent or identical to the present transcription factors. Overexpression may also occur in plant cells where endogenous expression of the present transcription factors or functionally equivalent molecules normally occurs, but such normal expression is at a lower level. Overexpression thus results in a greater than normal production, or "overproduction" of the transcription factor in the plant, cell or tissue. The term "transcription regulating region" refers to a DNA regulatory sequence that regulates expression of one or more genes in a plant when a transcription factor having one or more specific binding domains binds to the DNA regulatory sequence. Transcription factors of the present invention possess an AP2 domain, a B3 domain, or both of these binding domains. The AP2 domain of the transcription factor binds to a transcription regulating region comprising the motif CAACA, and the B3 domain of the same transcription factor binds to a transcription regulating region comprising the motif CACCTG. The transcription factors of the invention also comprise an amino acid subsequence that forms a transcription activation domain that regulates expression of one or more abiotic stress tolerance genes in a plant when the transcription factor binds to the regulating region. A "sample" with respect to a material containing nucleic acid molecules may comprise a bodily fluid; an extract from a cell, chromosome, organelle, or membrane isolated from a cell; genomic DNA, RNA, or cDNA in solution or bound to a substrate; a cell; a tissue; a tissue print; a forensic sample; and the like. In this context "substrate" refers to any rigid or semi-rigid support to which nucleic acid molecules or proteins are bound and includes membranes, filters, chips, slides, wafers, fibers, magnetic or nonmagnetic beads, gels, capillaries or other tubing, plates, polymers, and microparticles with a variety of surface forms including wells, trenches, pins, channels and pores. A substrate may also refer to a reactant in a chemical or biological reaction, or a substance acted upon (for example, by an enzyme). Transcription Factors Modify Expression of Endogenous Genes A transcription factor may include, but is not limited to, any polypeptide that can activate or repress transcription of a single gene or a number of genes. As one of ordinary skill in the art recognizes, transcription factors can be identified by the presence of a region or domain of structural similarity or identity to a specific consensus sequence or the presence of a specific consensus DNA-binding site or DNA-binding site motif (for example, Riechmann et al. (2000) supra). The plant transcription factors may belong to the AP2 protein transcription factor family (Riechmann and Meyerowitz (1998) supra). Generally, the transcription factors encoded by the present sequences are involved in cell differentiation and proliferation and the regulation of growth. Accordingly, one skilled in the art would recognize that by expressing the present sequences in a plant, one may change the expression of autologous genes or induce the expression of introduced genes. By affecting the expression of similar autologous sequences in a plant that have the biological activity of the present sequences, or by introducing the present sequences into a plant, one may alter a plant's phenotype to one with improved traits related to abiotic stresses. The sequences of the invention may also be used to transform a plant and introduce desirable traits not found in the wild-type cultivar or strain. Plants may then be selected for those that produce the most desirable degree of over- or under-expression of target genes of interest and coincident trait improvement. The sequences of the present invention may be from any species, particularly plant species, in a naturally occurring form or from any source whether natural, synthetic, semi-synthetic or recombinant. The sequences of the invention may also include fragments of the present amino acid sequences. Where "amino acid sequence" is recited to refer to an amino acid sequence of a naturally occurring protein molecule, "amino acid sequence" and like terms are not meant to limit the amino acid sequence to the complete native amino acid sequence associated with the recited protein molecule. In addition to methods for modifying a plant phenotype by employing one or more polynucleotides and polypeptides of the invention described herein, the polynucleotides and polypeptides of the invention have a variety of additional uses. These uses include their use in the recombinant production (i.e., expression) of proteins; as regulators of plant gene expression, as diagnostic probes for the presence of complementary or partially complementary nucleic acids (including for detection of natural coding nucleic acids); as substrates for further reactions, for example, mutation reactions, PCR reactions, or the like; as substrates for cloning for example, including digestion or ligation reactions; and for identifying exogenous or endogenous modulators of the transcription factors. In many instances, a polynucleotide comprises a nucleotide sequence encoding a polypeptide (or protein) or a domain or fragment thereof. Additionally, the polynucleotide may comprise a promoter, an intron, an enhancer region, a polyadenylation site, a translation initiation site, 5' or 3' untranslated regions, a reporter gene, a selectable marker, or the like. The polynucleotide can be single-stranded or double-stranded DNA or RNA. The polynucleotide optionally comprises modified bases or a modified backbone. The polynucleotide can be, for example, genomic DNA or RNA, a transcript (such as an mRNA), a cDNA, a PCR product, a cloned DNA, a synthetic DNA or RNA, or the like. The polynucleotide can comprise a sequence in either sense or antisense orientations. Expression of genes that encode transcription factors that modify expression of endogenous genes, polynucleotides, and proteins are well known in the art. In addition, transgenic plants comprising isolated polynucleotides encoding transcription factors may also modify expression of endogenous genes, polynucleotides, and proteins. Examples include Peng et al. (1997) Genes Development 11 : 3194-3205, and Peng et al. (1999) Nature, 400: 256-261). In addition, many others have demonstrated that an Arabidopsis transcription factor expressed in an exogenous plant species elicits the same or very similar phenotypic response (for example, Fu et al. (2001) Plant Cell 13: 1791-1802; Nandi et al. (2000) Curr. Biol. 10: 215-218; Coupland (1995) Nature 377: 482-483; and Weigel and Nilsson (1995) Nature 377: 482-500). In another example, Mandel et al. (1992) Cell 71-133-143, and Suzuki et al.(2001) Plant J. 28: 409-418, teach that a transcription factor expressed in another plant species elicits the same or very similar phenotypic response of the endogenous sequence, as often predicted in earlier studies of

Arabidopsis transcription factors in Arabidopsis (Mandel et al. (1992) supra; Suzuki et al. (2001) supra). Other examples include Miiller et al. (2001) Plant J. 28: 169-179; Kim et al. (2001) Plant J. 25: 247-259; Kyozuka and Shimamoto (2002) Plant Cell Physiol. 43: 130-135; Boss and Thomas (2002) Nature, 416: 847-850; He et al. (2000) Transgenic Res. 9: 223-227; and Robson et al. (2001) Plant J. 28: 619-631. In yet another example, Gilmour et al. (1998) Plant J. 16: 433-442, teach an Arabidopsis AP2 transcription factor, CBF1 that, when overexpressed in transgenic plants, increases plant freezing tolerance. Jaglo et al. (2001 ) Plant Physiol. 127: 910-917, further identified sequences in Brassica napus which encode CBF-like genes and that transcripts for these genes accumulated rapidly in response to low temperature. Transcripts encoding CBF-like proteins were also found to accumulate rapidly in response to low temperature in wheat, as well as in tomato. An alignment of the CBF proteins from Arabidopsis, B. napus, wheat, rye, and tomato revealed the presence of conserved consecutive amino acid residues, PKK/RPAGRxKFxETRHP and DSAWR, that bracket the AP2/EREBP DNA binding domains of the proteins and distinguish them from other members of the AP2/EREBP protein family (Jaglo et al. (2001) supra). Transcription factors mediate cellular responses and control traits through altered expression of genes containing cis-acting nucleotide sequences that are targets of the introduced transcription factor. It is well appreciated in the art that the effect of a transcription factor on cellular responses or a cellular trait is determined by the particular genes whose expression is either directly or indirectly (for example, by a cascade of transcription factor binding events and transcriptional changes) altered by transcription factor binding. In a global analysis of transcription comparing a standard condition with one in which a transcription factor is overexpressed, the resulting transcript profile associated with transcription factor overexpression is related to the trait or cellular process controlled by that transcription factor. For example, the PAP2 gene (and other genes in the MYB family) have been shown to control anthocyanin biosynthesis through regulation of the expression of genes known to be involved in the anthocyanin biosynthetic pathway (Bruce et al. (2000) Plant Cell, 12: 65-79; Borevitz et al. (2000) Plant Cell 12: 2383-93). Further, global transcript profiles have been used successfully as diagnostic tools for specific cellular states (for example, cancerous vs. non-cancerous; Bhattacharjee et al. (2001) Proc Natl. Acad. Sci., USA, 98: 13790-13795; Xu et al. (2001) Proc. Natl. Acad. Sci., USA, 98: 15089-15094). Consequently, it is evident to one skilled in the art that similarity of transcript profile upon overexpression of different transcription factors would indicate similarity of transcription factor function.

Polypeptides and Polynucleotides of the Invention The present invention provides, among other things, transcription factors (TFs), and transcription factor homolog polypeptides, and isolated or recombinant polynucleotides encoding the polypeptides, or novel sequence variant polypeptides or polynucleotides encoding novel variants of transcription factors derived from the specific sequences provided here. These polypeptides and polynucleotides may be employed to modify a plant's characteristics. The sequences of G867 and G9 were previously identified in U.S. provisional patent application 60/101,349, filed September 22, 1998, at which time these sequences were identified as encoding or being transcription factors, which were defined as polypeptides having the ability to effect transcription of a target gene. Sequences that have gene-regulating activity have been determined to have specific and substantial utility (Federal Register (2001) 66(4): 1095). The functions of G867 and G9 were previously disclosed in U.S. provisional patent applications 60/227,439, filed August 22, 2000, and 60/166,228, filed November 17, 1999, respectively. The sequence of G993 was previously identified in U.S. provisional applications 60/108,734, filed November 17, 1998, and 60/125,814, filed March 23, 1999. The function of G993 was implied from its homologous relationship with G867, as disclosed in U.S. non- provisional application 09/934,455, filed August 22, 2001. The sequence of G1930 was previously identified in U.S. non-provisional application 09/934,455, filed August 22, 2001. The functions of G1930 were previously disclosed in U.S. non-provisional patent application 09/934,455, filed August 22, 2001. In some cases, exemplary polynucleotides encoding the polypeptides of the invention were identified in the Arabidopsis thaliana GenBank database using publicly available sequence analysis programs and parameters. Sequences initially identified were then further characterized to identify sequences comprising specified sequence strings corresponding to sequence motifs present in families of known transcription factors. In addition, further exemplary polynucleotides encoding the polypeptides of the invention were identified in the plant GenBank database using publicly available sequence analysis programs and parameters. Sequences initially identified were then further characterized to identify sequences comprising specified sequence strings corresponding to sequence motifs present in families of known transcription factors. Polynucleotide sequences meeting such criteria were confirmed as transcription factors. Additional polynucleotides of the invention were identified by screening Arabidopsis thaliana and/or other plant cDNA libraries with probes corresponding to known transcription factors under low stringency hybridization conditions. Additional sequences, including full length coding sequences, were subsequently recovered by the rapid amplification of cDNA ends (RACE) procedure. Where necessary, multiple rounds of RACE are performed to isolate 5' and 3' ends. The full-length cDNA was then recovered by a routine end-to-end polymerase chain reaction (PCR) using primers specific to the isolated 5' and 3' ends. Exemplary sequences are provided in the Sequence Listing. The polynucleotides of the invention can be or were ectopically expressed in overexpressor or knockout plants and the changes in the characteristic(s) or trait(s) of the plants observed. Therefore, the polynucleotides and polypeptides can be employed to improve the characteristics of plants. The polynucleotides of the invention can be or were ectopically expressed in overexpressor plant cells and the changes in the expression levels of a number of genes, polynucleotides, and/or proteins of the plant cells observed. Therefore, the polynucleotides and polypeptides can be employed to change expression levels of a genes, polynucleotides, and/or proteins of plants. G867, which we have determined to confer abiotic stress tolerance in plants when overexpressed, has been described in the literature as related to BI3/1 1 (RA VI; Kagaya et al. (1999) Nucleic Acids Res. 27: 470-478) based on the presence of a B3 domain (which is also found in the ABI3/VP1 family of transcription factors). The protein also contains an AP2 domain, and is therefore presently included in the AP2/ERF family of transcription factors. Both the AP2 domain transcription factors and the B3 domain transcription factors are described below. AP2 domain transcription factors. Ohme-Takagi and Shinshi (1995) Plant Cell 1, 173-182) determined that the function of the AP2 domain is DNA binding. The AP2 region of the putative tobacco transcription factor EREBP2 is responsible for its binding to the cis-acting ethylene response DNA element referred to as the GCC-repeat. As discussed by Ohme-Takagi and Shinshi (1995) supra), the DNA-binding or AP2 domain of EREBP2 contains no significant amino acid sequence similarities or obvious structural similarities with other known transcription factors or DNA binding motifs beyond AP2 transcription factors. Thus, the domain appears to be a novel DNA-binding motif that, to date, has only been found in plant proteins. The RAV-like proteins, including G867 and other members of the G867 clade of transcription factor polypeptides, form a small subgroup in the AP2/ERF family of AP2 transcription factors. This large gene family includes at least 145 transcription factors, and can be further divided in three larger subfamilies: (a) The APETALA2 class is characterized by the presence of two AP2 DNA binding domains, and contains fourteen genes. (b) The RAV subgroup, which includes six genes, is characterized by the presence of a B3 DNA binding domain in addition to the AP2 DNA binding domain. (c) The AP2/ERF subfamily, which is the largest subfamily and includes 125 genes, is characterized by the presence of only one AP2 DNA binding domain, and includes genes that are involved in abiotic and biotic stress responses. This subfamily is composed of two relatively equal size subgroups, the DREB and ERF subgroups (Sakuma et al. (2002) Biochem. Biophys. Res. Comm. 290: 998-1009), which are distinguished on the basis of specific residues in the AP2 DNA binding domain. The binding characteristics of G867 (RAVI) have been characterized by Kagaya et al. ((1999) Nucleic Acids Res. 27: 470-478). There is no published information on the biological function of the RAV-like transcription factors. B3 domain transcription factors. Transcription factors of the ABI3 VP1 family have been implicated in seed maturation processes. ABB (G621) plays an important role in the acquisition of desiccation tolerance in late embryogenesis. This process is related to dehydration tolerance as evidenced by the protective function of late embryogenesis abundant (LEA) genes such as HVAl (Xu et al. (1996) Plant Physiol. 110: 249-257; Sivamani et al. (2000) Plant Science 155: 1 -9). Mutants for Arabidopsis ABI3 (Ooms et al. (1993) Plant Physiol. 102: 1185-1191) and the maize ortholog VP1 (Carson et al. (1997) Plant J. 12: 1231-1240) and references therein) show severe defects in the attainment of desiccation tolerance. Also, 5?>::ABI3 overexpression in combination with increased levels of abscisic acid results in an induction of several ABA/cold/drought-responsive genes such as RAB18 and RD29A and increased freezing tolerance in Arabidopsis (Tamminen et al. (2001) Plant J. 25: 1-8). This illustrates the relatedness of desiccation and dehydration tolerance and demonstrates that the seed-specific ABI3 transcription factor does not require additional seed-specific proteins to function in vegetative tissues. Both in Arabidopsis and maize, the B3 domain of ABI3NP1 binds the RY/SPH motif (Ezcurra et al. (2000) Plant J. 24: 57-66); Carson et al. (1997) supra) while the B2 domain interacts with the

ABRE elements in a complex involving bZIP transcription factors (TRAB1 in maize, Hobo et al. (1999) Proc. Natl. Acad. Sci. USA 96: 15348-15353). While in Arabidopsis the B3 domain of ABI3 is essential for abscisic acid dependent activation of late embryogenesis genes (Ezcurra et al. (2000) supra), the B3 domain of VP1 is not essential for ABA regulated gene expression in maize seed (Carson et al. (1997) supra; McCarty et al. (1989) Plant Cell 1 : 523-532). This difference in the regulatory network between Arabidopsis and maize can be explained by differential usage of the RY/SPH versus the ABRE element in the control of seed maturation gene expression (motif (Ezcurra et al. (2000) supra). The RY/SPH element is a key element in gene regulation during late embryogenesis in Arabidopsis (Reidt et al. (2000) Plant J. 21 : 401-408) while it seems to be less important for seed maturation in maize (McCarty et al. (19S9) supra). Mutations in two other B3 domain transcription factors, FUS3 (G1014) and LEC2 (G3035) result in pleiotropic effects. In the case of fus3, these effects are mainly restricted to seed development during late embryogenesis (Keith et al. (1994) Plant Cell 6: 589-600). Overexpression of LEC2 results in somatic embryo formation on the cotyledons (Stone et al. (2001) Proc. Natl. Acad. Sci. USA 98: 11806- 1 1811). The FUS3 protein can be considered as a natural truncation of the ABI3 protein (Luerssen et al. (1998) Plant J. 15: 755-764); like the latter, it binds to the RY/SPH element, and can activate the expression from target promoters even in non-seed tissues (Reidt et al. (2000) supra). Singh et al. have recently submitted an EST polynucleotide sequence (NCBI Accession No. CB686050, April 9, 2003) from a transgenic Brassica napus that has been shown to be constitutively frost resistant as a result of overexpressing CBF17. The predicted polypeptide sequence has a DML motif that is 90% identical, and a B3 domain that is 95% identical, to the DML motif and B3 domain of G867, respectively. The protein predicted from this sequence does not comprise an AP2 domain.

The G867 clade of Transcription Factor Polypeptides Kagaya et al. ((1999) supra) cloned and characterized G867 (RAVI) and G9 (RAV2) from Arabidopsis thaliana. The two transcription factors were found to contain two distinct amino acid sequence domains found only in higher plant species, the AP2 and B3 domains. The N-terminal regions of G867 and G9 were shown to be homologous to the AP2 DNA-binding domain present in the Arabidopsis APETALA2 and tobacco EREBP proteins families, while the C-terminal region exhibited homology to the B3 domain of VP1/ABI3 transcription factors. Binding site selection assays using a recombinant glutathione S-transferase fusion protein revealed that G867 bound specifically to bipartite recognition sequences composed of two unrelated motifs, 5'-CAACA-3' and 5'-CACCTG-3', separated by various spacings in two different relative orientations. Analyses using various deletion derivatives of the RAVI fusion protein showed that the AP2 and B3-like domains of RAVI bind autonomously to the CAACA and CACCTG motifs, respectively, and together achieve a high affinity and specificity of binding. Kagaya et al. concluded that the AP2 and B3-like domains of RAVI are connected by a highly flexible structure enabling the two domains to bind to the CAACA and CACCTG motifs in various spacings and orientations. This binding of the AP2 and B3 domains cooperatively enhances the DNA binding affinity of a G867 clade member. The RAV-like proteins, including G867 and other members of the G867 clade of transcription factor polypeptides (for example, G9, G993, G1930, G3389, G3390, G3391 , G3432, G3451, G3452, G3453, and G3455), generally have both AP2 and B3 domains. Within the G867 clade of transcription factor polypeptides, there is a high degree of conservation of the AP2 and B3 domains in all members of the clade. The proteins in the G867 clade were also found to possess a subsequence with a high degree of conservation between the AP2 and B3 domains. This subsequence was designated DML motif. The DML motif does not appear to be present in transcription factors outside of the G867 clade (more detailed description of the DML motif appears below, and a list of DML motif sequences may be found in Table 2) Table 1 shows the polypeptides identified by polypeptide SEQ ID NO and Mendel Gene ID (GID) No , presented m order of similarity to the G867 AP2 domam, and includes the AP2 and B3 bindmg domams of the polypeptide in amino acid coordmates, the AP2 domain sequences, the percent identity to the AP2 domam of G867, the B3 domams, and the percent identity to the B3 domam of G867 A number of these sequences have been shown to confer abiotic stress tolerant phenotypes when overexpressed m plants, as mdicated in the last column of Table 1 (an empty cell in the last column indicates msufficient or no data) These functional polypeptide sequences mclude AP2 and B3 domams with at least 77% and 70% identity to the AP2 and B3 domams of G867, respectively

Table 1 Binding domams of the G867 clade of transcription factor polypeptides

* preliminary data indicates an ABA insensitive phenotype in one overexpressing line; one line is also more tolerant to sucrose and cold in germination assays

Generally, the transcription factors of the present invention possess an AP2 domain and a B3 domain (examples of useful artificial constructs that lack one or the other domain are noted in Example X). The present invention also includes fragments of such transcription factors, which may be comprised of both, or only one, of these binding domains (several orthologs of G867 found by BLAST analysis as described below comprise only one of the domains). Generally, the AP2 domain of the transcription factors will bind to a transcription-regulating region comprising the motif CAACA, and the B3 domain of the same transcription factor binds to a second transcription regulating region comprising the motif CACCTG. Each of these transcription factors also comprise an amino acid subsequence that forms a transcription activation domain that regulates expression of one or more abiotic stress tolerance genes in a plant when the transcription factor binds to the regulating region. As shown in Table 1, the AP2 and B3 domains of the transcription factors within the G867 clade are at least 75% and 69% identical to the respective corresponding domains of G867, and all of the transcription factors studied thus far, which rely on the binding specificity of their conserved AP2 and B3 domains, have very similar or identical functions in plants as they confer increased abiotic stress tolerance when overexpressed. Therefore, the invention provides polynucleotides comprising Arabidopsis SEQ ID NOs: 1, 3, 5, 7, and fragments thereof; and non-Arabidopsis sequences SEQ ID NOs: 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, paralogs, orthologs, equivalogs, and fragments thereof. The invention also provides polypeptides and the polynucleotides that encode them, said polypeptides comprising Arabidopsis SEQ ID NOs: 2, 4, 6, 8, and fragments thereof; and non-Arabidopsis SEQ ID NOs: 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, 53, paralogs, orthologs, equivalogs, and fragments thereof. A number of these polynucleotides have been shown to have a strong association with abiotic stress tolerance, in that plants that overexpress these sequences are more tolerant to one or more of these stresses. The invention thus pertains to transgenic plants that overexpress the sequences in the sequence listing and that are more tolerant to abiotic stress. The invention also encompasses a complement of the polynucleotides. The polynucleotides are useful for screening libraries of molecules or compounds for specific binding and for creating transgenic plants having increased abiotic stress tolerance. A number of the polynucleotides of the invention have been, and the remainder of the polynucleotides of the invention may be, ectopically expressed in overexpressor or knockout plants and the changes in the characteristic(s) or trait(s) of the plants observed. Therefore, the polynucleotides and polypeptides can be employed to improve the characteristics of plants. The polynucleotides are particularly useful when they are hybridizable array elements in a microarray. The array elements are organized in an ordered fashion so that each element is present at a specified location on the substrate. Because the array elements are at specified locations on the substrate, the hybridization patterns and intensities (which together create a unique expression profile) can be interpreted in terms of expression levels of particular genes and can be correlated with a particular stress, pathology, or treatment. Such a microarray can be employed to monitor the expression of genes that are differentially expressed in response to abiotic stresses. The microarray can be used in large scale genetic or gene expression analysis of a large number of polynucleotides; or in the diagnosis of abiotic stress before phenotypic symptoms are evident. Furthermore, the microarray can be employed to investigate cellular responses, such as cell proliferation, transformation, and the like. The invention also entails an agronomic composition comprising a polynucleotide of the invention in conjunction with a suitable carrier and a method for altering a plant's trait using the composition.

Characteristics of the G867 clade of transcription factor polypeptides The polypeptide members of the invention belong to the G867 clade of transcription factor polypeptides, and, being a set of equivalogs, have conserved functional characteristics and derive from a last common ancestor as shown in Figure 3. Examination of the alignment of only those sequences in the G867 clade (having monocot and dicot subnodes), indicates 1) a high degree of conservation of the AP2 domains in all members of the clade, 2) a high degree of conservation of the B3 domains in all members of the clade; and 3) a high degree of conservation of an additional motif, the "DML motif, found between the AP2 and B3 domains in members of the clade. The DML motif may be represented by: H/R-S-K-Xa-E/G-I/V-V-D-M-L-R-K/R-H-T-Y-Xa-E/D/N-E-L F-Xa-Q/H-S/N R/G where Xa is any amino acid). SEQ ID NO: 54 (corresponding to positions 135-152 in G867) is an example of a DML motif. The consensus DML motif sequence determined from an alignment of multiple G867 clade member sequences is represented by the shorter SEQ ID NO: 55 and has been found in all functional members of the G867 clade studied thus far: S-K-Xa-Xa-Xa-V-D-M-L-R-Xa-H-T-Y-Xa-Xa-E. As a conserved motif found in G867 and its paralogs, the DML motif was used to identify additional orthologs of SEQ ID NO: 2. A significant number of sequences were found that had a minimum of 71% identity to the 22 residue DML motif of G867. A number of these motifs are shown in Table 2. The five amino acid residue sequence: V-D-M-L-R represents the most conserved subsequence within the DML motif and has been found in all putative members of the G867 clade identified by BLAST. The invention thus encompasses polypeptides with substantially similar activity with SEQ ID

NO: 2, and generally comprise, in order from N- to C-termini, an AP2 domain, a DML motif, and a B3 domain (some functional members of the clade may lack most or all of the AP2 or B3 domains, noted in Example X). The DML motif is represented by motifs that have 71% or greater identity to the DML motif of G867 or the consensus DML motif SEQ ID NO: 55, and include the highly conserved subsequence V- D-M-L-R. Arabidopsis sequences thought to be paralogous or otherwise highly related evolutionarily to G867 were aligned using Clustal X (version 1.81, June 2000). BLASTP analysis of proprietary and public databases with protein sequences of this set identified additional sequences with a high degree of sequence relatedness to G867. A number of these genes and encoded sequences that are members of the G867 clade of transcription factor polypeptides are now known to produce an abiotic stress tolerance phenotype when the polypeptides are overexpressed, and include Arabidopsis sequences G867 (SEQ ID NO: 1 and 2), G9 (SEQ ID NO: 3 and 4), G993 (SEQ ID NO: 5 and 6), G1930 (SEQ ID NO: 7 and 8), rice sequences G3389 (SEQ ID NO: 31 and 32), G3390 (SEQ ID NO: 33 and 34) and G3391 (SEQ ID NO: 35 an 36), soy sequences G3451 (SEQ ID NO: 17 and 18), G3452 (SEQ ID NO: 21 and 22), G3453 (SEQ ID NO: 23 and 34) and G3455 (SEQ D NO: 27 and 28) and corn sequence G3432 (SEQ ID NO: 37 and 38). These sequences were then aligned again, and a neighbor-joining algorithm used to generate a phylogenetic tree, using Clustal X vl .81 's phylogenetic capabilities. In this alignment, G867 and it paralogs G9, G993 and G1930 appeared in a clade along with two soybean sequences and several rice sequences. Based on the utility of the Arabidopsis sequences, as noted below, and the evolutionary history revealed by analysis of the phylogenetic tree (that the last common ancestor of the monocots and the eudicots had only one gene corresponding to transcription factors of the present invention, which functioned in abiotic stress tolerance), transcription factors of the G867 clade comprise a number of genes mvolved in the control of abiotic stress tolerance. Upon translation of these nucleotide sequences in a frame that provided the identified conserved motif, all the resulting protein sequences were found to have either a conserved AP2 domain before the DML motif, or a B3 domain after the DML motif (i.e., m BU024575, BQ405698, BF424857, BZ458719, AP002913, and AX654438). The protein sequences having conserved AP2 and or B3 domains in the expected location were aligned with the previously aligned set of AP2 and B3 sequences, and a neighbor- joining algorithm was used to generate a phylogenetic tree, as described above. In this tree, the additional sequences identified through the DML motif all were found within the G867 clade identified previously, indicatmg that the DML motif was successfully used to identify new orthologs of G867, listed in Table 2. Many of these sequences confer abiotic stress tolerant phenotypes when overexpressed in plants, as indicated m the last column of Table 2 (an empty cell in this column mdicates a sequence that has not yet been tested in this study). These functional polypeptide sequences mclude DML motifs with 71% or greater identity to the DML motif of G867.

Table 2. Similarity of the DML motifs in G867 clade sequences

Based on experimental observations with sequences that have significant homology to the G867 clade polypeptides, we have determined that the clade member sequences with an AP2 domain, a DML motif and a B3 domain function to confer stress tolerance in plants when the sequences are overexpressed. Variations of these sequences have also been shown to confer stress tolerance, as noted below (e.g., sequences that lack all or part of an AP2 or B3 domain). One may also envision conferring stress tolerance by overexpressing sequences having an AP2 domain, a DML motif and a B3 domain in an order different from that used in the experiments disclosed in the Examples. Other variant sequences may be envisioned that would also function to confer tolerance, such as those with "domain swaps", that is, sequences having AP2, DML and/or B3 domains derived from more than one species, or produced synthetically. Domain swapping methods may be found in, for example, Bhullar et al. (2003) Plant Physiol. 132: 988-998, or Lamb et al. (2003) Proc. Natl. Acad. Sci. USA 100: 6558-6563. Domain swap experiments may also be useful in elucidating the roles each domain plays in stress tolerance, particularly by substituting domains from sequences with lesser or greater ability to confer stress tolerance.

Producing Polypeptides The polynucleotides of the invention include sequences that encode transcription factors and transcription factor homolog polypeptides and sequences complementary thereto, as well as unique fragments of coding sequence, or sequences complementary thereto. Such polynucleotides can be, for example, DNA or RNA, the latter including mRNA, cRNA, synthetic RNA, genomic DNA, cDNA synthetic DNA, oligonucleotides, etc. The polynucleotides are either double-stranded or single-stranded, and include either, or both sense (i.e., coding) sequences and antisense (i.e., non-coding, complementary) sequences. The polynucleotides include the coding sequence of a transcription factor, or transcription factor homolog polypeptide, in isolation, in combination with additional coding sequences (e.g., a purification tag, a localization signal, as a fusion-protein, as a pre-protein, or the like), in combination with non-coding sequences (for example, introns or inteins, regulatory elements such as promoters, enhancers, terminators, and the like), and or in a vector or host environment in which the polynucleotide encoding a transcription factor or transcription factor homolog polypeptide is an endogenous or exogenous gene. A variety of methods exist for producing the polynucleotides of the invention. Procedures for identifying and isolating DNA clones are well known to those of skill in the art, and are described in, for example, Berger and Kimmel, Guide to Molecular Cloning Techniques, Methods in Enzymology, vol. 152 Academic Press, Inc., San Diego, CA ("Berger"); Sambrook et al. Molecular Cloning - A Laboratory Manual (2nd Ed.), Vol. 1 -3, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1989

("Sambrook") and Current Protocols in Molecular Biology. Ausubel et al. eds., Current Protocols, a joint venture between Greene Publishing Associates, Inc. and John Wiley & Sons, Inc., (supplemented through 2000) ("Ausubel"). Alternatively, polynucleotides of the invention can be produced by a variety of in vitro amplification methods adapted to the present invention by appropriate selection of specific or degenerate primers. Examples of protocols sufficient to direct persons of skill through in vitro amplification methods, include polymerase chain reaction (PCR), ligase chain reaction (LCR), Qβ-replicase amplification and other RNA polymerase mediated techniques (for example, NASBA), e.g., for the production of the homologous nucleic acids of the invention are found in Berger (supra), Sambrook (supra), and Ausubel (supra), as well as Mullis et al. (1987) PCR Protocols A Guide to Methods and Applications, Innis et al. eds., Academic Press Inc., San Diego. Improved methods for cloning in vitro amplified nucleic acids are described in Wallace et al., US Pat. No. 5,426,039. Methods for amplifying large nucleic acids by PCR are summarized in Cheng et al. (1994) Nature 369: 684-685 and the references cited therein, in which PCR amplicons of up to 40 kb are generated. One of skill will appreciate that essentially any RNA can be converted into a double-stranded DNA suitable for restriction digestion, PCR expansion and sequencing using reverse transcriptase and a polymerase (for example, Ausubel, Sambrook and Berger, all supra). Alternatively, polynucleotides and oligonucleotides of the invention can be assembled from fragments produced by solid-phase synthesis methods. Typically, fragments of up to approximately 100 bases are individually synthesized and then enzymatically or chemically ligated to produce a desired sequence, e.g., a polynucleotide encoding all or part of a transcription factor. For example, chemical synthesis using the phosphoramidite method is described, e.g., by Beaucage et al. (1981) Tetrahedron Letters 22: 1859-1869; and Matthes et al. (1984) EMBOJ. 3: 801-805. According to such methods, oligonucleotides are synthesized, purified, annealed to their complementary strand, ligated and then optionally cloned into suitable vectors. And if so desired, the polynucleotides and polypeptides of the invention can be custom ordered from any of a number of commercial suppliers. Homologous Sequences Sequences homologous, i.e., that share significant sequence identity or similarity, to those provided in the Sequence Listing, derived from Arabidopsis thaliana or from other plants of choice, are also an aspect of the invention. Homologous sequences can be derived from any plant including monocots and dicots and in particular agriculturally important plant species, including but not limited to, crops such as soybean, wheat, corn (maize), potato, cotton, rice, rape, oilseed rape (including canola), sunflower, alfalfa, clover, sugarcane, and turf; or fruits or fruit trees, vegetables such as banana, blackberry, blueberry, strawberry, and raspberry, cantaloupe, carrot, cauliflower, coffee, cucumber, eggplant, grapes, honeydew, lettuce, mango, melon, onion, papaya, peas, peppers, pineapple, pumpkin, spinach, squash, sweet corn, tobacco, tomato, tomatillo, watermelon, rosaceous fruits (such as apple, peach, pear, cherry and plum) and brassicas (such as broccoli, cabbage, cauliflower, Brussels sprouts, and kohlrabi). Other crops, including fruits and vegetables, whose phenotype can be changed and which comprise homologous sequences include barley; rye; millet; sorghum; currant; avocado; citrus fruits such as oranges, lemons, grapefruit and tangerines, artichoke, cherries; nuts such as the walnut and peanut; endive; leek; roots such as arrowroot, beet, cassava, turnip, radish, yam, and sweet potato; and beans. The homologous sequences may also be derived from woody species, such as pine, poplar and eucalyptus, or mint or other labiates. In addition, homologous sequences may be derived from plants that are evolutionarily related to crop plants, but which may not have yet been used as crop plants. Examples include deadly nightshade (Atropa belladona), related to tomato; jimson weed (Datura strommium), related to peyote; and teosinte (Zea species), related to corn (maize).

Orthologs and Paralogs Homologous sequences as described above can comprise orthologous or paralogous sequences.

Several different methods are known by those of skill in the art for identifying and defining these functionally homologous sequences. General methods for defining orthologs and paralogs are described; an ortholog, paralog or homolog may be identified by one or more of the methods described below. Within a single plant species, gene duplication may cause two copies of a particular gene, giving rise to two or more genes with similar sequence, and are known as paralogs. A paralog is therefore a similar gene formed by duplication within the same species. Paralogous genes may retain similar functions of the encoded proteins. In such cases, paralogs can be used interchangeably with respect to certain embodiments of the instant invention (for example, transgenic expression of a coding sequence).

Example of related paralogs includes the CBF family, with three well-defined members in Arabidopsis and one ortholog in Brassica napus, all of which control pathways involved in both freezing and drought stress (Gilmour et al. (1998) Plant J. 16: 433-442; Jaglo et al. (1998) Plant Physiol. Ill: 910-917). Paralogs typically cluster together or in the same clade (a group of similar genes) when a gene family phylogeny is analyzed using programs such as CLUSTAL (Thompson et al. (1994) Nucleic Acids Res. 22: 4673-4680; Higgins et al. (1996) Methods Enzymol. 266: 383-402). Groups of similar genes can also be identified with pair-wise BLAST analysis (Feng and Doolittle (1987) J. Mol. Evol. 25: 351-360). For example, a clade of very similar MADS domain transcription factors from Arabidopsis all share a common function in flowering time (Ratcliffe et al. (2001) Plant Physiol. 126: 122-132), and a group of very similar AP2 domain transcription factors from Arabidopsis are involved in tolerance of plants to freezing (Gilmour et al. (1998) Plant J. 16: 433-442). Analysis of groups of similar genes with similar function that fall within one clade can yield sub-sequences that are particular to the clade. These subsequences, known as consensus sequences, can not only be used to define the sequences within each clade, but define the functions of these genes; genes within a clade may contain paralogous sequences, or orthologous sequences that share the same function (for example, Mount (2001), in Bioinformatics:

Sequence and Genome Analysis Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, page 543). Speciation, the production of new species from a parental species, can also give rise to two or more genes with similar sequence and similar function. Transcription factor gene sequences are conserved across diverse eukaryotic species lines (Goodrich et al. (1993) Cell 75: 519-530; Lin et al. (1991) Nature 353: 569-571; Sadowski et al. (1988) Nature 335: 563-564). Plants are no exception to this observation; diverse plant species possess transcription factors that have similar sequences and functions. Because plants have common ancestors, many genes in any plant species will have a corresponding highly similar or orthologous gene in another plant species. Orthologous sequences from different organisms have highly and often essentially identical functions (Lee et al. (2002) Genome Res. 12: 493-502; Remm et al. (2001) J. Mol. Biol. 314: 1041-1052), and are often interchangeable between species without losing function. Once a phylogenic tree for a gene family of one species has been constructed using a program such as CLUSTAL (Thompson et al. (1994) Nucleic Acids Res. 22: 4673- 4680; Higgins et al. (1996) supra) potential orthologous sequences can be placed into the phylogenetic tree and their relationship to genes from the species of interest can be determined. Orthologous sequences can also be identified by a reciprocal BLAST strategy. Once an orthologous sequence has been identified, the function of the ortholog can be deduced from the identified function of the reference sequence. The following references represent a small sampling of the many studies that demonstrate that conserved transcription factor genes from diverse species are likely to function similarly (i.e., regulate similar target sequences and control the same traits), and that transcription factors may be transfonned into diverse species to confer or improve traits. (1) Distinct Arabidopsis transcription factors, including G28 (found in US Patent 6,664,446), G482 (found in US Patent Application 20040045049), G867 (found in US Patent Application 20040098764), and G1073 (found in US Patent 6,717,034), have been shown to confer stress tolerance or increased biomass when the sequences are overexpressed. The polypeptides sequences belong to distinct clades of transcription factor polypeptides that include members from diverse species. In each case, a significant number of clade member sequences derived from both dicots and monocots have been shown to confer increased biomass or tolerance to stress when the sequences were overexpressed (unpublished data).

(2) The Arabidopsis NPRl gene regulates systemic acquired resistance (SAR); over-expression of NPRl leads to enhanced resistance in Arabidopsis. When either Arabidopsis NPRl or the rice NPRl ortholog was overexpressed in rice (which, as a monocot, is diverse from Arabidopsis), challenge with the rice bacterial blight pathogen Xanthomonas oryzae pv. Oryzae, the transgenic plants displayed enhanced resistance (Chern et al. (2001) Plant J. 27: 101-113). NPRl acts through activation of expression of transcription factor genes, such as TGA2 (Fan and Dong (2002) Plant Cell 14: 1377-1389).

(3) E2F genes are involved in transcription of plant genes for proliferating cell nuclear antigen (PCNA). Plant E2Fs share a high degree of similarity in amino acid sequence between monocots and dicots, and are even similar to the conserved domains of the animal E2Fs. Such conservation indicates a functional similarity between plant and animal E2Fs. E2F transcription factors that regulate meristem development act through common cis-elements, and regulate related (PCNA) genes. (Kosugi and Ohashi, (2002) Plant J. 29: 45-59).

(4) The ABI5 gene (ABA insensitive 5) encodes a basic leucine zipper factor required for ABA response in the seed and vegetative tissues. Co-transfonnation experiments with ABI5 cDNA constructs in rice protoplasts resulted in specific transactivation of the ABA-inducible wheat, Arabidopsis, bean, and barley promoters. These results demonstrate that sequentially similar ABI5 transcription factors are key targets of a conserved ABA signaling pathway in diverse plants. (Gampala et al. (2001) J. Biol. Chem. 277: 1689-1694).

(5) Sequences of three Arabidopsis GAMYB-like genes were obtained on the basis of sequence similarity to GAMYB genes from barley, rice, and L. temulentum. These three Arabidopsis genes were determined to encode transcription factors (AtMYB33, AtMYB65, and AtMYBlOl) and could substitute for a barley GAMYB and control α-amylase expression. (Gocal et al. (2001) Plant Physiol. 127: 1682-1693).

(6) The floral control gene LEAFY from Arabidopsis can dramatically accelerate flowering in numerous dictoyledonous plants. Constitutive expression of Arabidopsis LEAFY also caused early flowering in transgenic rice (a monocot), with a heading date that was 26-34 days earlier than that of wild-type plants. These observations indicate that floral regulatory genes from Arabidopsis are useful tools for heading date improvement in cereal crops. (He et al. (2000) Transgenic Res. 9: 223-227).

(7) Bioactive gibberellins (GAs) are essential endogenous regulators of plant growth. GA signaling tends to be conserved across the plant kingdom. GA signaling is mediated via GAI, a nuclear member of the GRAS family of plant transcription factors. Arabidopsis GAI has been shown to function in rice to inhibit gibberellin response pathways. (Fu et al. (2001) Plant Cell 13: 1791- 1802). (8) The Arabidopsis gene SUPERMAN (SUP), encodes a putative transcription factor that maintains the boundary between stamens and carpels. By over-expressing Arabidopsis SUP in rice, the effect of the gene's presence on whorl boundaries was shown to be conserved. This demonstrated that SUP is a conserved regulator of floral whorl boundaries and affects cell proliferation. (Nandi et al. (2000) Curr. Biol. 10: 215-218), (9) Maize, petunia and Arabidopsis myb transcription factors that regulate flavonoid biosynthesis are genetically similar and affect the same trait in their native species. Therefore, sequence and function of these myb transcription factors correlate with each other in these diverse species (Borevitz et al. (2000) Plant Cell 12: 2383-2394). (10) Wheat reduced height- 1 (Rht-Bl/Rht-Dl) and maize dwarf-8 (d8) genes are orthologs of the Arabidopsis gibberellin insensitive (GAI) gene. Both of these genes have been used to produce dwarf grain varieties that have improved grain yield. These genes encode proteins that resemble nuclear transcription factors and contain an SH2-like domain, indicating that phosphotyrosine may participate in gibberellin signaling. Transgenic rice plants containing a mutant GAI allele from Arabidopsis have been shown to produce reduced responses to gibberellin and are dwarfed, indicating that mutant GAI orthologs could be used to increase yield in a wide range of crop species (Peng et al. (1999) Nature 400: 256-261).

Transcription factors that are homologous to the listed sequences will typically share at least about 75% and 69% amino acid sequence identity in the AP2 and B3 domains, respectively. More closely related transcription factors can share at least about 81% and about 78% amino acid sequence identity in the AP2 and B3 domains, respectively, or more sequence identity with the listed sequences, or with the listed sequences but excluding or outside a known consensus sequence or consensus DNA-binding site, or with the listed sequences excluding one or all conserved domains. At the nucleotide level, the sequences will typically share at least about 40% nucleotide sequence identity, preferably at least about 50%o, about 60%, about 70% or about 80% sequence identity, and more preferably about 85%, about 90%, about 95% or about 97% or more sequence identity to one or more of the listed sequences, or to a listed sequence but excluding or outside a known consensus sequence or consensus DNA-binding site, or outside one or all conserved domain. The degeneracy of the genetic code enables major variations in the nucleotide sequence of a polynucleotide while maintaining the amino acid sequence of the encoded protein. AP2 domains within the AP2 transcription factor family may exhibit a higher degree of sequence homology, such as at least 77% amino acid sequence identity including conservative substitutions, and preferably at least 80% sequence identity, and more preferably at least 85%, or at least about 86%, or at least about 87%, or at least about 88%, or at least about 90%, or at least about 95%, or at least about 98% sequence identity. Transcription factors that are homologous to the listed sequences should share at least 30%, or at least about 60%, or at least about 75%, or at least about 80%, or at least about 90%, or at least about 95%o amino acid sequence identity over the entire length of the polypeptide or the homolog. Percent identity can be determined electronically, e.g., by using the MEGALIGN program (DNASTAR, Inc. Madison, Wis.). The MEGALIGN program can create alignments between two or more sequences according to different methods, for example, the clustal method (for example, Higgins and Sharp (1988) Gene 73: 237-244). The clustal algorithm groups sequences into clusters by examining the distances between all pairs. The clusters are aligned pairwise and then in groups. Other alignment algorithms or programs may be used, including FASTA, BLAST, or ENTREZ, FASTA and BLAST, and which may be used to calculate percent similarity. These are available as a part of the GCG sequence analysis package (University of Wisconsin, Madison, Wis.), and can be used with or without default settings. ENTREZ is available through the National Center for Biotechnology Information. In one embodiment, the percent identity of two sequences can be determined by the GCG program with a gap weight of 1, e.g., each amino acid gap is weighted as if it were a single amino acid or nucleotide mismatch between the two sequences (for example, USPN 6,262,333). In the present application, the percentage similarity between two polypeptide sequences, e.g., sequence A and sequence B, is calculated by dividing the length of sequence A, minus the number of gap residues in sequence A, minus the number of gap residues in sequence B, into the sum of the residue matches between sequence A and sequence B, times one hundred. Gaps of low or of no similarity between the two amino acid sequences are not included in determining percentage similarity. Percent identity between polynucleotide sequences can also be counted or calculated by other methods known in the art, e.g., the Jotun Hein method (for example, Hein (1990) Methods Enzymol. 183: 626-645). Identity between sequences can also be determined by other methods known in the art, e.g., by varying hybridization conditions (US Patent Application No. 20010010913). Techniques for alignment are described in Methods in Enzymology, vol. 266, Computer Methods for Macromolecular Sequence Analysis (1996), ed. Doolittle, Academic Press, Inc., San Diego, Calif, USA. Preferably, an alignment program that permits gaps in the sequence is used to align the sequences. The Smith-Waterman is one type of algorithm that permits gaps in sequence alignments (Shpaer (1997) Methods Mol. Biol. 70: 173-187). Also, the GAP program using the Needleman and Wunsch alignment method can be utilized to align sequences. An alternative search strategy uses MPSRCH software, which runs on a MASPAR computer. MPSRCH uses a Smith-Waterman algorithm to score sequences on a massively parallel computer. This approach improves ability to pick up distantly related matches, and is especially tolerant of small gaps and nucleotide sequence errors. Nucleic acid-encoded amino acid sequences can be used to search both protein and DNA databases. Thus, the invention provides methods for identifying a sequence similar or paralogous or orthologous or homologous to one or more polynucleotides as noted herein, or one or more target polypeptides encoded by the polynucleotides, or otherwise noted herein and may include linking or associating a given plant phenotype or gene function with a sequence. In the methods, a sequence database is provided (locally or across an internet or intranet) and a query is made against the sequence database using the relevant sequences herein and associated plant phenotypes or gene functions. In addition, one or more polynucleotide sequences or one or more polypeptides encoded by the polynucleotide sequences may be used to search against a BLOCKS (Bairoch et al. (1997) Nucleic Acids Res. 25: 217-221), PFAM, and other databases which contain previously identified and annotated motifs, sequences and gene functions. Methods that search for primary sequence patterns with secondary structure gap penalties (Smith et al. (1992) Protein Engineering 5: 35-51) as well as algorithms such as Basic Local Alignment Search Tool (BLAST; Altschul (1993) J. Mol. Evol. 36: 290-300; Altschul et al. (1990) J. Mol. Biol. 215: 403-410), BLOCKS (Henikoff and Henikoff (1991) Nucleic Acids Res. 19: 6565-6572), Hidden Markov Models (HMM; Eddy (1996) Curr. Opin. Str. Biol. 6: 361-365; Sonnhammer et al. (1997) Proteins 28: 405-420), and the like, can be used to manipulate and analyze polynucleotide and polypeptide sequences encoded by polynucleotides. These databases, algorithms and other methods are well known in the art and are described in Ausubel et al. (1997; Short Protocols in Molecular Biology. John Wiley & Sons, New York, NY, unit 7.7) and in Meyers (1995; Molecular Biology and Biotechnology. Wiley VCH, New York, NY, p 856-853). A further method for identifying or confirming that specific homologous sequences control the same function is by comparison of the transcript profile(s) obtained upon overexpression or knockout of two or more related transcription factors. Since transcript profiles are diagnostic for specific cellular states, one skilled in the art will appreciate that genes that have a highly similar transcript profile (e.g., with greater than 50% regulated transcripts in common, more preferably with greater than 70% regulated transcripts in common, most preferably with greater than 90% regulated transcripts in common) will have highly similar functions. Fowler et al. ((2002) Plant Cell 14: 1675-1679) have shown that three paralogous AP2 family genes (CBFl, CBF2 and CBF3), each of which is induced upon cold treatment, and each of which can condition improved freezing tolerance, have highly similar transcript profiles. Once a transcription factor has been shown to provide a specific function, its transcript profile becomes a diagnostic tool to determine whether putative paralogs or orthologs have the same function. Furthermore, methods using manual alignment of sequences similar or homologous to one or more polynucleotide sequences or one or more polypeptides encoded by the polynucleotide sequences may be used to identify regions of similarity and AP2 binding domains. Such manual methods are well known of those of skill in the art and can include, for example, comparisons of tertiary structure between a polypeptide sequence encoded by a polynucleotide that comprises a known function, with a polypeptide sequence encoded by a polynucleotide sequence which has a function not yet determined. Such examples of tertiary structure may comprise predicted α-helices, β-sheets, amphipathic helices, leucine zipper motifs, zinc finger motifs, proline-rich regions, cysteine repeat motifs, and the like. Orthologs and paralogs of presently disclosed transcription factors may be cloned using compositions provided by the present invention according to methods well known in the art. cDNAs can be cloned using mRNA from a plant cell or tissue that expresses one of the present transcription factors. Appropriate mRNA sources may be identified by interrogating Northern blots with probes designed from the present transcription factor sequences, after which a library is prepared from the mRNA obtained from a positive cell or tissue. Transcription factor-encoding cDNA is then isolated using an amplification method, for example, PCR, with primers designed from a presently disclosed transcription factor gene sequence, or by probing with a partial or complete cDNA or with one or more sets of degenerate probes based on the disclosed sequences. The cDNA library may be used to transform plant cells. Expression of the cDNAs of interest is detected using, for example, methods disclosed herein such as microarrays,

Northern blots, quantitative PCR, or any other technique for monitoring changes in expression. Genomic clones may be isolated using similar techniques to those.

Identifying Polynucleotides or Nucleic Acids by Hybridization Polynucleotides homologous to the sequences illustrated in the Sequence Listing and tables can be identified, e.g., by hybridization to each other under stringent conditions. Single-stranded polynucleotides hybridize when they associate based on a variety of well characterized physical-chemical forces, such as hydrogen bonding, solvent exclusion, base stacking and the like. The stringency of a hybridization reflects the degree of sequence identity of the nucleic acids involved, such that the higher the stringency, the more similar are the two polynucleotide strands. Stringency is influenced by a variety of factors, including temperature, salt concentration and composition, organic and non-organic additives, solvents, etc., present in both the hybridization and wash solutions and incubations (and number thereof), as described in more detail in the references cited above. Encompassed by the invention are polynucleotide sequences that are capable of hybridizing to the claimed polynucleotide sequences, including any of the transcription factor polynucleotides within the Sequence Listing, and fragments thereof under various conditions of stringency (for example, Wahl and Berger (1987) Methods Enzymol. 152: 399-407; and Kimmel (1987) Methods Enzymol. 152: 507-51 1). In addition to the nucleotide sequences in the Sequence Listing, full-length cDNA, orthologs, and paralogs of the present nucleotide sequences may be identified and isolated using well-known methods. The cDNA libraries, orthologs, and paralogs of the present nucleotide sequences may be screened using hybridization methods to determine their utility as hybridization target or amplification probes. With regard to hybridization, conditions that are highly stringent, and means for achieving them, are well known in the art (for example, Sambrook et al. (1989) "Molecular Cloning: A Laboratory Manuar (2nd ed., Cold Spring Harbor Laboratory); Berger and Kimmel, eds., (1987) "Guide to Molecular Cloning Techniques", in Methods in Enzymology: 152: 467-469; and Anderson and Young (1985) "Quantitative Filter Hybridisation", in Hames and Higgins, ed., Nucleic Acid Hybridisation, A Practical Approach. Oxford, IRL Press, 73-111 ). Stability of DNA duplexes is affected by such factors as base composition, length, and degree of base pair mismatch. Hybridization conditions may be adjusted to allow DNAs of different sequence relatedness to hybridize. The melting temperature (T_m) is defined as the temperature when 50% of the duplex molecules have dissociated into their constituent single strands. The melting temperature of a perfectly matched duplex, where the hybridization buffer contains formamide as a denaturing agent, may be estimated by the following equations:

(I) DNA-DNA: T_m(° C)=81.5+16.6(log [Na+])+0.41(% G+C)- 0.62(% formamide)-500/L

(II) DNA-RNA:

T_m(° C)=79.8+18.5(log [Na+])+0.58(% G+C)+ 0.12(%G+C)²- 0.5(% formamide) - 820/L

(III) RNA-RNA: „(° C)=79.8+18.5(log [Na+])+0.58(% G+Q+ 0.12(%G+C)²- 0.35(% formamide) - 820/L where L is the length of the duplex formed, [Na+] is the molar concentration of the sodium ion in the hybridization or washing solution, and % G+C is the percentage of (guanine+cytosine) bases in the hybrid. For imperfectly matched hybrids, approximately 1° C is required to reduce the melting temperature for each 1% mismatch. Hybridization experiments are generally conducted in a buffer of pH between 6.8 to 7.4, although the rate of hybridization is nearly independent of pH at ionic strengths likely to be used in the hybridization buffer (Anderson and Young (1985) supra). In addition, one or more of the following may be used to reduce non-specific hybridization: sonicated salmon sperm DNA or another non- complementary DNA, bovine serum albumin, sodium pyrophosphate, sodium dodecylsulfate (SDS), polyvinyl-pyrrolidone, ficoll and Denhardt's solution. Dextran sulfate and polyethylene glycol 6000 act to exclude DNA from solution, thus raising the effective probe DNA concentration and the hybridization signal within a given unit of time. In some instances, conditions of even greater stringency may be desirable or required to reduce non-specific and/or background hybridization. These conditions may be created with the use of higher temperature, lower ionic strength and higher concentration of a denaturing agent such as formamide. Stringency conditions can be adjusted to screen for moderately similar fragments such as homologous sequences from distantly-related organisms, or to highly similar fragments such as genes that duplicate functional enzymes from closely-related organisms. The stringency can be adjusted either during the hybridization step or in the post-hybridization washes. Salt concentration, formamide concentration, hybridization temperature and probe lengths are variables that can be used to alter stringency (as described by the formula above). As a general guideline, high stringency is typically performed at T_m-5° C to T_m-20° C, moderate stringency at T_m-20° C to T_m-35° C and low stringency at T_m- 35° C to T_m-50° C for duplex >150 base pairs. Hybridization may be performed at low to moderate stringency (25-50° C below T_m), followed by post-hybridization washes at increasing stringencies. Maximum rates of hybridization in solution are determined empirically to occur at T_m-25° C for DNA- DNA duplex and T_m-15° C for RNA-DNA duplex. Optionally, the degree of dissociation may be assessed after each wash step to determine the need for subsequent, higher stringency wash steps. High stringency conditions may be used to select for nucleic acid sequences with high degrees of identity to the disclosed sequences. An example of stringent hybridization conditions obtained in a filter- based method such as a Southern or northern blot for hybridization of complementary nucleic acids that have more than 100 complementary residues is about 5°C to 20°C lower than the thermal melting point (T_m) for the specific sequence at a defined ionic strength and pH. Conditions used for hybridization may include about 0.02 M to about 0.15 M sodium chloride, about 0.5% to about 5% casein, about 0.02%> SDS or about 0.1% N-laurylsarcosine, about 0.001 M to about 0.03 M sodium citrate, at hybridization temperatures between about 50° C and about 70° C. More preferably, high stringency conditions are about 0.02 M sodium chloride, about 0.5% casein, about 0.02% SDS, about 0.001 M sodium citrate, at a temperature of about 50° C. Nucleic acid molecules that hybridize under stringent conditions will typically hybridize to a probe based on either the entire DNA molecule or selected portions, e.g., to a unique subsequence, of the DNA. Stringent salt concentration will ordinarily be less than about 750 mM NaCl and 75 mM trisodium citrate. Increasingly stringent conditions may be obtained with less than about 500 mM NaCl and 50 mM trisodium citrate, to even greater stringency with less than about 250 mM NaCl and 25 mM trisodium citrate. Low stringency hybridization can be obtained in the absence of organic solvent, e.g., formamide, whereas high stringency hybridization may be obtained in the presence of at least about 35% formamide, and more preferably at least about 50% formamide. Stringent temperature conditions will ordinarily include temperatures of at least about 30° C, more preferably of at least about 37° C, and most preferably of at least about 42° C with formamide present. Varying additional parameters, such as hybridization time, the concentration of detergent, e.g., sodium dodecyl sulfate (SDS) and ionic strength, are well known to those skilled in the art. Various levels of stringency are accomplished by combining these various conditions as needed. The washing steps that follow hybridization may also vary in stringency; the post-hybridization wash steps primarily determine hybridization specificity, with the most critical factors being temperature and the ionic strength of the final wash solution. Wash stringency can be increased by decreasing salt concentration or by increasing temperature. Stringent salt concentration for the wash steps will preferably be less than about 30 mM NaCl and 3 mM trisodium citrate, and most preferably less than about 15 mM NaCl and 1.5 mM trisodium citrate. Thus, hybridization and wash conditions that may be used to bind and remove polynucleotides with less than the desired homology to the nucleic acid sequences or their complements that encode the present transcription factors include, for example: 6X SSC at 65° C; 50% formamide, 4X SSC at 42° C; or 0.5X SSC, 0.1% SDS at 65° C; with, for example, two wash steps of 10 - 30 minutes each. Useful variations on these conditions will be readily apparent to those skilled in the art. A person of skill in the art would not expect substantial variation among polynucleotide species encompassed within the scope of the present invention because the highly stringent conditions set forth in the above formulae yield structurally similar polynucleotides. If desired, one may employ wash steps of even greater stringency, including about 0.2X SSC, 0.1% SDS at 65° C and washing twice, each wash step being about 30 min, or about 0.1 X SSC, 0.1% SDS at 65° C and washing twice for 30 min. The temperature for the wash solutions will ordinarily be at least about 25° C, and for greater stringency at least about 42° C. Hybridization stringency may be increased further by using the same conditions as in the hybridization steps, with the wash temperature raised about 3° C to about 5° C, and stringency may be increased even further by using the same conditions except the wash temperature is raised about 6° C to about 9° C. For identification of less closely related homologs, wash steps may be performed at a lower temperature, e.g., 50° C. An example of a low stringency wash step employs a solution and conditions of at least 25° C in 30 mM NaCl, 3 mM trisodium citrate, and 0.1% SDS over 30 min. Greater stringency may be obtained at 42° C in 15 mM NaCl, with 1.5 mM trisodium citrate, and 0.1% SDS over 30 min. Even higher stringency wash conditions are obtained at 65° C -68° C in a solution of 15 mM NaCl, 1.5 mM trisodium citrate, and 0.1% SDS. Wash procedures will generally employ at least two final wash steps. Additional variations on these conditions will be readily apparent to those skilled in the art (for example, US Patent Application No. 20010010913). Stringency conditions can be selected such that an oligonucleotide that is perfectly complementary to the coding oligonucleotide hybridizes to the coding oligonucleotide with at least about a 5-1 Ox higher signal to noise ratio than the ratio for hybridization of the perfectly complementary oligonucleotide to a nucleic acid encoding a transcription factor known as of the filing date of the application. It may be desirable to select conditions for a particular assay such that a higher signal to noise ratio, that is, about 15x or more, is obtained. Accordingly, a subject nucleic acid will hybridize to a unique coding oligonucleotide with at least a 2x or greater signal to noise ratio as compared to hybridization of the coding oligonucleotide to a nucleic acid encoding known polypeptide. The particular signal will depend on the label used in the relevant assay, e.g., a fluorescent label, a colorimetric label, a radioactive label, or the like. Labeled hybridization or PCR probes for detecting related polynucleotide sequences may be produced by oligolabeling, nick translation, end-labeling, or PCR amplification using a labeled nucleotide. Encompassed by the invention are polynucleotide sequences capable of hybridizing to the claimed polynucleotide sequences, and, in particular, to those shown in SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, and fragments thereof under various conditions of stringency, (for example, Wahl and Berger (1987) Methods Enzymol. 152: 399-407; Kimmel (1987) Methods Enzymol. 152: 507-511). Estimates of homology are provided by either DNA- DNA or DNA-RNA hybridization under conditions of stringency as is well understood by those skilled in the art (Hames and Higgins, Eds. (1985) Nucleic Acid Hybridisation, IRL Press, Oxford, U.K.). Stringency conditions can be adjusted to screen for moderately similar fragments, such as homologous sequences from distantly related organisms, to highly similar fragments, such as genes that duplicate functional enzymes from closely related organisms. Post-hybridization washes determine stringency conditions.

Identifying Polynucleotides or Nucleic Acids with Expression Libraries In addition to hybridization methods, transcription factor homolog polypeptides can be obtained by screening an expression library using antibodies specific for one or more transcription factors. With the provision herein of the disclosed transcription factor, and transcription factor homolog nucleic acid sequences, the encoded polypeptide(s) can be expressed and purified in a heterologous expression system (e.g., E. coli) and used to raise antibodies (monoclonal or polyclonal) specific for the polypeptide(s) in question. Antibodies can also be raised against synthetic peptides derived from the sequences of transcription factors or homologous sequences. Methods of raising antibodies are well known in the art and are described in Harlow and Lane (1988), Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, New York. Such antibodies can then be used to screen an expression library produced from the plant from which it is desired to clone additional transcription factor homologs, using the methods described above. The selected cDNAs can be confirmed by sequencing and enzymatic activity.

Sequence Variations It will readily be appreciated by those of skill in the art, that any of a variety of polynucleotide sequences are capable of encoding the transcription factors and transcription factor homolog polypeptides of the invention. Due to the degeneracy of the genetic code, many different polynucleotides can encode identical and/or substantially similar polypeptides in addition to those sequences illustrated in the Sequence Listing. Nucleic acids having a sequence that differs from the sequences shown in the Sequence Listing, or complementary sequences, that encode functionally equivalent peptides (i.e., peptides having some degree of equivalent or similar biological activity) but differ in sequence from the sequence shown in the Sequence Listing due to degeneracy in the genetic code, are also within the scope of the invention. Altered polynucleotide sequences encoding polypeptides include sequences with deletions, insertions, or substitutions of different nucleotides, resulting in a polynucleotide encoding a polypeptide with at least one functional characteristic of the instant polypeptides. Included within this definition are polymorphisms which may or may not be readily detectable using a particular oligonucleotide probe of the polynucleotide encoding the instant polypeptides, and improper or unexpected hybridization to allelic variants, with a locus other than the normal chromosomal locus for the polynucleotide sequence encoding the instant polypeptides. Allelic variant refers to any of two or more alternative forms of a gene occupying the same chromosomal locus. Allelic variation arises naturally through mutation, and may result in phenotypic polymorphism within populations. Gene mutations can be silent (i.e., no change in the encoded polypeptide) or may encode polypeptides having altered amino acid sequence. The term allelic variant is also used herein to denote a protein encoded by an allelic variant of a gene. Splice variant refers to alternative fonns of RNA transcribed from a gene. Splice variation arises naturally through use of alternative splicing sites within a transcribed RNA molecule, or less commonly between separately transcribed RNA molecules, and may result in several mRNAs transcribed from the same gene. Splice variants may encode polypeptides having altered amino acid sequence. The term splice variant is also used herein to denote a protein encoded by a splice variant of an mRNA transcribed from a gene. Those skilled in the art would recognize that, for example, G867, SEQ ID NO: 2, represents a single transcription factor; allelic variation and alternative splicing may be expected to occur. Allelic variants of SEQ ID NO: 1 can be cloned by probing cDNA or genomic libraries from different individual organisms according to standard procedures. Allelic variants of the DNA sequence shown in SEQ ID NO: 1, including those containing silent mutations and those in which mutations result in amino acid sequence changes, are within the scope of the present invention, as are proteins which are allelic variants of SEQ ID NO: 2. cDNAs generated from alternatively spliced mRNAs, which retain the properties of the transcription factor are included within the scope of the present invention, as are polypeptides encoded by such cDNAs and mRNAs. Allelic variants and splice variants of these sequences can be cloned by probing cDNA or genomic libraries from different individual organisms or tissues according to standard procedures known in the art (for example, USPN 6,388,064). Thus, in addition to the sequences set forth in the Sequence Listing, the invention also encompasses related nucleic acid molecules that include allelic or splice variants of SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, and include sequences which are complementary to any of the above nucleotide sequences. Related nucleic acid molecules also include nucleotide sequences encoding a polypeptide comprising a substitution, modification, addition and/or deletion of one or more amino acid residues compared to the polypeptide as set forth in any of SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, or 53. Such related polypeptides may comprise, for example, additions and/or deletions of one or more N-linked or O-linked glycosylation sites, or an addition and/or a deletion of one or more cysteine residues. For example, Table 3 illustrates, e.g., that the codons AGC, AGT, TCA, TCC, TCG, and TCT all encode the same amino acid: serine. Accordingly, at each position in the sequence where there is a codon encoding serine, any of the above trinucleotide sequences can be used without altering the encoded polypeptide.

Table 3 Amino acid Possible Codons Alanine Ala A GCA GCC GCG GCT Cysteine Cys C TGC TGT Aspartic acid Asp D GAC GAT Glutamic acid Glu E GAA GAG Phenylalanine Phe F TTC TTT Glycine Gly G GGA GGC GGG GGT Histidine His H CAC CAT Isoleucine He I ATA ATC ATT Lysine Lys K AAA AAG Leucine Leu L TTA TTG CTA CTC CTG CTT Methionine Met M ATG Asparagine Asn N AAC AAT Proline Pro P CCA CCC CCG CCT Glutamine Gin Q CAA CAG Arginine Arg R AGA AGG CGA CGC CGG CGT Serine Ser S AGC AGT TCA TCC TCG TCT Threonine Thr T ACA ACC ACG ACT Valine Val V GTA GTC GTG GTT Tryptophan Trp w TGG Tyrosine Tyr Y TAC TAT Sequence alterations that do not change the amino acid sequence encoded by the polynucleotide are termed "silent" variations. With the exception of the codons ATG and TGG, encoding methionine and tryptophan, respectively, any of the possible codons for the same amino acid can be substituted by a variety of techniques, e.g., site-directed mutagenesis, available in the art. Accordingly, any and all such variations of a sequence selected from the above table are a feature of the invention. In addition to silent variations, other conservative variations that alter one, or a few amino acid residues in the encoded polypeptide, can be made without altering the function of the polypeptide, these conservative variants are, likewise, a feature of the invention. For example, substitutions, deletions and insertions introduced into the sequences provided in the Sequence Listing, are also envisioned by the invention. Such sequence modifications can be engineered into a sequence by site-directed mutagenesis (Wu (ed.) Methods Enzymol. (1993) vol. 217, Academic Press) or the other methods noted below. Amino acid substitutions are typically of single residues; insertions usually will be on the order of about from 1 to 10 amino acid residues; and deletions will range about from 1 to 30 residues. In some embodiments, deletions or insertions are made in adjacent pairs, e.g., a deletion of two residues or insertion of two residues. Substitutions, deletions, insertions or any combination thereof can be combined to arrive at a sequence. The mutations that are made in the polynucleotide encoding the transcription factor should not place the sequence out of reading frame and should not create complementary regions that could produce secondary mRNA structure. Preferably, the polypeptide encoded by the DNA performs the desired function. Conservative substitutions are those in which at least one residue in the amino acid sequence has been removed and a different residue inserted in its place. Such substitutions generally are made in accordance with the Table 4 when it is desired to maintain the activity of the protein. Table 4 shows amino acids which can be substituted for an amino acid in a protein and which are typically regarded as conservative substitutions.

Table 4

Similar substitutions are those in which at least one residue in the amino acid sequence has been removed and a different residue inserted in its place. Such substitutions generally are made in accordance with the Table 5 when it is desired to maintain the activity of the protein. Table 5 shows amino acids which can be substituted for an amino acid in a protein and which are typically regarded as structural and functional substitutions. For example, a residue in column 1 of Table 5 may be substituted with a residue in column 2; in addition, a residue in column 2 of Table 5 may be substituted with the residue of column 1.

Table 5

Substitutions that are less conservative than those in Table 5 can be selected by picking residues that differ more significantly in their effect on maintaining (a) the structure of the polypeptide backbone in the area of the substitution, for example, as a sheet or helical conformation, (b) the charge or hydrophobicity of the molecule at the target site, or (c) the bulk of the side chain. The substitutions which in general are expected to produce the greatest changes in protein properties will be those in which (a) a hydrophilic residue, e.g., seryl or threonyl, is substituted for (or by) a hydrophobic residue, e.g., leucyl, isoleucyl, phenylalanyl, valyl or alanyl; (b) a cysteine or proline is substituted for (or by) any other residue; (c) a residue having an electropositive side chain, e.g., lysyl, arginyl, or histidyl, is substituted for (or by) an electronegative residue, e.g., glutamyl or aspartyl; or (d) a residue having a bulky side chain, e.g., phenylalanine, is substituted for (or by) one not having a side chain, e.g., glycine.

Further Modifying Sequences of the Invention - Mutation/Forced Evolution In addition to generating silent or conservative substitutions as noted, above, the present invention optionally includes methods of modifying the sequences of the Sequence Listing. In the methods, nucleic acid or protein modification methods are used to alter the given sequences to produce new sequences and/or to chemically or enzymatically modify given sequences to change the properties of the nucleic acids or proteins. Thus, in one embodiment, given nucleic acid sequences are modified, e.g., according to standard mutagenesis or artificial evolution methods to produce modified sequences. The modified sequences may be created using purified natural polynucleotides isolated from any organism or may be synthesized from purified compositions and chemicals using chemical means well known to those of skill in the art. For example, Ausubel, supra, provides additional details on mutagenesis methods. Artificial forced evolution methods are described, for example, by Stemmer (1994) Nature 370: 389-391, Stemmer (1994) Proc. Natl. Acad. Sci. 91: 10747-10751, and US Patents 5,811,238, 5,837,500, and 6,242,568. Methods for engineering synthetic transcription factors and other polypeptides are described, for example, by Zhang et al. (2000) J. Biol. Chem. 275: 33850-33860, Liu et al. (2001) J. Biol. Chem. 276: 11323-1 1334, and Isalan et al. (2001) Nature Biotechnol. 19: 656-660. Many other mutation and evolution methods are also available and expected to be within the skill of the practitioner. Similarly, chemical or enzymatic alteration of expressed nucleic acids and polypeptides can be performed by standard methods. For example, sequence can be modified by addition of lipids, sugars, peptides, organic or inorganic compounds, by the inclusion of modified nucleotides or amino acids, or the like. For example, protein modification techniques are illustrated in Ausubel, supra. Further details on chemical and enzymatic modifications can be found herein. These modification methods can be used to modify any given sequence, or to modify any sequence produced by the various mutation and artificial evolution modification methods noted herein. Accordingly, the invention provides for modification of any given nucleic acid by mutation, evolution, chemical or enzymatic modification, or other available methods, as well as for the products produced by practicing such methods, e.g., using the sequences herein as a starting substrate for the various modification approaches. For example, optimized coding sequence containing codons preferred by a particular prokaryotic or eukaryotic host can be used e.g., to increase the rate of translation or to produce recombinant RNA transcripts having desirable properties, such as a longer half-life, as compared with transcripts produced using a non-optimized sequence. Translation stop codons can also be modified to reflect host preference. For example, preferred stop codons for Saccharomyces cerevisiae and mammals are TAA and TGA, respectively. The preferred stop codon for monocotyledonous plants is TGA, whereas insects and E. coli prefer to use TAA as the stop codon. The polynucleotide sequences of the present invention can also be engineered in order to alter a coding sequence for a variety of reasons, including but not limited to, alterations which modify the sequence to facilitate cloning, processing and/or expression of the gene product. For example, alterations are optionally introduced using techniques which are well known in the art, e.g., site-directed mutagenesis, to insert new restriction sites, to alter glycosylation patterns, to change codon preference, to introduce splice sites, etc. Furthermore, a fragment or domain derived from any of the polypeptides of the invention can be combined with domains derived from other transcription factors or synthetic domains to modify the biological activity of a transcription factor. For instance, a DNA-binding domain derived from a transcription factor of the invention can be combined with the activation domain of another transcription factor or with a synthetic activation domain. A transcription activation domain assists in initiating transcription from a DNA-binding site. Examples include the transcription activation region of VP 16 or GAL4 (Moore et al. (1998) Proc. Natl. Acad. Sci. 95: 376-381; Aoyama et al. (1995) Plant Cell 7: 1773- 1785), peptides derived from bacterial sequences (Ma and Ptashne (1987) Cell 51: 113-119) and synthetic peptides (Giniger and Ptashne (1987) Nature 330: 670-672).

Expression and Modification of Polypeptides Typically, polynucleotide sequences of the invention are incorporated into recombinant DNA (or RNA) molecules that direct expression of polypeptides of the invention in appropriate host cells, transgenic plants, in vitro translation systems, or the like. Due to the inherent degeneracy of the genetic code, nucleic acid sequences which encode substantially the same or a functionally equivalent amino acid sequence can be substituted for any listed sequence to provide for cloning and expressing the relevant homolog. The transgenic plants of the present invention comprising recombinant polynucleotide sequences are generally derived from parental plants, which may themselves be non-transformed (or non-transgenic) plants. These transgenic plants may either have a transcription factor gene "knocked out" (for example, with a genomic insertion by homologous recombination, an antisense or ribozyme construct) or expressed to a normal or wild-type extent. However, overexpressing transgenic "progeny" plants will exhibit greater mRNA levels, wherein the mRNA encodes a transcription factor, that is, a DNA-binding protein that is capable of binding to a DNA regulatory sequence and inducing transcription, and preferably, expression of a plant trait gene. Preferably, the mRNA expression level will be at least three-fold greater than that of the parental plant, or more preferably at least ten-fold greater mRNA levels compared to said parental plant, and most preferably at least fifty-fold greater compared to said parental plant.

Vectors. Promoters, and Expression Systems The present invention includes recombinant constructs comprising one or more of the nucleic acid sequences herein. The constructs typically comprise a vector, such as a plasmid, a cosmid, a phage, a virus (e.g., a plant virus), a bacterial artificial chromosome (BAC), a yeast artificial chromosome (YAC), or the like, into which a nucleic acid sequence of the invention has been inserted, in a forward or reverse orientation. In a preferred aspect of this embodiment, the construct further comprises regulatory sequences, including, for example, a promoter, operably linked to the sequence. Large numbers of suitable vectors and promoters are known to those of skill in the art, and are commercially available. General texts that describe molecular biological techniques useful herein, including the use and production of vectors, promoters and many other relevant topics, include Berger, Sambrook, supra and Ausubel, supra. Any of the identified sequences can be incorporated into a cassette or vector, e.g., for expression in plants. A number of expression vectors suitable for stable transformation of plant cells or for the establishment of transgenic plants have been described including those described in Weissbach and Weissbach (1989) Methods for Plant Molecular Biology, Academic Press, and Gelvin et al. (1990) Plant Molecular Biology Manual, Kluwer Academic Publishers. Specific examples include those derived from a Ti plasmid of Agrobacterium tumefaciens, as well as those disclosed by Herrera-Estrella et al. (1983) Nature 303: 209, Bevan (1984) Nucleic Acids Res. 12: 8711-8721, Klee ( 1985) Bio/Technology 3: 637-642, for dicotyledonous plants. Alternatively, non-Ti vectors can be used to transfer the DNA into monocotyledonous plants and cells by using free DNA delivery techniques. Such methods can involve, for example, the use of liposomes, electroporation, microprojectile bombardment, silicon carbide whiskers, and viruses. By using these methods transgenic plants such as wheat, rice (Christou (1991) Bio/Technology 9: 957-962) and corn (Gordon-Kamm (1990) Plant Cell 2: 603-618) can be produced. An immature embryo can also be a good target tissue for monocots for direct DNA delivery techniques by using the particle gun (Weeks et al. (1993) Plant Physiol. 102: 1077-1084; Vasil (1993) Bio/Technology 10: 667-674; Wan and Lemeaux (1994) Plant Physiol. 104: 37-48, and for Agrobacterium-mediated DNA transfer (Ishida et al. (1996) Nature Biotechnol. 14: 745-750). Typically, plant transformation vectors include one or more cloned plant coding sequence (genomic or cDNA) under the transcriptional control of 5' and 3' regulatory sequences and a dominant selectable marker. Such plant transformation vectors typically also contain a promoter (e.g., a regulatory region controlling inducible or constitutive, environmentally-or developmentally-regulated, or cell- or tissue-specific expression), a transcription initiation start site, an RNA processing signal (such as intron splice sites), a transcription termination site, and/or a polyadenylation signal. A potential utility for the transcription factor polynucleotides disclosed herein is the isolation of promoter elements from these genes that can be used to program expression in plants of any genes. Each transcription factor gene disclosed herein is expressed in a unique fashion, as determined by promoter elements located upstream of the start of translation, and additionally within an intron of the transcription factor gene or downstream of the termination codon of the gene. As is well known in the art, for a significant portion of genes, the promoter sequences are located entirely in the region directly upstream of the start of translation. In such cases, typically the promoter sequences are located within 2.0 kb of the start of translation, or within 1.5 kb of the start of translation, frequently within 1.0 kb of the start of translation, and sometimes within 0.5 kb of the start of translation. The promoter sequences can be isolated according to methods known to one skilled in the art. Examples of constitutive plant promoters which can be useful for expressing the TF sequence include: the cauliflower mosaic virus (CaMV) 35S promoter, which confers constitutive, high-level expression in most plant tissues (for example, Odell et al. (1985) Nature 313: 810-812); the nopaline synthase promoter (An et al. (1988) Plant Physiol. 88: 547-552); and the octopine synthase promoter (Fromm et al. (1989) Plant Cell 1 : 977-984). The transcription factors of the invention may be operably linked with a specific promoter that causes the transcription factor to be expressed in response to environmental, tissue-specific or temporal signals. A variety of plant gene promoters that regulate gene expression in response to environmental, hormonal, chemical, developmental signals, and in a tissue-active manner can be used for expression of a TF sequence in plants. Choice of a promoter is based largely on the phenotype of interest and is determined by such factors as tissue (e.g., seed, fruit, root, pollen, vascular tissue, flower, carpel, etc.), inducibility (e.g., in response to drought, wounding, heat, cold, light, pathogens, etc.), timing, developmental stage, and the like. Numerous known promoters have been characterized and can favorably be employed to promote expression of a polynucleotide of the invention in a transgenic plant or cell of interest. For example, tissue-specific promoters include: seed-specific promoters (such as the napin, phaseolin or DC3 promoter described in US Pat. No. 5,773,697), fruit-specific promoters that are active during fruit ripening (such as the dru 1 promoter (US Pat. No. 5,783,393), or the 2A11 promoter (US Pat. No. 4,943,674) and the tomato polygalacturonase promoter (Bird et al. (1988) Plant Mol. Biol. 11 : 651 -662), root-specific promoters, such as those disclosed in US Patent Nos. 5,618,988, 5,837,848 and 5,905,186, pollen-active promoters such as PTA29, PTA26 and PTA13 (US Pat. No. 5,792,929), promoters active in vascular tissue (Ringli and Keller (1998) Plant Mol. Biol. 37: 977-988), flower- specific (Kaiser et al. (1995) Plant Mol. Biol. 28: 231-243), pollen (Baerson et al. (1994) Plant Mol. Biol. 26: 1947-1959), carpels (Ohl et al. (1990) Plant Cell 2: 837-848), pollen and ovules (Baerson et al.

(1993) Plant Mol. Biol. 22: 255-267), auxin-inducible promoters (such as that described in van der Kop et al. (1999) Plant Mol. Biol. 39: 979-990 or Baumann et al., (1999) Plant Cell 11 : 323-334), cytokinin- inducible promoter (Guevara-Garcia (1998) Plant Mol. Biol. 38: 743-753), promoters responsive to gibberellin (Shi et al. (1998) Plant Mol. Biol. 38: 1053-1060, Willmott et al. (1998) Plant Molec. Biol. 38: 817-825) and the like. Additional promoters are those that elicit expression in response to heat (Ainley et al. (1993) Plant Mol. Biol. 22: 13-23), light (e.g., the pea rbcS-3A promoter, Kuhlemeier et al. (1989) Plant Cell 1 : 471-478, and the maize rbcS promoter, Schaffner and Sheen (1991) Plant Cell 3: 997-1012); wounding (e.g., wunl, Siebertz et al. (1989) Plant Cell 1 : 961-968); pathogens (such as the PR-1 promoter described in Buchel et al. (1999) Plant Mol. Biol. 40: 387-396, and the PDF1.2 promoter described in Manners et al. (1998) Plant Mol. Biol. 38: 1071-1080), and chemicals such as methyl jasmonate or salicylic acid (Gatz (1997) Annu. Rev. Plant Physiol. Plant Mol. Biol. 48: 89-108). In addition, the timing of the expression can be controlled by using promoters such as those acting at senescence (Gan and Amasino (1995) Science 270: 1986-1988); or late seed development (Odell et al. (1994) Plant Physiol. 106: 447-458). Plant expression vectors can also include RNA processing signals that can be positioned within, upstream or downstream of the coding sequence. In addition, the expression vectors can include additional regulatory sequences from the 3 '-untranslated region of plant genes, e.g., a 3' terminator region to increase mRNA stability of the mRNA, such as the PI-II terminator region of potato or the octopine or nopaline synthase 3' terminator regions.

Additional Expression Elements Specific initiation signals can aid in efficient translation of coding sequences. These signals can include, e.g., the ATG initiation codon and adjacent sequences. In cases where a coding sequence, its initiation codon, and upstream sequences are inserted into the appropriate expression vector, no additional translational control signals may be needed. However, in cases where only coding sequence (e.g., a mature protein coding sequence) or a portion thereof is inserted, exogenous transcriptional control signals including the ATG initiation codon can be separately provided. The initiation codon is provided in the correct reading frame to facilitate transcription. Exogenous transcriptional elements and initiation codons can be of various origins, both natural and synthetic. The efficiency of expression can be enhanced by the inclusion of enhancers appropriate to the cell system in use.

Expression Hosts The present invention also relates to host cells which are transduced with vectors of the invention, and the production of polypeptides of the invention (including fragments thereof) by recombinant techniques. Host cells are genetically engineered (i.e., nucleic acids are introduced, e.g., transduced, transformed or transfected) with the vectors of this invention, which may be, for example, a cloning vector or an expression vector comprising the relevant nucleic acids herein. The vector is optionally a plasmid, a viral particle, a phage, a naked nucleic acid, etc. The engineered host cells can be cultured in conventional nutrient media modified as appropriate for activating promoters, selecting transformants, or amplifying the relevant gene. The culture conditions, such as temperature, pH and the like, are those previously used with the host cell selected for expression, and will be apparent to those skilled in the art and in the references cited herein, including, Sambrook, supra and Ausubel, supra. The host cell can be a eukaryotic cell, such as a yeast cell, or a plant cell, or the host cell can be a prokaryotic cell, such as a bacterial cell. Plant protoplasts are also suitable for some applications. For example, the DNA fragments are introduced into plant tissues, cultured plant cells or plant protoplasts by standard methods including electroporation (Fromm et al. (1985) Proc. Natl. Acad. Sci. 82: 5824-5828), infection by viral vectors such as cauliflower mosaic virus (CaMV) (Hohn et al. (1982) Molecular

Biology of Plant Tumors Academic Press, New York, NY, pp. 549-560; US 4,407,956), high velocity ballistic penetration by small particles with the nucleic acid either within the matrix of small beads or particles, or on the surface (Klein et al. (1987) Nature 327: 70-73), use of pollen as vector (WO 85/01856), or use of Agrobacterium tumefaciens or A. rhizogenes carrying a T-DNA plasmid in which DNA fragments are cloned. The T-DNA plasmid is transmitted to plant cells upon infection by

Agrobacterium tumefaciens, and a portion is stably integrated into the plant genome (Horsch et al. (1984) Science 233: 496-498; Fraley et al. (1983) Proc. Natl. Acad. Sci. 80: 4803-4807). The cell can include a nucleic acid of the invention that encodes a polypeptide, wherein the cell expresses a polypeptide of the invention. The cell can also include vector sequences, or the like. Furthermore, cells and transgenic plants that include any polypeptide or nucleic acid above or throughout this specification, e.g., produced by transduction of a vector of the invention, are an additional feature of the invention. For long-term, high-yield production of recombinant proteins, stable expression can be used. Host cells transformed with a nucleotide sequence encoding a polypeptide of the invention are optionally cultured under conditions suitable for the expression and recovery of the encoded protein from cell culture. The protein or fragment thereof produced by a recombinant cell may be secreted, membrane- bound, or contained intracellularly, depending on the sequence and/or the vector used. As will be understood by those of skill in the art, expression vectors containing polynucleotides encoding mature proteins of the invention can be designed with signal sequences which direct secretion of the mature polypeptides through a prokaryotic or eukaryotic cell membrane.

Modified Amino Acid Residues Polypeptides of the invention may contain one or more modified amino acid residues. The presence of modified amino acids may be advantageous in, for example, increasing polypeptide half-life, reducing polypeptide antigenicity or toxicity, increasing polypeptide storage stability, or the like. Amino acid residue(s) are modified, for example, co-translationally or post-translationally during recombinant production or modified by synthetic or chemical means. Non-limiting examples of a modified amino acid residue include incorporation or other use of acetylated amino acids, glycosylated amino acids, sulfated amino acids, prenylated (e.g., farnesylated, geranylgeranylated) amino acids, PEG modified (e.g., "PEGylated") amino acids, biotinylated amino acids, carbpxylated amino acids, phosphorylated amino acids, etc. References adequate to guide one of skill in the modification of amino acid residues are replete throughout the literature. The modified amino acid residues may prevent or increase affinity of the polypeptide for another molecule, including, but not limited to, polynucleotide, proteins, carbohydrates, lipids and lipid derivatives, and other organic or synthetic compounds.

Identification of Additional Protein Factors A transcription factor provided by the present invention can also be used to identify additional endogenous or exogenous molecules that can affect a phenotype or trait of interest. Such molecules include endogenous molecules that are acted upon either at a transcriptional level by a transcription factor of the invention to modify a phenotype as desired. For example, the transcription factors can be employed to identify one or more downstream genes that are subject to a regulatory effect of the transcription factor. In one approach, a transcription factor or transcription factor homolog of the invention is expressed in a host cell, e.g., a transgenic plant cell, tissue or explant, and expression products, either RNA or protein, of likely or random targets are monitored, e.g., by hybridization to a microarray of nucleic acid probes corresponding to genes expressed in a tissue or cell type of interest, by two-dimensional gel electrophoresis of protein products, or by any other method known in the art for assessing expression of gene products at the level of RNA or protein. Alternatively, a transcription factor of the invention can be used to identify promoter sequences (such as binding sites on DNA sequences) involved in the regulation of a downstream target. After identifying a promoter sequence, interactions between the transcription factor and the promoter sequence can be modified by changing specific nucleotides in the promoter sequence or specific amino acids in the transcription factor that interact with the promoter sequence to alter a plant trait. Typically, transcription factor DNA-binding sites are identified by gel shift assays. After identifying the promoter regions, the promoter region sequences can be employed in double- stranded DNA arrays to identify molecules that affect the interactions of the transcription factors with their promoters (Bulyk et al. (1999) Nature Biotechnol. 17: 573-577). The identified transcription factors are also useful to identify proteins that modify the activity of the transcription factor. Such modification can occur by covalent modification, such as by phosphorylation, or by protein-protein (homo or-heteropolymer) interactions. Any method suitable for detecting protein-protein interactions can be employed. Among the methods that can be employed are co- immunoprecipitation, cross-linking and co-purification through gradients or chromatographic columns, and the two-hybrid yeast system. The two-hybrid system detects protein interactions in vivo and is described in Chien et al. ((1991) Proc. Natl. Acad. Sci. 88: 9578-9582) and is commercially available from Clontech (Palo Alto, Calif). In such a system, plasmids are constructed that encode two hybrid proteins: one consists of the DNA- binding domain of a transcription activator protein fused to the TF polypeptide and the other consists of the transcription activator protein's activation domain fused to an unknown protein that is encoded by a cDNA that has been recombined into the plasmid as part of a cDNA library. The DNA-binding domain fusion plasmid and the cDNA library are transformed into a strain of the yeast Saccharomyces cerevisiae that contains a reporter gene (e.g., lacZ) whose regulatory region contains the transcription activator's binding site. Either hybrid protein alone cannot activate transcription of the reporter gene. Interaction of the two hybrid proteins reconstitutes the functional activator protein and results in expression of the reporter gene, which is detected by an assay for the reporter gene product. Then, the library plasmids responsible for reporter gene expression are isolated and sequenced to identify the proteins encoded by the library plasmids. After identifying proteins that interact with the transcription factors, assays for compounds that interfere with the TF protein-protein interactions can be performed.

Subsequences Also contemplated are uses of polynucleotides, also referred to herein as oligonucleotides, typically having at least 12 bases, preferably at least 15, more preferably at least 20, 30, or 50 bases, which hybridize under stringent conditions to a polynucleotide sequence described above. The polynucleotides may be used as probes, primers, sense and antisense agents, and the like, according to methods as noted supra. Subsequences of the polynucleotides of the invention, including polynucleotide fragments and oligonucleotides, are useful as nucleic acid probes and primers. An oligonucleotide suitable for use as a probe or primer is typically at least about 15 nucleotides in length, and frequently at least about 30 or 40 or more nucleotides in length. A nucleic acid probe is useful in hybridization protocols, for example, to identify additional polypeptide homologs of the invention, including protocols for microarray experiments. Primers can be annealed to a complementary target DNA strand by nucleic acid hybridization to form a hybrid between the primer and the target DNA strand, and then extended along the target DNA strand with DNA polymerase. Primer pairs can be used for amplification of a nucleic acid sequence, e.g., by the polymerase chain reaction (PCR) or other nucleic-acid amplification methods (Sambrook, supra, and Ausubel, supra). In addition, the invention includes an isolated or recombinant polypeptide including a subsequence of at least about 15 contiguous amino acids encoded by the recombinant or isolated polynucleotides of the invention. For example, such polypeptides, or domains or fragments thereof, can be used as immunogens, e.g., to produce antibodies specific for the polypeptide sequence, or as probes for detecting a sequence of interest. A subsequence can range in size from about 15 amino acids in length up to and including the full length of the polypeptide. To be encompassed by the present invention, an expressed polypeptide which comprises such a polypeptide subsequence performs at least one biological function of the intact polypeptide in substantially the same manner, or to a similar extent, as does the intact polypeptide. For example, a polypeptide fragment can comprise a recognizable structural motif or functional domain such as a DNA binding domain that activates transcription, e.g., by binding to a specific DNA promoter region an activation domain, or a domain for protein-protein interactions.

Production of Transgenic Plants Modification of Traits The polynucleotides of the invention are favorably employed to produce transgenic plants with various traits, or characteristics, that have been modified in a desirable manner, e.g., to improve the seed characteristics of a plant. For example, alteration of expression levels or patterns (e.g., spatial or temporal expression patterns) of one or more of the transcription factors (or transcription factor homologs) of the invention, as compared with the levels of the same protein found in a wild-type plant, can be used to modify a plant's traits. An illustrative example of trait modification, improved characteristics, by altering expression levels of a particular transcription factor is described further in the Examples and the Sequence Listing. Arabidopsis as a model system Arabidopsis thaliana is the object of rapidly growing attention as a model for genetics and metabolism in plants. Arabidopsis has a small genome, and well-documented studies are available. It is easy to grow in large numbers and mutants defining important genetically controlled mechanisms are either available, or can readily be obtained. Various methods to introduce and express isolated homologous genes are available (Koncz et al., eds., Methods in Arabidopsis Research (1992) World Scientific, New Jersey, NJ, in "Preface"). Because of its small size, short life cycle, obligate autogamy and high fertility, Arabidopsis is also a choice organism for the isolation of mutants and studies in morphogenetic and development pathways, and control of these pathways by transcription factors (Koncz (1992) supra, p. 72). A number of studies introducing transcription factors into A. thaliana have demonstrated the utility of this plant for understanding the mechanisms of gene regulation and trait alteration in plants (for example, Koncz (1992) supra, and US Patent Number 6,417,428).

Arabidopsis genes in transgenic plants. Expression of genes that encode transcription factors that modify expression of endogenous genes, polynucleotides, and proteins are well known in the art. In addition, transgenic plants comprising isolated polynucleotides encoding transcription factors may also modify expression of endogenous genes, polynucleotides, and proteins. Examples include Peng et al. (1997) Genes and Development 11 : 3194- 3205, and Peng et al. (1999) Nature 400: 256-261. In addition, many others have demonstrated that an Arabidopsis transcription factor expressed in an exogenous plant species elicits the same or very similar phenotypic response (for example, Fu et al. (2001) Plant Cell 13: 1791-1802; Nandi et al. (2000) Curr. Biol. 10: 215-218; Coupland (1995) Nature 311: 482-483; and Weigel and Nilsson (1995) Nature 377: 482-500).

Homologous genes introduced into transgenic plants. Homologous genes that may be derived from any plant, or from any source whether natural, synthetic, semi-synthetic or recombinant, and that share significant sequence identity or similarity to those provided by the present invention, may be introduced into plants, for example, crop plants, to confer desirable or improved traits. Consequently, transgenic plants may be produced that comprise a recombinant expression vector or cassette with a promoter operably linked to one or more sequences homologous to presently disclosed sequences. The promoter may be, for example, a plant or viral promoter. The invention thus provides for methods for preparing transgenic plants, and for modifying plant traits. These methods include introducing into a plant a recombinant expression vector or cassette comprising a functional promoter operably linked to one or more sequences homologous to presently disclosed sequences. Plants and kits for producing these plants that result from the application of these methods are also encompassed by the present invention.

Transcription factors of interest for the modification of plant traits Currently, the existence of a series of maturity groups for different latitudes represents a major barrier to the introduction of new valuable traits. Any trait (e.g. drought tolerance) has to be bred into each of the different maturity groups separately, a laborious and costly exercise. The availability of single strain, which could be grown at any latitude, would therefore greatly increase the potential for introducing new traits to crop species such as soybean and cotton. For the specific effects, traits and utilities conferred to plants, one or more transcription factor genes of the present invention may be used to increase or decrease, or improve or prove deleterious to a given trait. For example, knocking out a transcription factor gene that naturally occurs in a plant, or suppressing the gene (with, for example, antisense suppression), may cause decreased tolerance to an abiotic stress relative to non-transformed or wild-type plants. By overexpressing this gene, the plant may experience increased tolerance to the same stress. More than one transcription factor gene may be introduced into a plant, either by transforming the plant with one or more vectors comprising two or more transcription factors, or by selective breeding of plants to yield hybrid crosses that comprise more than one introduced transcription factor. Genes, traits and utilities that affect plant characteristics Plant transcription factors can modulate gene expression, and, in turn, be modulated by the environmental experience of a plant. Significant alterations in a plant's environment invariably result in a change in the plant's transcription factor gene expression pattern. Altered transcription factor expression patterns generally result in phenotypic changes in the plant. Transcription factor gene product(s) in transgenic plants then differ(s) in amounts or proportions from that found in wild-type or non-transformed plants, and those transcription factors likely represent polypeptides that are used to alter the response to the environmental change. By way of example, it is well accepted in the art that analytical methods based on altered expression patterns may be used to screen for phenotypic changes in a plant far more effectively than can be achieved using traditional methods. Sugar sensing and effects on yield. In addition to their important role as an energy source and structural component of the plant cell, sugars are central regulatory molecules that control several aspects of plant physiology, metabolism and development (Hsieh et al. (1998) Proc. Natl. Acad. Sci. 95: 13965-13970). It is thought that this control is achieved by regulating gene expression and, in higher plants, sugars have been shown to repress or activate plant genes involved in many essential processes such as photosynthesis, glyoxylate metabolism, respiration, starch and sucrose synthesis and degradation, pathogen response, wounding response, cell cycle regulation, pigmentation, flowering and senescence. The mechanisms by which sugars control gene expression are not understood. Several sugar sensing mutants have turned out to be allelic to abscisic acid (ABA) and ethylene mutants. ABA is found in all photosynthetic organisms and acts as a key regulator of transpiration, stress responses, embryogenesis, and seed germination. Most ABA effects are related to the compound acting as a signal of decreased water availability, whereby it triggers a reduction in water loss, slows growth, and mediates adaptive responses. However, ABA also influences plant growth and development via interactions with other phytohormones. Physiological and molecular studies indicate that maize and Arabidopsis have almost identical pathways with regard to ABA biosynthesis and signal transduction. For further review, see Finkelstein and Rock (2002) "Abscisic acid biosynthesis and response", in The Arabidopsis Book, Somerville and Meyerowitz (editors), American Society of Plant Biologists, Rockville, MD. This potentially implicates G867, G9, G993, G1930, G3389, G3391 , G3451, G3452, and G3455 in hormone signaling based on the sucrose sensing phenotype of transgenic lines overexpressing these sequences (Example VIII, below). On the other hand, the sucrose treatment used in these experiments (9.4%) w/v) could also be an osmotic stress. Therefore, one could interpret these data as an indication that these transgenic lines are more tolerant to osmotic stress. However, it is well known that plant responses to ABA, osmotic and other stress may be linked, and these different treatments may even act in a synergistic manner to increase the degree of a response. For example, Xiong, Ishitani, and Zhu ((1999) Plant Physiol. 119: 205-212) have shown that genetic and molecular studies may be used to show extensive interaction between osmotic stress, temperature stress, and ABA responses in plants. These investigators analyzed the expression of RD 29 A-LUC in response to various treatment regimes in Arabidopsis. The RD29A promoter contains both the ABA-responsive and the dehydration-responsive element - also termed the C-repeat - and can be activated by osmotic stress, low temperature, or ABA treatment;. Transcription of the RD29A gene in response to osmotic and cold stresses is mediated by both ABA-dependent and ABA-independent pathways (Xiong, Ishitani, and Zhu (1999) supra). LUC refers to the firefly luciferase coding sequence, which, in this case, was driven by the stress responsive RD29A promoter. The results revealed both positive and negative interactions, depending on the nature and duration of the treatments. Low temperature stress was found to impair osmotic signaling but moderate heat stress strongly enhanced osmotic stress induction, thus acting synergistically with osmotic signaling pathways. In this study, the authors reported that osmotic stress and ABA can act synergistically by showing that the treatments simultaneously induced transgene and endogenous gene expression. Similar results were reported by Bostock and Quatrano ((1992) Plant Physiol. 98: 1356-1363), who found that osmotic stress and ABA act synergistically and induce maize Em gene expression. Ishitani et al (1997) Plant Cell 9: 1935-1949) isolated a group of Arabidopsis single-gene mutations that confer enhanced responses to both osmotic stress and ABA. The nature of the recovery of these mutants from osmotic stress and ABA treatment suggested that although separate signaling pathways exist for osmotic stress and ABA, the pathways share a number of components; these common components may mediate synergistic interactions between osmotic stress and ABA. Thus, contrary to the previously-held belief that ABA-dependent and ABA-independent stress signaling pathways act in a parallel manner, our data reveal that these pathways cross-talk and converge to activate stress gene expression. Because sugars are important signaling molecules, the ability to control either the concentration of a signaling sugar or how the plant perceives or responds to a signaling sugar could be used to control plant development, physiology or metabolism. For example, the flux of sucrose (a disaccharide sugar used for systemically transporting carbon and energy in most plants) has been shown to affect gene expression and alter storage compound accumulation in seeds. Manipulation of the sucrose signaling pathway in seeds may therefore cause seeds to have more protein, oil or carbohydrate, depending on the type of manipulation. Similarly, in tubers, sucrose is converted to starch which is used as an energy store. It is thought that sugar signaling pathways may partially determine the levels of starch synthesized in the tubers. The manipulation of sugar signaling in tubers could lead to tubers with a higher starch content. Thus, the presently disclosed transcription factor genes that manipulate the sugar signal transduction pathway, including G867, G9, G993, G1930, G3389, G3391, G3451, G3452, and G3455, along with their equivalogs, may lead to altered gene expression to produce plants with desirable traits. In particular, manipulation of sugar signal transduction pathways could be used to alter source-sink relationships in seeds, tubers, roots and other storage organs leading to increase in yield. Abiotic stress: drought and low humidity tolerance and effects on yield. Exposure to dehydration invokes similar survival strategies in plants as does freezing stress (for example, Yelenosky (1989) Plant Physiol 89: 444-451) and drought stress induces freezing tolerance (for example, Siminovitch et al. (1982) Plant Physiol 69: 250-255; and Guy et al. (1992) Planta 188: 265-270). In addition to the induction of cold-acclimation proteins, strategies that allow plants to survive in low water conditions may include, for example, reduced surface area, or surface oil or wax production. Modifying the expression of a number of presently disclosed transcription factor genes, such as G9, G867, G3432 and G3451 may be used to increase a plant's tolerance to low water conditions and provide the benefits of improved increased yield, survival and an extended geographic and temporal planting range. Osmotic stress and effects on yield. Modification of the expression of a number of presently disclosed transcription factor genes, e.g., G867, G9, G993, G1930, G3389, G3390, G3391, G3451, G3452, and G3455, and their equivalogs, may be used to increase germination rate or growth under adverse osmotic conditions, which could impact survival and yield of seeds and plants. Osmotic stresses may be regulated by specific molecular control mechanisms that include genes controlling water and ion movements, functional and structural stress-induced proteins, signal perception and transduction, and free radical scavenging, and many others (Wang et al. (2001) Ada Hort. (ISHS) 560: 285-292). Instigators of osmotic stress include freezing, drought and high salinity, each of which are discussed below. In many ways, freezing, high salt and drought have similar effects on plants, not the least of which is induction of common polypeptides that respond to these different stresses. For example, freezing is similar to water deficit in that freezing reduces the amount of water available to a plant. Exposure to freezing temperatures may lead to cellular dehydration as water leaves cells and forms ice crystals in intercellular spaces (Buchanan (2000) supra). As with high salt concentration and freezing, problems for plants caused by low water availability include mechanical stresses caused by the withdrawal of cellular water. Thus, the incorporation of transcription factors that modify a plant's response to osmotic stress into, for example, a crop or ornamental plant, may be useful in increasing yield by reducing damage or loss. Specific effects caused by freezing, high salt and drought are addressed below. Salt and drought tolerance Plants are subject to a range of environmental challenges. Several of these, including salt stress, general osmotic stress, drought stress and freezing stress, have the ability to impact whole plant and cellular water availability. Not surprisingly, then, plant responses to this collection of stresses are related. In a recent review, Zhu notes that "most studies on water stress signaling have focused on salt stress primarily because plant responses to salt and drought are closely related and the mechanisms overlap" (Zhu (2002) Ann. Rev. Plant Biol. 53: 247-273). Many examples of similar responses and pathways to this set of stresses have been documented. For example, the CBF transcription factors have been shown to condition resistance to salt, freezing and drought (Kasuga et al. (1999) Nature Biotech. 17: 287-291). The Arabidopsis rd29B gene is induced in response to both salt and dehydration stress, a process that is mediated largely through an ABA signal transduction process (Uno et al. (2000) Proc. Natl. Acad. Sci. USA 97: 11632-11637), resulting in altered activity of transcription factors that bind to an upstream element within the rd29B promoter. In Mesembiγanthemum crystallinum (ice plant), Patharker and Cushman have shown that a calcium-dependent protein kinase (McCDPKl) is induced by exposure to both drought and salt stresses (Patharker and Cushman (2000) Plant J. 24: 679-691). The stress-induced kinase was also shown to phosphorylate a transcription factor, presumably altering its activity, although transcript levels of the target transcription factor are not altered in response to salt or drought stress. Similarly, Saijo et al. demonstrated that a rice salt/drought-induced calmodulin-dependent protein kinase (OsCDPK7) conferred increased salt and drought tolerance to rice when overexpressed (Saijo et al. (2000) Plant J. 23: 319-327). Exposure to dehydration invokes similar survival strategies in plants as does freezing stress (for example, Yelenosky (1989) Plant Physiol 89: 444-451) and drought stress induces freezing tolerance (for example, Siminovitch et al. (1982) Plant Physiol 69: 250-255; and Guy et al. (1992) Planta 188: 265- 270). In addition to the induction of cold-acclimation proteins, strategies that allow plants to survive in low water conditions may include, for example, reduced surface area, or surface oil or wax production. Consequently, one skilled in the art would expect that some pathways involved in resistance to one of these stresses, and hence regulated by an individual transcription factor, will also be involved in resistance to another of these stresses, regulated by the same or homologous transcription factors. Of course, the overall resistance pathways are related, not identical, and therefore not all transcription factors controlling resistance to one stress will control resistance to the other stresses. Nonetheless, if a transcription factor conditions resistance to one of these stresses, it would be apparent to one skilled in the art to test for resistance to these related stresses. The genes of the sequence listing, including, for example, G867, G9, G993, G1930, G3389, G3391 , G3451 , G3452 and their equivalogs, that provide tolerance to salt may be used to engineer salt tolerant crops and trees that can flourish in soils with high saline content or under drought conditions. In particular, increased salt tolerance during the germination stage of a plant enhances yield and survival. Presently disclosed transcription factor genes that provide increased salt tolerance during germination, the seedling stage, and throughout a plant's life cycle, would find particular value for imparting survival and yield in areas where a particular crop would not normally prosper. Root growth and vigor. Some of the genes in the Sequence Listing, including G867, G9, G993, G3390, G3451, G3452 and G3455 have been shown to increase root growth and to produce hairy roots on media containing methyl jasmonate. Thus, these genes could potentially be used to increase root growth and vigor, which might in turn allow better plant growth and yield during periods of osmotic stress, or limited nutrient availability. Summary of altered plant characteristics. A clade of structurally and functionally related sequences that derive from a wide range of plants, including polynucleotide SEQ ID NOs 1 , 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, or 51, polynucleotides that encode polypeptide SEQ ID NOs: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, or 53, fragments thereof, paralogs, orthologs, equivalogs, and fragments thereof, is provided. These sequences have been shown in laboratory and field experiments to confer increased size in transformed plants as compared to wild-type plants in conditions of abiotic stress. The invention also provides polypeptides comprising SEQ ID NOs: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, or 53, and fragments thereof, conserved domains thereof, paralogs, orthologs, equivalogs, and fragments thereof. Plants that overexpress these sequences have been observed to be more tolerant to a wide variety of abiotic stresses, including, germination in heat and cold, and osmotic stresses such as drought and high salt levels. Many of the orthologs of these sequences are listed in the Sequence Listing, and due to the high degree of structural similarity to the sequences of the invention, it is expected that these sequences may also function to increase abiotic stress tolerance and yield. The invention also encompasses the complements of the polynucleotides. The polynucleotides are useful for screening libraries of molecules or compounds for specific binding and for creating transgenic plants having increased abiotic stress tolerance.

Antisense and Co-suppression In addition to expression of the nucleic acids of the invention as gene replacement or plant phenotype modification nucleic acids, the nucleic acids are also useful for sense and anti-sense suppression of expression, e.g. to down-regulate expression of a nucleic acid of the invention, e.g. as a further mechanism for modulating plant phenotype. That is, the nucleic acids of the invention, or subsequences or anti-sense sequences thereof, can be used to block expression of naturally occurring homologous nucleic acids. A variety of sense and anti-sense technologies are known in the art, e.g. as set forth in Lichtenstein and Nellen (1997) Antisense Technology: A Practical Approach, IRL Press at Oxford University Press, Oxford, U.K. Antisense regulation is also described in Crowley et al. (1985) Cell 43: 633-641 ; Rosenberg et al. (1985) Nature 313: 703-706; Preiss et al. (1985) Nature 313: 27-32; Melton (1985) Proc. Natl. Acad. Sci. 82: 144-148; Izant and Weintraub (1985) Science 229: 345-352; and Kim and Wold (1985) Cell 42: 129-138. Additional methods for antisense regulation are known in the art. Antisense regulation has been used to reduce or inhibit expression of plant genes in, for example in European Patent Publication No. 271988. Antisense RNA may be used to reduce gene expression to produce a visible or biochemical phenotypic change in a plant (Smith et al. (1988) Nature, 334: 724-726; Smith et al. (1990) Plant Mol. Biol. 14: 369-379). In general, sense or anti-sense sequences are introduced into a cell, where they are optionally amplified, e.g. by transcription. Such sequences include both simple oligonucleotide sequences and catalytic sequences such as ribozymes. For example, a reduction or elimination of expression (i.e., a "knock-out") of a transcription factor or transcription factor homolog polypeptide in a transgenic plant, e.g., to modify a plant trait, can be obtained by introducing an antisense construct corresponding to the polypeptide of interest as a cDNA. For antisense suppression, the transcription factor or homolog cDNA is arranged in reverse orientation (with respect to the coding sequence) relative to the promoter sequence in the expression vector. The introduced sequence need not be the full-length cDNA or gene, and need not be identical to the cDNA or gene found in the plant type to be transformed. Typically, the antisense sequence need only be capable of hybridizing to the target gene or RNA of interest. Thus, where the introduced sequence is of shorter length, a higher degree of homology to the endogenous transcription factor sequence will be needed for effective antisense suppression. While antisense sequences of various lengths can be used, preferably, the introduced antisense sequence in the vector will be at least 30 nucleotides in length, and improved antisense suppression will typically be observed as the length of the antisense sequence increases. Preferably, the length of the antisense sequence in the vector will be greater than 100 nucleotides. Transcription of an antisense construct as described results in the production of RNA molecules that are the reverse complement of mRNA molecules transcribed from the endogenous transcription factor gene in the plant cell. Suppression of endogenous transcription factor gene expression can also be achieved using RNA interference, or RNAi. RNAi is a post-transcriptional, targeted gene-silencing technique that uses double- stranded RNA (dsRNA) to incite degradation of messenger RNA (mRNA) containing the same sequence as the dsRNA (Constans (2002) The Scientist 16: 36). Small interfering RNAs, or siRNAs are produced in at least two steps: an endogenous ribonuclease cleaves longer dsRNA into shorter, 21-23 nucleotide- long RNAs. The siRNA segments then mediate the degradation of the target mRNA (Zamore (2001) Nature Struct. Biol. 8: 746-750). RNAi has been used for gene function determination in a manner similar to antisense oligonucleotides (Constans (2002) supra). Expression vectors that continually express siRNAs in transiently and stably transfected cells have been engineered to express small hairpin RNAs (shRNAs), which get processed in vivo into siRNAs-like molecules capable of carrying out gene-specific silencing (Brummelkamp et al. (2002) Science 296: 550-553, and Paddison et al. (2002) Genes & Dev. 16: 948-958). Post-transcriptional gene silencing by double-stranded RNA is discussed in further detail by Hammond et al. (2001) Nature Rev Gen 2: 110-119, Fire et al. (1998) Nature 391 : 806-811 and Timmons and Fire (1998) Nature 395: 854. Vectors in which RNA encoded by a transcription factor or transcription factor homolog cDNA is over-expressed can also be used to obtain co-suppression of a corresponding endogenous gene, e.g., in the manner described in US Patent No. 5,231,020 by Jorgensen. Such co-suppression (also termed sense suppression) does not require that the entire transcription factor cDNA be introduced into the plant cells, nor does it require that the introduced sequence be exactly identical to the endogenous transcription factor gene of interest. However, as with antisense suppression, the suppressive efficiency will be enhanced as specificity of hybridization is increased, e.g., as the introduced sequence is lengthened, and/or as the sequence similarity between the introduced sequence and the endogenous transcription factor gene is increased. Vectors expressing an untranslatable form of the transcription factor mRNA, e.g., sequences comprising one or more stop codon, or nonsense mutation, can also be used to suppress expression of an endogenous transcription factor, thereby reducing or eliminating its activity and modifying one or more traits. Methods for producing such constructs are described in US Patent No. 5,583,021. Preferably, such constructs are made by introducing a premature stop codon into the transcription factor gene. Alternatively, a plant trait can be modified by gene silencing using double-stranded RNA (Sharp (1999) Genes and Development 13: 139-141). Another method for abolishing the expression of a gene is by insertion mutagenesis using the T-DNA oi Agrobacterium tumefaciens. After generating the insertion mutants, the mutants can be screened to identify those containing the insertion in a transcription factor or transcription factor homolog gene. Plants containing a single transgene insertion event at the desired gene can be crossed to generate homozygous plants for the mutation. Such methods are well known to those of skill in the art (for example, Koncz et al. (1992) Methods in Arabidopsis Research, World Scientific Publishing Co. Pte. Ltd., River Edge, NJ). Alternatively, a plant phenotype can be altered by eliminating an endogenous gene, such as a transcription factor or transcription factor homolog, e.g., by homologous recombination (Kempin et al. (1997) Nature 389: 802-803). A plant trait can also be modified by using the Cre-lox system (for example, as described in US Pat. No. 5,658,772). A plant genome can be modified to include first and second lox sites that are then contacted with a Cre recombinase. If the lox sites are in the same orientation, the intervening DNA sequence between the two sites is excised. If the lox sites are in the opposite orientation, the intervening sequence is inverted. The polynucleotides and polypeptides of this invention can also be expressed in a plant in the absence of an expression cassette by manipulating the activity or expression level of the endogenous gene by other means, such as, for example, by ectopically expressing a gene by T-DNA activation tagging

(Ichikawa et al. (1997) Nature 390 698-701 ; Kakimoto et al. (1996) Science 274: 982-985). This method entails transforming a plant with a gene tag containing multiple transcriptional enhancers and once the tag has inserted into the genome, expression of a flanking gene coding sequence becomes deregulated. In another example, the transcriptional machinery in a plant can be modified so as to increase transcription levels of a polynucleotide of the invention (e.g., PCT Publications WO 96/06166 and WO 98/53057 which describe the modification of the DNA-binding specificity of zinc finger proteins by changing particular amino acids in the DNA-binding motif). The transgenic plant can also include the machinery necessary for expressing or altering the activity of a polypeptide encoded by an endogenous gene, for example, by altering the phosphorylation state of the polypeptide to maintain it in an activated state. Transgenic plants (or plant cells, or plant explants, or plant tissues) incorporating the polynucleotides of the invention and/or expressing the polypeptides of the invention can be produced by a variety of well established techniques as described above. Following construction of a vector, most typically an expression cassette, including a polynucleotide, e.g., encoding a transcription factor or transcription factor homolog, of the invention, standard techniques can be used to introduce the polynucleotide into a plant, a plant cell, a plant explant or a plant tissue of interest. Optionally, the plant cell, explant or tissue can be regenerated to produce a transgenic plant. The plant can be any higher plant, including gymnosperms, monocotyledonous and dicotyledonous plants. Suitable protocols are available for Leguminosae (alfalfa, soybean, clover, etc.), Umbelliferae (carrot, celery, parsnip), Cruciferae (cabbage, radish, rapeseed, broccoli, etc.), Curcurbitaceae (melons and cucumber), Gramineae (wheat, com, rice, barley, millet, etc.), Solanaceae (potato, tomato, tobacco, peppers, etc.), and various other crops. Examples of these protocols are described in Ammirato et al., eds., (1984) Handbook of Plant Cell Culture -Crop Species, Macmillan Publ. Co., New York, NY; Shimamoto et al. (1989) Nature 338: 274-276; Fromm et al. (1990) Bio/Technol. 8: 833-839; and Vasil et al. (1990) Bio/Technol. 8: 429-434. Transformation and regeneration of both monocotyledonous and dicotyledonous plant cells is now routine, and the selection of the most appropriate transformation technique will be determined by the practitioner. The choice of method will vary with the type of plant to be transformed; those skilled in the art will recognize the suitability of particular methods for given plant types. Suitable methods can include, but are not limited to: electroporation of plant protoplasts; liposome-mediated transformation; polyethylene glycol (PEG) mediated transformation; transformation using viruses; micro-injection of plant cells; micro-projectile bombardment of plant cells; vacuum infiltration; and Agrobacterium tumefaciens mediated transformation. Transformation means introducing a nucleotide sequence into a plant in a manner to cause stable or transient expression of the sequence. Successful examples of the modification of plant characteristics by transformation with cloned sequences which serve to illustrate the current knowledge in this field of technology, and which are herein incorporated by reference, include: US Patent Nos. 5,571,706; 5,677,175; 5,510,471; 5,750,386; 5,597,945; 5,589,615; 5,750,871 ; 5,268,526; 5,780,708; 5,538,880; 5,773,269; 5,736,369 and 5,610,042. Following transformation, plants are preferably selected using a dominant selectable marker incorporated into the transformation vector. Typically, such a marker will confer antibiotic or herbicide resistance on the transformed plants, and selection of transformants can be accomplished by exposing the plants to appropriate concentrations of the antibiotic or herbicide. After transformed plants are selected and grown to maturity, plants showing a modified trait are identified. The modified trait may be abiotic stress tolerance. To confirm that the modified trait is due to changes in expression levels or activity of the polypeptide or polynucleotide of the invention can be determined by analyzing mRNA expression using, for example, Northern blots, RT-PCR or microarrays, or protein expression using, for example, immunoblots, Western blots or gel shift assays. Integrated Systems - Sequence Identity Additionally, the present invention may be an integrated system, computer or computer readable medium that comprises an instruction set for determining the identity of one or more sequences in a database. In addition, the instruction set can be used to generate or identify sequences that meet any specified criteria. Furthermore, the instruction set may be used to associate or link certain functional benefits, such improved characteristics, with one or more identified sequence. For example, the instruction set can include, e.g., a sequence comparison or other alignment program, e.g., an available program such as, for example, the Wisconsin Package Version 10.0, such as BLAST, FASTA, PILEUP, FINDPATTERNS or the like (GCG, Madison, WI). Public sequence databases such as GenBank, EMBL, Swiss-Prot and PIR or private sequence databases such as PHYTOSEQ sequence database (Incyte Genomics, Palo Alto, CA) can be searched. Alignment of sequences for comparison can be conducted by the local homology algorithm of Smith and Waterman (1981) Adv. Appl. Math. 2: 482-489, by the homology alignment algorithm of Needleman and Wunsch (1970) J. Mol. Biol. 48: 443-453, by the search for similarity method of Pearson and Lipman (1988) Proc. Natl. Acad. Sci. 85: 2444-2448, by computerized implementations of these algorithms. After alignment, sequence comparisons between two (or more) polynucleotides or polypeptides are typically performed by comparing sequences of the two sequences over a comparison window to identify and compare local regions of sequence similarity. The comparison window can be a segment of at least about 20 contiguous positions, usually about 50 to about 200, more usually about 100 to about 150 contiguous positions. A description of the method is provided in Ausubel et al. supra. A variety of methods for determining sequence relationships can be used, including manual alignment and computer assisted sequence alignment and analysis. This later approach is a preferred approach in the present invention, due to the increased throughput afforded by computer assisted methods. As noted above, a variety of computer programs for performing sequence alignment are available, or can be produced by one of skill. One example algorithm that is suitable for determining percent sequence identity and sequence similarity is the BLAST algorithm, which is described in Altschul et al. (1990) J. Mol. Biol. 215: 403- 410. Software for performing BLAST analyses is publicly available, e.g., through the National Library of Medicine's National Center for Biotechnology Information (ncbi.nlm.nih; at world wide web (www) National Institutes of Health US government (gov) website). This algorithm involves first identifying high scoring sequence pairs (HSPs) by identifying short words of length W in the query sequence, which either match or satisfy some positive-valued threshold score T when aligned with a word of the same length in a database sequence. T is referred to as the neighborhood word score threshold (Altschul et al. supra). These initial neighborhood word hits act as seeds for initiating searches to find longer HSPs containing them. The word hits are then extended in both directions along each sequence for as far as the cumulative alignment score can be increased. Cumulative scores are calculated using, for nucleotide sequences, the parameters M (reward score for a pair of matching residues; always > 0) and N (penalty score for mismatching residues; always < 0). For amino acid sequences, a scoring matrix is used to calculate the cumulative score. Extension of the word hits in each direction are halted when: the cumulative alignment score falls off by the quantity X from its maximum achieved value; the cumulative score goes to zero or below, due to the accumulation of one or more negative-scoring residue alignments; or the end of either sequence is reached. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment. The BLASTN program (for nucleotide sequences) uses as defaults a wordlength (W) of 1 1 , an expectation (E) of 10, a cutoff of 100, M=5, N=-4, and a comparison of both strands. For amino acid sequences, the BLASTP program uses as defaults a wordlength (W) of 3, an expectation (E) of 10, and the BLOSUM62 scoring matrix (for example, Henikoff and Henikoff (1992) Proc. Natl. Acad. Sci. 89: 10915-10919). Unless otherwise indicated, "sequence identity" here refers to the % sequence identity generated from a tblastx using the NCBI version of the algorithm at the default settings using gapped alignments with the filter "off (for example, NIH NLM NCBI website at ncbi.nlm.nih; world wide web (www) National Institutes of Health US government (gov) website). In addition to calculating percent sequence identity, the BLAST algorithm also performs a statistical analysis of the similarity between two sequences (for example, Karlin and Altschul (1993) Proc. Natl. Acad. Sci. 90: 5873-5787). One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which provides an indication of the probability by which a match between two nucleotide or amino acid sequences would occur by chance. For example, a nucleic acid is considered similar to a reference sequence (and, therefore, in this context, homologous) if the smallest sum probability in a comparison of the test nucleic acid to the reference nucleic acid is less than about 0.1, or less than about 0.01, and or even less than about 0.001. An additional example of a useful sequence alignment algorithm is PILEUP. PILEUP creates a multiple sequence alignment from a group of related sequences using progressive, pairwise alignments. The program can align, e.g., up to 300 sequences of a maximum length of 5,000 letters. The integrated system, or computer typically includes a user input interface allowing a user to selectively view one or more sequence records corresponding to the one or more character strings, as well as an instruction set which aligns the one or more character strings with each other or with an additional character string to identify one or more region of sequence similarity. The system may include a link of one or more character strings with a particular phenotype or gene function. Typically, the system includes a user readable output element that displays an alignment produced by the alignment instruction set. The methods of this invention can be implemented in a localized or distributed computing environment. In a distributed environment, the methods may be implemented on a single computer comprising multiple processors or on a multiplicity of computers. The computers can be linked, e.g. through a common bus, but more preferably the computer(s) are nodes on a network. The network can be a generalized or a dedicated local or wide-area network and, in certain preferred embodiments, the computers may be components of an infra-net or an internet. Thus, the invention provides methods for identifying a sequence similar or homologous to one or more polynucleotides as noted herein, or one or more target polypeptides encoded by the polynucleotides, or otherwise noted herein and may include linking or associating a given plant phenotype or gene function with a sequence. In the methods, a sequence database is provided (locally or across an inter or infra net) and a query is made against the sequence database using the relevant sequences herein and associated plant phenotypes or gene functions. Any sequence herein can be entered into the database, before or after querying the database. This provides for both expansion of the database and, if done before the querying step, for insertion of control sequences into the database. The control sequences can be detected by the query to ensure the general integrity of both the database and the query. As noted, the query can be performed using a web browser based interface. For example, the database can be a centralized public database such as those noted herein, and the querying can be done from a remote terminal or computer across an internet or intranet. Any sequence herein can be used to identify a similar, homologous, paralogous, or orthologous sequence in another plant. This provides means for identifying endogenous sequences in other plants that may be useful to alter a trait of progeny plants, which results from crossing two plants of different strains. Sequences that encode an ortholog of a sequence herein that naturally occurs in a plant with a desired trait can be identified. The plant is then crossed with a second plant of the same species but which does not have the desired trait to produce progeny which can then be used in further crossing experiments to produce the desired trait in progeny of the second plant. Therefore, the resulting progeny plants contain no fransgenes; expression of the endogenous sequence may also be regulated by treatment with a particular chemical or other means, such as EMR. Examples of such well-known compounds include: ethylene; cytokinins; phenolic compounds, which stimulate the transcription of the genes needed for infection; specific monosaccharides and acidic environments which potentiate vir gene induction; acidic polysaccharides which induce one or more chromosomal genes; and opines; other mechanisms include light or dark treatment (an example of a review of examples of such treatments is found in Winans (1992) Microbiol. Rev. 56: 12-31 ; Eyal et al. (1992) Plant Mol. Biol. 19: 589-599; Chrispeels et al. (2000) Plant Mol. Biol. 42: 279-290; and Piazza et al. (2002) Plant Physiol. 128: 1077-1086). Table 6 lists sequences discovered to be orthologous to transcription factors of the present invention, in decreasing order of similarity to G867. The column headings include the transcription factors listed by (a) the SEQ ID NO: of the homolog (paralog or ortholog) or the nucleotide encoding the homolog; (b) the GID sequence identifier; (c) the Sequence Identifier or GenBank Accession Number; (d) the species from which the homologs (orthologs or paralogs) to the transcription factors are derived; and (e) the smallest sum probability relationship to G867 determined by BLAST analysis. Table 6. Homologs of Representative Arabidopsis Transcription Factor Genes Identified using BLAST

Molecular Modeling Another means that may be used to confirm the utility and function of transcription factor sequences that are orthologous or paralogous to presently disclosed transcription factors is through the use of molecular modeling software. Molecular modeling is routinely used to predict polypeptide structure, and a variety of protein structure modeling programs, such as "Insight II" (Accelrys, Inc.) are commercially available for this purpose. Modeling can thus be used to predict which residues of a polypeptide can be changed without altering function (Crameri et al. (2003) U.S. Patent No. 6, 521, 453). Thus, polypeptides that are sequentially similar can be shown to have a high likelihood of similar function by their structural similarity, which may, for example, be established by comparison of regions of superstructure. The relative tendencies of amino acids to form regions of superstructure (for example, helixes and _-sheets) are well established. For example, O'Neil et al. ((1990) Science 250: 646-651) have discussed in detail the helix forming tendencies of amino acids. Tables of relative structure forming activity for amino acids can be used as substitution tables to predict which residues can be functionally substituted in a given region, for example, in DNA-binding domains of known transcription factors and equivalogs. Homologs that are likely to be functionally similar can then be identified. Of particular interest is the structure of a transcription factor in the region of its conserved domains, such as those identified in Table 1. Structural analyses may be performed by comparing the structure of the known transcription factor around its conserved domain with those of orthologs and paralogs. Analysis of a number of polypeptides within a transcription factor group or clade, including the functionally or sequentially similar polypeptides provided in the Sequence Listing, may also provide an understanding of structural elements required to regulate transcription within a given family.

EXAMPLES The invention, now being generally described, will be more readily understood by reference to the following examples, which are included merely for purposes of illustration of certain aspects and embodiments of the present invention and are not intended to limit the invention. It will be recognized by one of skill in the art that a transcription factor that is associated with a particular first trait may also be associated with at least one other, unrelated and inherent second trait which was not predicted by the first trait. The complete descriptions of the traits associated with each polynucleotide of the invention are fully disclosed in Example VIII. The complete description of the transcription factor gene family and identified AP2 binding domains and B3 domains of the polypeptide encoded by the polynucleotide is fully disclosed in Table 1. Example I: Full Length Gene Identification and Cloning Putative transcription factor sequences (genomic or ESTs) related to known transcription factors were identified in the Arabidopsis thaliana GenBank database using the tblastn sequence analysis program using default parameters and a P-value cutoff threshold of -4 or -5 or lower, depending on the length of the query sequence. Putative transcription factor sequence hits were then screened to identify those containing particular sequence strings. If the sequence hits contained such sequence strings, the sequences were confirmed as transcription factors. Alternatively, Arabidopsis thaliana cDNA libraries derived from different tissues or treatments, or genomic libraries were screened to identify novel members of a transcription family using a low stringency hybridization approach. Probes were synthesized using gene specific primers in a standard PCR reaction (annealing temperature 60° C) and labeled with ²P dCTP using the High Prime DNA Labeling Kit (Roche Diagnostics Corp., Indianapolis, IN). Purified radiolabelled probes were added to filters immersed in Church hybridization medium (0.5 M NaP0₄ pH 7.0, 7% SDS, 1% w/v bovine serum albumin) and hybridized overnight at 60° C with shaking. Filters were washed two times for 45 to 60 minutes with 1 x SCC, 1 % SDS at 60° C. To identify additional sequence 5' or 3' of a partial cDNA sequence in a cDNA library, 5' and 3' rapid amplification of cDNA ends (RACE) was performed using the MARATHON cDNA amplification kit (Clontech, Palo Alto, CA). Generally, the method entailed first isolating poly(A) mRNA, performing first and second strand cDNA synthesis to generate double-stranded cDNA, blunting cDNA ends, followed by ligation of the MARATHON Adaptor to the cDNA to form a library of adaptor-ligated ds cDNA. Gene-specific primers were designed to be used along with adaptor specific primers for both 5' and 3' RACE reactions. Nested primers, rather than single primers, were used to increase PCR specificity. Using 5 ' and 3 ' RACE reactions, 5 ' and 3 ' RACE fragments were obtained, sequenced and cloned. The process can be repeated until 5' and 3' ends of the full-length gene were identified. Then the full-length cDNA was generated by PCR using primers specific to 5' and 3' ends of the gene by end-to-end PCR.

Example II: Construction of Expression Vectors The sequence was amplified from a genomic or cDNA library using primers specific to sequences upstream and downstream of the coding region. For direct promoteπtranscription factor gene fusions, the expression vector was pMEN20 or pMEN65, which are both derived from pMON316 (Sanders et al. (1987) Nucleic Acids Res. 15:1543-1558) and contain the CaMV 35S promoter to express transgenes. To clone the sequence into the vector, both pMEN20 and the amplified DNA fragment were digested separately with Sail and Notl restriction enzymes at 37° C for 2 hours. The digestion products were subject to elecfrophoresis in a 0.8% agarose gel and visualized by ethidium bromide staining. The DNA fragments containing the sequence and the linearized plasmid were excised and purified by using a QIAQUICK gel extraction kit (Qiagen, Valencia C A). The fragments of interest were ligated at a ratio of 3:1 (vector to insert). Ligation reactions using T4 DNA ligase (New England Biolabs, Beverly MA) were carried out at 16° C for 16 hours. The ligated DNAs were transformed into competent cells of the E. coli strain DH5α by using the heat shock method. The transformations were plated on LB plates containing 50 mg/1 kanamycin (Sigma Chemical Co., St. Louis MO). Individual colonies were grown overnight in five milliliters of LB broth containing 50 mg/1 kanamycin at 37° C. Plasmid DNA was purified by using Qiaquick Mini Prep kits (Qiagen). For the two-component system, two separate constructs were used: Promoter:: Lex A-GAL4T A and opLexA::TF. The first of these (Promoter:: Lex A-GAL4T A) comprised a desired promoter cloned in front of (that is, at a position that is upstream or closer to the 5' end of the promoter region) a LexA DNA binding domain fused to a GAL4 activation domain. The construct vector backbone (pMEN48; P5375) also carried a kanamycin resistance marker, along with an opLexA::GFP reporter. Transgenic lines were obtained containing this first component, and a line was selected that showed reproducible expression of the reporter gene in the desired pattern through a number of generations. A homozygous population was established for that line, and the population was supertransformed with the second construct

(opLexA::TF) carrying the transcription factor of interest cloned behind or after (that is, at a position that is downstream or closer to the 3' end of the operator) a LexA operator site. This second construct vector backbone (pMEN53; P5381) also contained a sulfonamide resistance marker. One example of a desired promoter includes, but is not limited to, the RD29A promoter. Each of the above methods offered a number of pros and cons. A direct fusion approach allowed for much simpler genetic analysis if a given promoter-transcription factor line was to be crossed into different genetic backgrounds at a later date. The two-component method, on the other hand, potentially allowed for stronger expression to be obtained via an amplification of transcription. A range of two- component constructs were available at the start of this research program.

Example in: Transformation of Agrobacterium with the Expression Vector After the expression constructs were generated, the constructs were used to transform Agrobacterium tumefaciens cells expressing the gene products. The stock of Agrobacterium tumefaciens cells for transformation was made as described by Nagel et al. (1990) FEMS Microbiol Letts. 67: 325- 328. Agrobacterium strain ABI was grown in 250 ml LB medium (Sigma) overnight at 28°C with shaking until an absorbance over 1 cm at 600 nm (A₆₀o) of 0.5 - 1.0 was reached. Cells were harvested by centrifugation at 4,000 x g for 15 min at 4°C. Cells were then resuspended in 250 μl chilled buffer (1 mM HEPES, pH adjusted to 7.0 with KOH). Cells were centrifuged again as described above and resuspended in 125 μl chilled buffer. Cells were then centrifuged and resuspended two more times in the same HEPES buffer as described above at a volume of 100 μl and 750 μl, respectively. Resuspended cells were then distributed into 40 μl aliquots, quickly frozen in liquid nitrogen, and stored at -80° C. Agrobacterium cells were transformed with constructs prepared as described above following the protocol described by Nagel et al. (1990) supra. For each DNA construct to be transformed, 50 - 100 ng DNA (generally resuspended in 10 mM Tris-HCl, 1 mM EDTA, pH 8.0) was mixed with 40 μl of Agrobacterium cells. The DNA/cell mixture was then transferred to a chilled cuvette with a 2 mm electrode gap and subject to a 2.5 kV charge dissipated at 25 μF and 200 μF using a Gene Pulser H apparatus (Bio-Rad, Hercules, CA). After electroporation, cells were immediately resuspended in 1.0 ml LB and allowed to recover without antibiotic selection for 2 - 4 hours at 28° C in a shaking incubator. After recovery, cells were plated onto selective medium of LB broth containing 100 μg/ml spectinomycin (Sigma) and incubated for 24-48 hours at 28° C. Single colonies were then picked and inoculated in fresh medium. The presence of the plasmid construct was verified by PCR amplification and sequence analysis.

Example IV: Transformation of Arabidopsis Plants with Agrobacterium tumefaciens After transformation of Agrobacterium tumefaciens with the constructs or plasmid vectors containing the gene of interest, single Agrobacterium colonies were identified, propagated, and used to transform Arabidopsis plants. Briefly, 500 ml cultures of LB medium containing 50 mg/1 kanamycin were inoculated with the colonies and grown at 28° C with shaking for 2 days until an optical absorbance at 600 nm wavelength over 1 cm (A₆₀o) of > 2.0 is reached. Cells were then harvested by centrifugation at 4,000 x g for 10 min, and resuspended in infiltration medium (1/2 X Murashige and Skoog salts (Sigma), 1 X Gamborg's B-5 vitamins (Sigma), 5.0% (w/v) sucrose (Sigma), 0.044 μM benzylamino purine (Sigma), 200 μl/1 Silwet L-77 (Lehle Seeds) until an A₆₀o of 0.8 was reached. Prior to transformation, Arabidopsis thaliana seeds (ecotype Columbia) were sown at a density of ~10 plants per 4" pot onto Pro-Mix BX potting medium (Hummert International) covered with fiberglass mesh (18 mm X 16 mm). Plants were grown under continuous illumination (50-75 μE/m²/sec) at 22-23° C with 65-70% relative humidity. After about 4 weeks, primary inflorescence stems (bolts) are cut off to encourage growth of multiple secondary bolts. After flowering of the mature secondary bolts, plants were prepared for fransformation by removal of all siliques and opened flowers. The pots were then immersed upside down in the mixture of Agrobacterium infiltration medium as described above for 30 sec, and placed on their sides to allow draining into a 1 ' x 2' flat surface covered with plastic wrap. After 24 h, the plastic wrap was removed and pots are turned upright. The immersion procedure was repeated one week later, for a total of two immersions per pot. Seeds were then collected from each transformation pot and analyzed following the protocol described below.

Example V: Identification of Arabidopsis Primary Transformants Seeds collected from the transformation pots were sterilized essentially as follows. Seeds were dispersed into in a solution containing 0.1% (v/v) Triton X-100 (Sigma) and sterile water and washed by shaking the suspension for 20 min. The wash solution was then drained and replaced with fresh wash solution to wash the seeds for 20 min with shaking. After removal of the ethanol/detergent solution, a solution containing 0.1% (v/v) Triton X-100 and 30% (v/v) bleach (CLOROX; Clorox Corp. Oakland CA) was added to the seeds, and the suspension was shaken for 10 min. After removal of the bleach detergent solution, seeds were then washed five times in sterile distilled water. The seeds were stored in the last wash water at 4° C for 2 days in the dark before being plated onto antibiotic selection medium (1 X Murashige and Skoog salts (pH adjusted to 5.7 with IM KOH), 1 X Gamborg's B-5 vitamins, 0.9% phytagar (Life Technologies), and 50 mg/1 kanamycin). Seeds were germinated under continuous illumination (50-75 μE/m²/sec) at 22-23° C. After 7-10 days of growth under these conditions, kanamycin resistant primary transformants (TI generation) were visible and obtained. These seedlings were fransferred first to fresh selection plates where the seedlings continued to grow for 3-5 more days, and then to soil (Pro-Mix BX potting medium). Primary transformants were crossed and progeny seeds (T₂) collected; kanamycin resistant seedlings were selected and analyzed. The expression levels of the recombinant polynucleotides in the transformants varies from about a 5% expression level increase to a least a 100% expression level increase. Similar observations are made with respect to polypeptide level expression.

Example VI: Identification of Arabidopsis Plants with Transcription Factor Gene Knockouts The screening of insertion mutagenized A rabidopsis collections for null mutants in a known target gene was essentially as described in Krysan et al. (1999) Plant Cell 11 : 2283-2290. Briefly, gene- specific primers, nested by 5-250 base pairs to each other, were designed from the 5' and 3' regions of a known target gene. Similarly, nested sets of primers were also created specific to each of the T-DNA or transposon ends (the "right" and "left" borders). All possible combinations of gene specific and T- DNA/transposon primers were used to detect by PCR an insertion event within or close to the target gene. The amplified DNA fragments were then sequenced which allows the precise determination of the T- DNA/transposon insertion point relative to the target gene. Insertion events within the coding or intervening sequence of the genes were deconvoluted from a pool comprising a plurality of insertion events to a single unique mutant plant for functional characterization. The method is described in more detail in Yu and Adam, US Application Serial No. 09/177,733 filed October 23, 1998.

Example VH: Identification of Modified Phenotypes in Overexpressing Plants. In some instances, expression patterns of the stress-induced genes may be monitored by microarray experiments. In these experiments, cDNAs are generated by PCR and resuspended at a final concentration of ~ 100 ng/ μl in 3X SSC or 150 mM Na-phosphate (Eisen and Brown (1999) Methods Enzymol. 303: 179-205). The cDNAs are spotted on microscope glass slides coated with polylysine. The prepared cDNAs are aliquoted into 384 well plates and spotted on the slides using, for example, an x-y-z gantry (OmniGrid) which may be purchased from GeneMachines (Menlo Park, CA) outfitted with quill type pins which may be purchased from Telechem International (Sunnyvale, CA). After spotting, the arrays are cured for a minimum of one week at room temperature, rehydrated and blocked following the protocol recommended by Eisen and Brown (1999) supra. Sample total RNA (10 μg) samples are labeled using fluorescent Cy3 and Cy5 dyes. Labeled samples are resuspended in 4X SSC/0.03% SDS/4 μg salmon sperm DNA/2 μg tRNA/ 50mM Na- pyrophosphate, heated for 95° C for 2.5 minutes, spun down and placed on the array. The array is then covered with a glass coverslip and placed in a sealed chamber. The chamber is then kept in a water bath at 62° C overnight. The arrays are washed as described in Eisen and Brown (1999) supra, and scanned on a General Scanning 3000 laser scanner. The resulting files are subsequently quantified using IMAGENE, software (BioDiscovery, Los Angeles CA). Modified phenotypes observed for particular overexpressor plants may include increased or decreased abiotic stress tolerance or resistance. For a particular overexpressor that shows a less beneficial characteristic, such as reduced abiotic stress tolerance or resistance, it may be more useful to select a plant with a decreased expression of the particular transcription factor. For a particular knockout that shows a less beneficial characteristic, such as decreased abiotic stress tolerance, it may be more useful to select a plant with an increased expression of the particular transcription factor. The germination assays in this example followed modifications of the same basic protocol. Sterile seeds were sown on the conditional media listed below. Plates were incubated at 22° C under 24-hour light (120-130 μEin/m²/s) in a growth chamber. Evaluation of germination and seedling vigor was conducted 3 to 15 days after planting. The basal media was 80% Murashige-Skoog medium (MS) + vitamins. For sfress experiments conducted with more mature plants, seeds were germinated and grown for seven days on MS + vitamins + 1% sucrose at 22 °C and then transferred to cold and heat stress conditions. The plants were either exposed to cold stress (6 hour exposure to 4-8° C ), or heat stress (32° C was applied for five days, after which the plants were transferred back 22 °C for recovery and evaluated after 5 days relative to controls not exposed to the depressed or elevated temperature). The salt stress assays were intended to find genes that confer better germination, seedling vigor or growth in high salt. Evaporation from the soil surface causes upward water movement and salt accumulation in the upper soil layer where the seeds are placed. Thus, germination normally takes place at a salt concentration much higher than the mean salt concentration in the whole soil profile. Plants differ in their tolerance to NaCl depending on their stage of development, therefore seed germination, seedling vigor, and plant growth responses were evaluated. Osmotic sfress assays (including NaCl and mannitol assays) are intended to determine if an osmotic stress phenotype is NaCl-specific or if it is a general osmotic stress related phenotype. Plants tolerant to osmotic stress could also have more tolerance to drought and/or freezing. Experiments were performed to identify those transformants or knockouts that exhibited modified sugar-sensing. For such studies, seeds from transformants were germinated on media containing 5% glucose or 9.4% sucrose which normally partially restrict hypocotyl elongation. Plants with altered sugar sensing may have either longer or shorter hypocotyls than normal plants when grown on this media. Additionally, other plant traits may be varied such as root mass. For salt and osmotic stress germination experiments, the medium was supplemented with 150 mM NaCl or 300 mM mannitol. Growth regulator sensitivity assays were performed in MS media, vitamins, and either 0.3 μM ABA, 9.4% sucrose, or 5% glucose. Desiccation assays are intended to find genes that mediate better plant survival after short-term, severe water deprivation. Ion leakage may be measured. Desiccation tolerance indicates a drought- tolerant phenotype. Drought assays were performed to find genes that mediate better plant survival after short-term, severe water deprivation. Ion leakage is measured if needed. Positive osmotic stress tolerance results also support a drought-tolerant phenotype. Soil-based drought screens were performed with Arabidopsis plants overexpressing the transcription factors listed in the Sequence Listing, where noted below. Seeds from wild-type Arabidopsis plants, or plants overexpressing a polypeptide of the invention, were stratified for three days at 4° C in 0.1 % agarose. Fourteen seeds of each overexpressor or wild-type were then sown in three inch clay pots containing a 50:50 mix of vermiculite:perlite topped with a small layer of MetroMix 200 and grown for fifteen days under 24 hr light. Pots containing wild-type and overexpressing seedlings were placed in flats in random order. Drought stress was initiated by placing pots on absorbent paper for seven to eight days. The seedlings were considered to be sufficiently sfressed when the majority of the pots containing wild-type seedlings within a flat had become severely wilted. Pots were then re-watered and survival was scored four to seven days later. Plants were ranked against wild-type controls for each of two criteria: tolerance to the drought conditions and recovery (survival) following re- watering. At the end of the initial drought period, each pot was assigned a numeric value score depending on the above criteria. A low value was assigned to plants with an extremely poor appearance (i.e., the plants were uniformly brown) and a high value given to plants that were rated very healthy in appearance (i.e., the plants were all green). After the plants were rewatered and incubated an additional four to seven days, the plants were reevaluated to indicate the degree of recovery from the water deprivation treatment. An analysis was then conducted to determine which plants best survived water deprivation, identifying the transgenes that consistently conferred drought-tolerant phenotypes and their ability to recover from this treatment. The analysis was performed by comparing overall and within- flat tabulations with a set of statistical models to account for variations between batches. Several measures of survival were tabulated, including: (a) the average proportion of plants surviving relative to wild-type survival within the same flat; (b) the median proportion surviving relative to wild-type survival within the same flat; (c) the overall average survival (taken over all batches, flats, and pots); (d) the overall average survival relative to the overall wild-type survival; and (e) the average visual score of plant health before rewatering. Sugar sensing assays are intended to find genes involved in sugar sensing by germinating seeds on high concentrations of sucrose and glucose and looking for degrees of hypocotyl elongation. The germination assay on mannitol controls for responses related to osmotic stress. Sugars are key regulatory molecules that affect diverse processes in higher plants including germination, growth, flowering, senescence, sugar metabolism and photosynthesis. Sucrose is the major transport form of photosynfhate and its flux through cells has been shown to affect gene expression and alter storage compound accumulation in seeds (source-sink relationships). Glucose-specific hexose-sensing has also been described in plants and is implicated in cell division and repression of "famine" genes (photosynthetic or glyoxylate cycles). Seeds from transformants were germinated on high sugar-containing media (5% glucose, 9.4%) sucrose) that normally partially restrict hypocotyl elongation. Plants with altered sugar sensing may have either longer or shorter hypocotyls than normal plants when grown on this media. Additionally, other plant traits may be varied such as root mass. Temperature stress assays were carried out to find genes that confer better germination, seedling vigor or plant growth under temperature sfress (cold, freezing and heat). Temperature stress cold germination experiments were carried out at 8° C. Heat stress germination experiments were conducted at 32 °C to 37° C for 6 hours of exposure. Flowering time was measured by the number of rosette leaves present when a visible inflorescence of approximately 3 cm is apparent. Rosette and total leaf number on the progeny stem are tightly correlated with the timing of flowering (Koomneef et al. (1991) Mol. Gen. Genet. 229: 57-66).

The vernalization response was also measured. For vernalization treatments, seeds were sown to MS agar plates, sealed with micropore tape, and placed in a 4° C cold room with low light levels for 6-8 weeks. The plates were then transferred to the growth rooms alongside plates containing freshly sown non- vernalized controls. Rosette leaves were counted when a visible inflorescence of approximately 3 cm was apparent. The transcription factor sequences of the Sequence Listing, or those in the present Tables or Figures, and their equivalogs, can be used to prepare transgenic plants and plants with altered traits. The specific transgenic plants listed below are produced from the sequences of the Sequence Listing, as noted. The Sequence Listing provides exemplary polynucleotide and polypeptide sequences of the invention.

Example VHI: Genes that Confer Significant Improvements to Plants This example provides experimental evidence for increased tolerance to abiotic sfress of plants that are transformed and have increased expression of transcription factor polynucleotides of the invention, when the plants are subjected to abiotic stress. Increased tolerance to these stresses will generally result in increased yields as compared to wild type under conditions of stress. Examples of specific genes and homologs that confer improvements to plants overexpressing a number of members of the G867 clade of transcription factor polypeptides are noted below.

Results: As noted below, overexpression of G867 and many related sequences, including Arabidopsis sequences G9, G993, G1930, soy sequences G3451, G3452, G3455, and rice sequences G3389, G3390, and G3391 , has been shown to increase abiotic stress tolerance. In the tables in this and subsequent examples, empty cells indicate an observed stress response similar to wild-type controls, "+" indicates a phenotype more tolerant than wild-type controls, "++" indicates a highly tolerant phenotype, and "-" indicates stress tolerance that was less than wild-type controls; "supTfii" refers to a "supertransformation" approach in which a plant is transformed with a two vector method providing a high degree of overexpression.

G867 (SEQ ID NO: 1 and 2) Published information. There are six RAV-like proteins in Arabidopsis. One of them, G867, has been described in the literature as related to AB\3IV?\ (RAVI; Kagaya et al. (1999) Nucleic Acids Res. 27: 470-478) based on the presence of a B3 domain (which is also found in the ABI3NP1 family of transcription factors). GS61/RAV1 belongs to a small subgroup within the AP2/EREBP family of transcription factors, whose distinguishing characteristic is that its members contain a second DNA- binding domain, in addition to the conserved AP2 domain, that is related to the B3 domain of VP1/ABI3 (Kagaya et al. (1999) supra. Analyses using various deletion derivatives of the RAVI fusion protein showed that the two DNA-binding domains of G867, the AP2 and B3 domains, separately recognize each of two motifs that constitute a bipartite binding sequence, CAACA and CACCTG, respectively, and together cooperatively enhance the DNA-binding affinity and specificity of the transcription factor (Kagaya et al. (1999) supra). No functional data are available for GS61/RAV1. Experimental observations. Arabidopsis lines were generated in which G867 was overexpressed with a 35S direct promoter fusion construct (P383). These 35S::G867 plants displayed a number of pleiotropic and variable alterations in overall morphology relative to wild-type controls, including a reduction in overall size and alterations in leaf orientation. In some lines, changes in leaf shape, flowering time and non-specific floral abnormalities that reduced fertility were observed. Several lines had small and chlorotic seedlings and had a low germination efficiency. In our earlier genomics program, G867 overexpressing lines were shown to have increased seedling vigor in germination assays on both high salt and high sucrose containing media compared to wild-type controls. In later studies, indicated in the following table, these results were confirmed. Ten of the 35S::G867 direct-fusion lines were subjected to plate-based treatments. All of these lines out- performed wild-type controls in at least one or more of the assays, and particularly strong tolerance was observed to sodium chloride, sucrose, and ABA in germination tests, and to low temperatures in a growth assay. Two lines were also significantly more tolerant to drought stress in soil-based assays. In the present study, we also sought to test whether use of a two-component overexpression system would produce any strengthening of the phenotype relative to the use of a 35 S direct promoter- fusion. At various stages of growth, a number of lines were small in size and late developing. Other lines showed no consistent morphological differences with wild-type controls.

G9 (SEQ ID NO: 3 and 4) Published information. G9 was first identified in a partial cDNA clone, and the corresponding gene named RAP2.8 (Okamuro et al., 1997). It has also been named RAV2 (Kagaya et al. (1999) Nucleic Acids Res. 27: 470-478). G9IRAV2IRAP2.8 belongs to a small subgroup within the AP2/EREBP family of transcription factors, whose distinguishing characteristic is that its members contain a second DNA- binding domain, in addition to the conserved AP2 domain, that is related to the B3 domain of VP1/ABI3 (Kagaya et al. (1999) supra). It has been shown that the two DNA-binding domains of RAVI (another member of this subgroup of proteins) can separately recognize each of two motifs that constitute a bipartite binding sequence and together cooperatively enhance its DNA-binding affinity and specificity (Kagaya et al. (1999) supra). No functional data are available for G9/RAV2/RAP2.8 or RAVI. Experimental observations. The complete sequence of G9 was determined. G9 appeared to be constitutively expressed. However, overexpression of G9 caused phenotypic changes in the roots: more root growth on MS plates, and hairy roots on media containing 10 μM methyl jasmonate (MeJ). Increased seedling vigor, manifested by increased expansion of the cotyledons of G9 overexpressing plants, was observed in gennination assays on both high salt (150 mM NaCl) and high sucrose-containing media (9.4% sucrose) , as compared to wild-type controls. A number of these lines were also insensitive to ABA. 35S::G9 transgenic plants were more tolerant to chilling (4° -8° C for 6 h) compared to the wild-type controls in seedling growth assays (Table 8). Several G9 overexpressing lines were found to be more sensitive to 0.3 μM ABA. Three lines were significantly more tolerant to drought stress in soil-based assays.

Table 8. 35S::G9 Arabido sis Constitutive Promoter Abiotic Stress Assa Results

G993 (SEQ ID NO: 5 and 6) Published information. G993 corresponds to gene F2J7.3 (AAG12735). No information is available about the function(s) of G993. Experimental observations. The function of G993 was studied using transgenic plants in which the gene was expressed under the control of the 35 S promoter. Overexpression of G993 produced highly pleiotropic effects on plant development and influenced Growth rate, overall plant size, branching pattern and fertility. 35S::G993 seedlings were small, developed slowly, and produced inflorescences markedly later than wild-type controls. They also showed a reduction in apical dominance and disorganized rosettes, as multiple axillary shoots developed simultaneously. Inflorescence stems were generally shorter than wild type, and produced an increased number of cauline leaf nodes leading to a leafy, bushy, appearance. In addition, the seed yield of 35S::G993 plants was generally very poor, and senescence occurred later than in wild-type controls. The transformation rate attained with the G993 construct was relatively low, suggesting that high levels of G993 activity might produce lethal effects. No alterations were detected in 35S::G993 plants in the biochemical analyses that were performed. G993 is ubiquitously expressed and does not appear to be significantly induced by any of the conditions tested. Increased seedling vigor, manifested by increased expansion of the cotyledons of G993 overexpressing plants, was observed in germination assays on both high salt (150 mM) and high sucrose (9.4%) containing-media, as compared to wild-type controls. In addition, several 35S::G993 transgenic lines were more tolerant to cold germination (8° C) and numerous lines were more tolerant to chilling (4° -8° C for 6 h) compared to the wild-type confrols, in both germination as well as seedling growth assays, respectively.

G1930 (SEQ ID NO: 7 and 8) Published information. G1930 was identified in the sequence of PI clone K13N2 (gene K13N2.7, GenBank protein accession number BAA95760). No information is available about the function(s) of G1930. Experimental observations. G1930 is ubiquitously expressed and does not appear to be induced by any of the conditions tested._.The function of G 1930 was studied using transgenic plants in which this gene was expressed under the confrol of the 35S promoter. 35S::G1930 TI plants were generally small and developed spindly inflorescences. The fertility of these plants was low and flowers often failed to open or pollinate. G1930 overexpressors were more tolerant to osmotic stress. The plants responded to high NaCl (150 mM) and high sucrose (9.4%) on plates with more seedling vigor compared to wild-type confrols. In addition, an increase in the amount of chlorophylls a and b in seeds of two T2 lines was detected. Several 35S::G1930 transgenic lines were more tolerant to cold germination conditions (8° C for 6 h) and numerous transgenic lines overexpressing G1930 were more tolerant to chilling (4° -8° C for 6 h) compared to the wild-type controls, in both germination as well as seedling growth assays, respectively. Several G1930 overexpressing lines were found to be more sensitive to 0.3 μM ABA.

G3389 (SEQ ID NO: 31 and 32) G3389 is a rice ortholog of G867. This project determined whether G3389 has an equivalent function to G867 or other G867 clade members via the analysis of 35S::G3389 Arabidopsis lines. Experimental observations. Line 341 was slightly smaller than the wild-type controls. Lines 342- 344 were moφhologically similar to wild-type controls. Lines 341 and 344 flowered slightly early. Results of the abiotic stress assays showed that these overexpressors were more tolerant of high NaCl concentrations, and one line (#341) was also more tolerant to high sucrose and insensitive to ABA.

G3451 (SEQ ID NO: 17 and 18) G3451 is a soy gene that was identified as an ortholog of G867. The aim of this project was to determine whether G3451 had an equivalent function to G867 by analysis of 35S::G3451 Arabidopsis lines. A number of lines overexpressing a soybean ortholog of G867, G3451 lines have recently been isolated. These seedlings showed a reduction in overall size, similar to what was obtained for 35S::G867 lines. Experimental observations. Table 12 displays the increased tolerance to osmotic stress exhibited by G3451 overexpressing plants. A number of 35S::G3451 lines were more tolerant to high salt, high mannitol concentration, and particularly high sucrose concentration, than wild-type control plants that did not overexpress this gene when the overexpressors and wild-type confrol Arabidopsis plants were exposed to the same conditions. All three of the lines tested in a soil-based drought assay survived water deprivation treatment significantly better than wild-type control plants, and two of these lines recovered better than controls to a significant degree after rewatering.

G3452 (SEQ ID NO: 21 and 22) G3452 is a soy gene that was identified as being an ortholog of G867. The aim of this project was to determine whether G3452 has an equivalent function to G867 by analysis of 35S::G3452 Arabidopsis lines. Experimental observations. Overexpression of G3452 in Arabidopsis produced a reduction in overall size compared to wild-type. These 35S::G3452 lines also appeared pale in coloration. Such effects were highly penefrant and were observed in all eighteen of the TI lines obtained. A small number of the lines (e.g., #314, 316) also exhibited a slight acceleration in the onset of flowering.

Table 13. 35S::G3452 (soy) Direct Fusion Constitutive Promoter Abiotic Stress Assay Results

G3453 (SEQ ID NO: 23 and 24) G3453 is a soy ortholog of G867. Experimental observations. A number of 35S:: Some of the G3453 Arabidopsis lines were somewhat smaller than wild-type control plants at various stages of growth. Other plants show considerable size variation but no consistent moφhological differences to wild-type controls.

G3390 (SEQ ID NO: 33 and 34) G3390 is a rice gene identified as an ortholog of G867. Experimental observations. 35S::G3390 Arabidopsis lines were generally smaller and flowered later than wild-type controls, including line 304, which may have performed better than the other lines in these abiotic stress assays. However, a number of lines did not show any significant moφhological differences relative to wild-type controls.

Table 15. 35S::G3390 rice Direct Fusion Constitutive Promoter Abiotic Stress Assay Results

G3391 (SEQ ID NO: 35 and 36) G3391 is a rice ortholog of G867. The aim of this project was to determine whether G3391 has an equivalent function to G867 or other G867 clade members by analyzing 35S::G3391 Arabidopsis lines. Experimental observations. 35S::G3391 Arabidopsis lines were distinctly small and showed alterations in leaf shape, leaf orientation, flowering time, and floral defects that resulted in poor fertility. Of the lines for which results are shown in Table 16, all were markedly small, with narrow pointed leaves. Lines #322, 323, 327, 329, 332, 334 were early flowering. All lines had poor seed yield. A number of other lines were tiny and dark in coloration at the seedling stages. As they continued to develop, these plants remained small relative to wild-type confrols, and had pointed upright leaves. A number of lines produced plants that were early flowering. All lines showed poor fertility and yielded relatively few seeds.

Table 16. 35S::G3391 (rice) Direct Fusion Constitutive Promoter Abiotic Stress Assay Results

G3455 (SEQ ID NO: 27 and 28) G3455 is a soy sequence identified as an ortholog of G867. This project determined that G3455 has an equivalent function to G867 by analyzing 35S::G3391 Arabidopsis lines. Experimental observations. A number of 35S:: G3455 Arabidopsis lines were somewhat smaller than wild-type control plants at various stages of growth. The former tended to be marginally late developing with partially glabrous leaves. Other plants show considerable size variation but no consistent moφhological differences to wild-type controls. Almost all of the 35S::G3455 lines were more tolerant to high sucrose concentration, and all lines were more tolerant to cold growth conditions than wild-type control plants.

Table 17. 35S::G3455 so Direct Fusion Constitutive Promoter Abiotic Sfress Assa Results

G3432 (SEQ ID NO: 37 and 38) G3432 is a com ortholog of G867. this project determined whether G3432 has an equivalent function to G867 by analyzing 35S:: G3432 Arabidopsis lines. Experimental observations. A significant number of 35S:: G3432 Arabidopsis lines were late developing, spindly and poorly fertile. Some plants, including line 362, showed no consistent differences to wild-type controls. The abiotic stress data indicate that G3432 confers tolerance in only a few assays when the polypeptide is overexpressed. These borderline results may reflect the relatively low similarity of the DML motif (71% identity) to the analogous motif in G867. One line was significantly more tolerant to drought stress in soil-based assays.

Table 18. 35S::G3432 (com) Direct Fusion Constitutive Promoter Abiotic Stress Assay Results

Utilities of the G867 Clade of transcription factor polypeptides The enhanced performance of Arabidopsis plants constitutively overexpressing G867 and other clade members seedlings in various abiotic stress assays indicates that these sequences or their equivalogs might be applied to engineer crops that show better growth under abiotic stress conditions, which may extend a crops planting season or range, or improve yield or performance. Most ABA effects are related to the compound acting as a signal of decreased water availability, triggering a reduction in water loss, slowed growth, and mediation of adaptive responses. Thus increased ABA sensitivity is a likely indicator of an enhanced stress response. These observations and the salt and osmotic stress assays indicate that G867 or its equivalogs can increase or facilitate seed germination and seedling or plant growth under adverse conditions such osmotic stresses, including drought and salt stress, which may extend a crop's planting season or range, or improve yield or performance. We have obtained comparable developmental effects as well as a strong enhancement of drought- related stress tolerance in plants overexpressing lines for the G867 and its Arabidopsis paralogs, as well as soy com and rice orthologs. The similar phenotypic effects observed for these G867 clade sequences from diverse species indicate that they are functionally equivalent. The accelerated flowering in 35S::3391 and G3452 plants suggested that these sequences act to promote the floral transition, and could be used to manipulate flowering time. In particular, shortening generation times would also help speed-up breeding programs, particularly in species such as trees, which typically grow for many years before flowering. Conversely, it might be possible to modify the activity of G3452 (or its equivalogs) to delay flowering in order to achieve an increase in biomass and yield.

Example IX: Mitigation of Undesirable Morphological Effects Caused by Overexpression of G867 and other Clade Members. The abiotic stress results shown in Example VIII provide evidence that members of the G867 clade of transcription factor polypeptides may be used to create plants with the characteristics of improved yield, performance and/or range. However, overexpression of these clade members may also bring about undesirable moφhological effects, including low fertility and smaller plant size. This was observed with many, but not all, of the lines generated in the present study. Since it is often desirable to generate plants with wild-type or near-wild-type stature, overexpression of G867 or other clade members under the regulatory control of a constitutive promoter may not always be the optimal approach for improving the abiotic stress tolerance of plants. This present study also included an investigation in the use of alternative promoter or two- component overexpression systems for the puφose of conferring enhanced stress tolerance and eliminating developmental abnormalities such as reduced size that were associated with G867 constitutive overexpression. In this regard, the present invention also relates to methods and compositions for producing transgenic plants with improved stress performance achieved by altering the expression of G867 and homologous sequences with specific promoter-gene combinations or other regulatory means. These combinations may regulate transcription factor expression patterns in a transient, inducible, or organ- or tissue-specific manner. This approach may be used to generate plants that are moφhologically similar to wild-type control plants that have not been transformed with a polynucleotide encoding G867 or another member of the G867 clade of transcription factor polypeptides. Thus, specific regulatory elements may be used to control regulation of a G867 clade member transcription factor gene to alleviate undesirable developmental abnormalities or moφhology that would result from overexpressing of the same transcription factor genes with a constitutive promoter (e.g., the 35S promoter).

Experimental observations: SUC2 vascular specific promoter We isolated SUC2::G867 lines via both a direct-promoter fusion and a two-component approach. The two-component lines were generated by supertransforming an opLexA::G867 construct into a SUC2::LexA-GAL4TA promoter driver line. These lines appeared normal at all developmental stages. The promoter driver line used in this set of lines produced relatively low expression levels. A direct promoter-fusion construct (P21521) for SUC2::G867 was also available. Fourteen lines harboring this construct also showed no consistent differences to wild-type controls. All lines examined thus far displayed segregation on selection plates in the T2 generation that was compatible with transgene insertions at a single locus. Often 2-component lines tested in drought-related assays, seven showed enhanced resistance to sodium chloride in germination assays. Some of these lines also displayed better performance, to varying extents, in sucrose germination, cold germination, and plate-based desiccation assays. These effects were comparable, but slightly weaker that those shown by 35S::G867 plants in plate based assays. The particular SUC2 promoter driver line that was used in the lines isolated so far showed a relatively weak levels of expression, and that resistance might be enhanced by use of stronger promoter line.

Table 19. G867 (Arabidopsis) - Vascular SUC2 Promoter Abiotic Stress Assay Results

Experimental observations: RBCS3 Leaf specific Promoter We isolated RBCS3::G867 lines via a 2-component approach. Most RBCS3::G867 lines were moφhologically similar to wild-type. Results with abiotic stress tolerance tests are shown in Table 20.

Experimental observations: CUT1 Leaf specific Promoter CUT1 ::G867 lines were isolated with a 2-component method. Generally, CUT1 ::G867 lines were moφhologically indistinguishable from wild-type plants. Stress tolerance tests are shown in Table 21.

Experimental observations: ARSK1 Root-specific Promoter We isolated ARSK1 ::G867 lines with a 2-component approach. Several lines were small, but most were moφhologically similar to wild-type plants at most stages of growth. Results with plate-based abiotic stress tolerance tests are shown in Table 22. Three of three ARSK1::G867 overexpressors tested were more tolerant to drought than wild-type controls in soil-based assays.

Experimental observations: RD29A stress-inducible promoter We isolated RD29A::G867 lines with a two-component method with two different RD29A::LexA promoter driver lines. Line 2 had a higher level of background expression than line 5. Line 2 was observed to have constitutive basal expression of GFP, and a marked increase in GFP expression following onset of sfress. In contrast, line 5 exhibited very low background expression, although it still exhibited an up-regulation of expression following the onset of stress. However, the stress-induced levels of GFP expression observed in line 5 were lower than those observed for line 2. Most lines generated with line 2 and line 5 were slightly smaller than wild-type controls, but in other respects appeared normal. The size reduction in these lines was generally much less severe than in the 35S::G867 lines. Results with these abiotic stress tolerance tests are shown in Tables 23 (RD29A Line 5) and 24 (RD29A Line 2). One of the lines of RD29A Line 2 and one of the of RD29A Line 5 overexpressors were significantly more tolerant to drought than wild-type controls in soil-based assays.

Experimental observations with G867 (Arabidopsis) - artificial GAL4 activation domain (C-GAL4-TA) P21193 (SEQ ID NO: 59) is an overexpression construct encoding a G867 clone that has a GAL4 transactivation domain fused at the C terminus (35S::G867-GAL4). The construct carries a kanamycin resistance marker. Overexpression of this super-active form of G867, comprising a GAL4 transactivation domain fused to the C terminus of the protein, produced no consistent effects on Arabidopsis moφhology, except as noted below. Two batches of lines containing construct P21193 have so far been obtained: 521-531 and 641- 645. The majority of these plants appeared wild type, although several of lines (522, 523, 525) were noted to be small at early stages of development. Of the ten lines submitted for physiological assays, the following showed a segregation on kanamycin plates in the T2 generation that was compatible with the transgene being present at a single locus: 521, 523, 525, 528, 530, 531, 642, and 643. Lines 641 and 644 showed a segregation that was compatible with insertions at multiple loci. GAL4 super-activation of G867 produced three lines that performed markedly better than wild- type plants on plates containing sucrose in a germination assay. Two of these lines were also less sensitive to ABA in another germination assay, and two lines showed enhanced performance in a chilling growth assay. Two lines that showed enhanced abiotic stress tolerance (#523, 525) were small at early stages of development.

Table 25. G867 (Arabidopsis) - artificial GAL4 activation domain (C-GAL4-TA) Abiotic Stress Results

Experimental observations with G867 (Arabidopsis) - artificial GAL4 activation domain (N-GAL4-TA) P21201 (SEQ ID NO: 58) is an overexpression construct encoding a G867 clone that has a GAL4 transactivation domain fused at the N terminus (35S::GAL4-G867). The construct carries a kanamycin resistance marker. Overexpression of this super-active form of G867, comprising a GAL4 transactivation domain fused to the N terminus of the protein, produced no consistent effects on Arabidopsis moφhology. This result contrasts the effects of overexpression of the wild-type form of the G867 protein, which produces a marked reduction in overall size and other developmental abnormalities. Thus, it appears that the additional domain added at the N-terminus blocked the deleterious phenotypes. These 35S::G867-GAL4 lines have now been tested in plate based assays. Four of the ten lines were substantially more resistant to salt in germination assays than wild-type controls, one of these lines was more tolerant to drought in a plate-based assay, and a separate line was more tolerant to osmotic stress and chilling conditions. Therefore, compared to the regular 35S::G867 lines (where all of the lines tested showed abiotic sfress resistance), stress resistance phenotypes were seen at lower frequency in these 35S::G867-GAL4 lines. In this case, however, the lines had wild-type moφhologies. Table 26. G867 (Arabidopsis) - artificial GAL4 activation domain (N-GAL4-TA) Abiotic Stress Results

Utilities of GAL4 fusions, and inducible or tissue-specific regulation of G867 clade sequences Some of the undesirable moφhologies that arise from constitutive overexpression of these sequences might be optimized with the use of sequence modifications or tissue-specific or conditional promoters. Given that the G867 protein lacks any clearly identifiable activation or repression domains, we originally posited that it might be possible to enhance the efficacy of the protein by addition of an artificial GAL4 activation domain. Lines have been made that overexpress a version of G867 with a GAL4 fusion at either the N or the C terminus. Lines for the N-terminal fusion did not show any of the developmental phenotypes observed in the regular 35S::G867 lines. Lines for the C-terminal fusion, were also moφhologically wild type in the majority of cases, but a small number of lines showed a slight reduction in size at the seedling stage. The majority of 35S::G867-GAL4 lines gave a wild-type performance in physiology assays, but a number of lines displayed a markedly enhanced performance in stress assays. A small number of these transformants showed a slight to modest reduction in size, but the majority showed no consistent differences in moφhology compared to wild-type controls. These stress and moφhology results indicate that tissue-specific or conditional promoters or the addition of an artificial activation domain at the N-terminus or C-terminus alleviates size reduction while conferring stress tolerance. Example X. Analysis of Arabidopsis G867 Truncation Variants Because the four Arabidopsis RAV proteins G867, G9, G1930 and G993 could have common target promoters, it was thought that overexpression of a truncated form of G867 containing the DNA binding domain, might be used to generate a mutant phenotype by a dominant negative type mechanism. In this regard, two such "dominant negative" constructs were designed, one encoding a form of the protein that contained the AP2 domain, but not the B3 domain (construct P21275) and one containing the B3 domain, but not the AP2 domain (construct P21276). Lines have been obtained for each of two different G867 dominant negative constructs: P21275 (SEQ ID NO: 56) and P21276 (SEQ ID NO: 57). P21275 is an overexpression construct encoding a truncated version of the G867 protein containing the AP2 domain and DML motif but not the B3 domain of the G867 protein. This construct carries a kanamycin selection. P21276 is an overexpression construct encoding a truncated version of the G867 protein containing most of the DML motif and the B3 domain but not the AP2 domain of the G867 protein. This construct carries a kanamycin selection. Arabidopsis plants were transformed with P21275, a construct in which a truncated version of G867 comprising the AP2 domain. Plants overexpressing this construct under the regulatory control of the 35 S promoter exhibited no consistent differences in moφhology to wild-type controls. Arabidopsis plants were also transformed with P21276, a construct in which a truncated version of G867 containing the B3 domain, but not the AP2 domain. Plants overexpressing this truncated G867 under the regulatory control of the 35 S promoter showed a number pleiotropic but distinct alterations in moφhology. The plants generally formed narrow strap like leaves, were slightly reduced in overall size, had reductions in trichome density, showed increased activity of secondary shoot meristems (in the primary rosette leaf axils), and had abnormalities in shoot phyllotaxy. Some of the lines were also noted to flower early and develop rather more rapidly than wild type. Results of abiotic sfress assays (Tables 27 and 28) indicated that these deletion variants were able to confer abiotic stress tolerance in some of the lines transformed with P21275 and P21276.

Table 27. Abiotic stress results of Arabidopsis transformed with P21276 encoding a dominant negative deletion - truncated G867 protein containing the B3 domain but not the AP2 domain

Table 28. Abiotic stress results of Arabidopsis transformed with P21275 encoding a dominant negative deletion - truncated G867 protein containing the AP2 domain but not the B3 domain

Utilities of sequence modifications of G867 clade sequences Given the undesirable moφhologies that may arise from G867 overexpression, it might be advantageous to optimize the gene by use of alternative promoters or sequence modifications such as various truncations of the RAV proteins. Although a number of the lines transformed with truncated forms of the G867 polypeptide were significantly more abiotic stress tolerant than wild-type controls, many of these plants did not exhibit the adverse moφhological effects seen in plants constitutively overexpressing G867. These results indicated that such sequence modifications for reducing adverse moφhological effects in plants overexpressing G867 clade members may be effectively targeted outside of the AP2 domain.

Example XI: Identification of Homologous Sequences This example describes identification of genes that are orthologous to Arabidopsis thaliana transcription factors from a computer homology search. Homologous sequences, including those of paralogs and orthologs from Arabidopsis and other plant species, were identified using database sequence search tools, such as the Basic Local Alignment Search Tool (BLAST) (Altschul et al. (1990) J. Mol. Biol. 215: 403-410; and Altschul et al. (1997) Nucleic Acid Res. 25: 3389-3402). The tblastx sequence analysis programs were employed using the BLOSUM-62 scoring matrix (Henikoff and Henikoff (1992) Proc. Natl. Acad. Sci. 89: 10915-10919). The entire NCBI GenBank database was filtered for sequences from all plants except Arabidopsis thaliana by selecting all entries in the NCBI GenBank database associated with NCBI taxonomic ID 33090 (Viridiplantae; all plants) and excluding entries associated with taxonomic ID 3701 (Arabidopsis thaliana). These sequences are compared to sequences representing genes of SEQ ID NO: 1 , 3, 5, 7, 9, 11 , 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41 , 43, 45, 47, 49, 51, using the Washington University TBLASTX algorithm (version 2.0al9MP) at the default settings using gapped alignments with the filter "off. For each gene of SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, individual comparisons were ordered by probability score (P-value), where the score reflects the probability that a particular alignment occurred by chance. For example, a score of 3.6E-40 is 3.6 x 10-40. In addition to P-values, comparisons were also scored by percentage identity. Percentage identity reflects the degree to which two segments of DNA or protein are identical over a particular length. Examples of sequences so identified are presented in Table 6. The percent sequence identity among these sequences can be as low as 47%, or even lower sequence identity. Candidate paralogous sequences were identified among Arabidopsis transcription factors through alignment, identity, and phylogenic relationships. Paralogs of G867 determined in this manner include G9, G993 and GI 930. Candidate orthologous sequences were identified from proprietary unigene sets of plant gene sequences in Zea mays, Glycine max and Oryzα sαtivα based on significant homology to Arabidopsis transcription factors. These candidates were reciprocally compared to the set of Arabidopsis transcription factors. If the candidate showed maximal similarity in the protein domain to the eliciting transcription factor or to a paralog of the eliciting transcription factor, then it was considered to be an ortholog. Identified non-Arabidopsis sequences that were shown in this manner to be orthologous to the Arabidopsis sequences are provided in Table 6.

Example XII: Screen of Plant cDNA library for Sequence Encoding a Transcription Factor DNA Binding Domain That Binds To a Transcription Factor Binding Promoter Element and Demonstration of Protein Transcription Regulation Activity. The "one-hybrid" strategy (Li and Herskowitz (1993) Science 262: 1870-1874) is used to screen for plant cDNA clones encoding a polypeptide comprising a transcription factor DNA binding domain, a conserved domain. In brief, yeast strains are constructed that contain a lacZ reporter gene with either wild-type or mutant transcription factor binding promoter element sequences in place of the normal UAS (upstream activator sequence) of the GAL4 promoter. Yeast reporter strains are constructed that carry transcription factor binding promoter element sequences as UAS elements are operably linked upstream (5') of a lacZ reporter gene with a minimal GAL4 promoter. The strains are transformed with a plant expression library that contains random cDNA inserts fused to the GAL4 activation domain (GAL4- ACT) and screened for blue colony formation on X-gal-treated filters (X-gal: 5-bromo-4-chloro-3-indolyl- β-D-galactoside; Invitrogen Coφoration, Carlsbad CA). Alternatively, the strains are transformed with a cDNA polynucleotide encoding a known transcription factor DNA binding domain polypeptide sequence. Yeast strains carrying these reporter constructs produce low levels of β-galactosidase and form white colonies on filters containing X-gal. The reporter strains carrying wild-type transcription factor binding promoter element sequences are transformed with a polynucleotide that encodes a polypeptide comprising a plant transcription factor DNA binding domain operably linked to the acidic activator domain of the yeast GAL4 transcription factor, "GAL4-ACT". The clones that contain a polynucleotide encoding a transcription factor DNA binding domain operably linked to GAL4-ACT can bind upstream of the lacZ reporter genes carrying the wild-type transcription factor binding promoter element sequence, activate transcription of the lacZ gene and result in yeast forming blue colonies on X-gal-treated filters. Upon screening about 2 x 10⁶ yeast transformants, positive cDNA clones are isolated; i.e., clones that cause yeast strains carrying lacZ reporters operably linked to wild-type transcription factor binding promoter elements to form blue colonies on X-gal-treated filters. The cDNA clones do not cause a yeast strain carrying a mutant type transcription factor binding promoter elements fused to LacZ to turn blue. Thus, a polynucleotide encoding transcription factor DNA binding domain, a conserved domain, is shown to activate transcription of a gene.

Example XIII: Gel Shift Assays The presence of a transcription factor comprising a DNA binding domain which binds to a DNA transcription factor binding element is evaluated using the following gel shift assay. The transcription factor is recombinantly expressed and isolated from E. coli or isolated from plant material. Total soluble protein, including transcription factor, (40 ng) is incubated at room temperature in 10 μl of 1 x binding buffer (15 mM HEPES (pH 7.9), 1 mM EDTA, 30 mM KC1, 5% glycerol, 5% bovine serum albumin, 1 mM DTT) plus 50 ng poly(dl-dC):poly(dl-dC) (Pharmacia, Piscataway NJ) with or without 100 ng competitor DNA. After 10 minutes incubation, probe DNA comprising a DNA transcription factor binding element (1 ng) that has been ³²P-labeled by end-filling (Sambrook, supra) is added and the mixture incubated for an additional 10 minutes. Samples are loaded onto polyacrylamide gels (4% w/v) and fractionated by elecfrophoresis at 150V for 2h (Sambrook, supra). The degree of transcription factor- probe DNA binding is visualized using autoradiography. Probes and competitor DNAs are prepared from oligonucleotide inserts ligated into the BamHI site of pUCl 18 (Vieira et al. (1987) Methods Enzymol. 153: 3-11). Orientation and concatenation number of the inserts are determined by dideoxy DNA sequence analysis (Sambrook, supra). Inserts are recovered after restriction digestion with EcoRI and HindHI and fractionation on polyacrylamide gels (12% w/v) (Sambrook, supra).

Example XIV. Introduction of Polynucleotides into Dicotyledonous Plants SEQ ID NO: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21 , 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 49, 51, or polynucleotide sequences encoding SEQ ID NO: 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, 53, paralogous, and orthologous sequences, including truncated sequences and sequences combined with an artificial activation domain, are recombined into pMEN20 or pMEN65 expression vectors and transformed into a plant. The vector may be introduced into a variety of monocot plants by well-known means, including direct DNA transfer or Agrobacterium tumefaciens-mediated transformation. It is routine to produce transgenic plants using most dicot plants (for example, Weissbach and Weissbach, (1989) supra; Gelvin et al. (1990) supra; Herrera- Estrella et al. (1983) supra; Bevan (1984) supra; and Klee (1985) supra).

Example XV: Transformation of Cereal Plants with an Expression Vector Cereal plants such as, but not limited to, com, wheat, rice, sorghum, or barley, may also be transformed with the present polynucleotide sequences in pMEN20 or pMEN65 expression vectors for the puφose of modifying plant traits. For example, pMEN020 may be modified to replace the Nptπ coding region with the BAR gene of Streptomyces hygroscopicus that confers resistance to phosphinothricin. The Kpnl and Bglll sites of the Bar gene are removed by site-directed mutagenesis with silent codon changes. The cloning vector may be introduced into a variety of cereal plants by means well known in the art such as, for example, direct DNA transfer or Agrobacterium tumefaciens-mediated transformation. It is now routine to produce transgenic plants of most cereal crops (Vasil (1994) Plant Mol. Biol. 25: 925- 937) such as com, wheat, rice, sorghum (Cassas et al. (1993) Proc. Natl. Acad. Sci. 90: 11212-11216), and barley (Wan and Lemeaux (1994) Plant Physiol. 104:37-48). DNA transfer methods such as the microprojectile can be used for com (Fromm et al. (1990) Bio/Technol. 8: 833-839; Gordon-Kamm et al. (1990) Plant Cell 2: 603-618; and Ishida (1990) Nature Biotechnol. 14:745-750), wheat (Vasil et al. (1992) Bio/Technol. 10:667-674; Vasil et al. (1993) Bio/Technol. 11 :1553-1558; Weeks et al. (1993) Plant Physiol. 102:1077- 1084),and rice (Christou (1991) Bio/Technol. 9:957-962; Hiei et al. (1994) Plant J. 6:271-282; Aldemita and Hodges (1996) Planta 199:612-617; and Hiei et al. (1997) Plant Mol. Biol. 35:205-218). For most cereal plants, embryogenic cells derived from immature scutellum tissues are the preferred cellular targets for fransformation (Hiei et al. (1997) Plant Mol. Biol. 35:205-218; Vasil (1994) Plant Mol. Biol. 25: 925-937). Vectors according to the present invention may be transformed into com embryogenic cells derived from immature scutellar tissue by using microprojectile bombardment, with the Al 88XB73 genotype as the preferred genotype (Fromm et al. (1990) Bio/Technol. 8: 833-839; and Gordon-Kamm et al. (1990) Plant Cell 2: 603-618). After microprojectile bombardment the tissues are selected on phosphinothricin to identify the transgenic embryogenic cells (Gordon-Kamm et al. (1990) Plant Cell 2: 603-618). Transgenic plants are regenerated by standard com regeneration techniques (Fromm et al. (1990) Bio/Technol. 8: 833-839; and Gordon-Kamm et al. (1990) Plant Cell 2: 603-618). The plasmids prepared as described above can also be used to produce transgenic wheat and rice plants (Christou (1991) Bio/Technol. 9:957-962; Hiei et al. (1994) Plant J. 6:271-282; Aldemita and Hodges (1996) Planta 199:612-617; and Hiei et al. (1997) Plant Mol. Biol. 35:205-218) that coordinate^ express genes of interest by following standard transformation protocols known to those skilled in the art for rice and wheat (Vasil et al. (1992) Bio/Technol. 10:667-674; Vasil et al. (1993) Bio/Technol. 11:1553-1558; and Weeks et al. (1993) Plant Physiol. 102:1077-1084), where the bar gene is used as the selectable marker.

Example XVI: Transformation of Dicots Numerous protocols for the transformation of dicots, including tomato and soy plants have been previously described, and are well known in the art. Gruber et al. ((1993) in Methods in Plant Molecular Biology and Biotechnology, p. 89-1 19, Glick and Thompson, eds., CRC Press, Inc., Boca Raton) describe several expression vectors and culture methods that may be used for cell or tissue transformation and subsequent regeneration. For soybean transformation, methods are described by Miki et al. (1993) in Methods in Plant Molecular Biology and Biotechnology, p. 67-88, Glick and Thompson, eds., CRC Press, Inc., Boca Raton; and U.S. Pat. No. 5,563,055, (Townsend and Thomas), issued Oct. 8, 1996. There are a substantial number of alternatives to Agrobacterium-mediated transformation protocols, other methods for the puφose of transferring exogenous genes into soybeans or tomatoes. One such method is microprojectile-mediated transformation, in which DNA on the surface of microprojectile particles is driven into plant tissues with a biolistic device (for example, Sanford et al., (1987) Part. Sci. Technol. 5:27-37; Christou et al. (\992) Plant. J. 2: 275-281; Sanford (1993) Methods Enzymol. 217: 483-509; Klein et al. (1987) Nature 327: 70-73; U.S. Pat. No. 5,015,580 (Christou et al), issued May 14, 1991; and U.S. Pat. No. 5,322,783 (Tomes et al.), issued Jun. 21, 1994). Alternatively, sonication methods (for example, Zhang et al. (1991) Bio/Technology 9: 996-997); direct uptake of DNA into protoplasts using CaCl₂ precipitation, polyvinyl alcohol or poly-L-ornithine (for example, Hain et al. (1985) Mol. Gen. Genet. 199: 161-168; Draper et al., Plant Cell Physiol. 23: 451-458 (1982)); liposome or spheroplast fusion (for example, Deshayes et al. (1985) EMBOJ., 4: 2731- 2737; Christou et al. (1987) Proc. Natl. Acad. Sci. U.S.A. 84: 3962-3966); and electroporation of protoplasts and whole cells and tissues (for example, Dorm et al.(1990) in Abstracts of Vllth International Congress on Plant Cell and Tissue Culture IAPTC. A2-38: 53; D'Halluin et al. (1992) Plant Cell 4: 1495- 1505; and Spencer et al. (1994) Plant Mol. Biol. 24: 51-61) have been used to introduce foreign DNA and expression vectors into plants. After plants or plant cells are transformed (and the latter regenerated into plants) the transgenic plant thus generated may be crossed with itself ("selfing") or a plant from the same line, a non- transformed or wild-type plant, or another transformed plant from a different transgenic line of plants. Crossing provides the advantages of being able to produce new and perhaps stable transgenic varieties. Genes and the traits they confer that have been introduced into a tomato or soybean line may be moved into distinct line of plants using traditional backcrossing techniques well known in the art. Transformation of tomato plants may be conducted using the protocols of Koomneef et al (1986) In Tomato Biotechnology: Alan R. Liss, Inc., 169-178, and in U.S. Patent 6,613,962, the latter method described in brief here. Eight day old cotyledon explants are precultured for 24 hours in Petri dishes containing a feeder layer of Petunia hybrida suspension cells plated on MS medium with 2% (w/v) sucrose and 0.8% agar supplemented with 10 μM α-naphthalene acetic acid and 4.4 μM 6- benzylaminopurine. The explants are then infected with a diluted ovemight culture oϊ Agrobacterium tumefaciens containing an expression vector comprising a polynucleotide of the invention for 5-10 minutes, blotted dry on sterile filter paper and cocultured for 48 hours on the original feeder layer plates. Culture conditions are as described above. Overnight cultures oϊ Agrobacterium tumefaciens are diluted in liquid MS medium with 2% (w/v/) sucrose, pH 5.7) to an OD₆₀o of 0.8. Following the cocultivation, the cotyledon explants are transferred to Petri dishes with selective medium consisting of MS medium supplemented with 4.56 μM zeatin, 67.3 μM vancomycin, 418.9 μM cefotaxime and 171.6 μM kanamycin sulfate, and cultured under the culture conditions described above. The explants are subcultured every three weeks onto fresh medium. Emerging shoots are dissected from the underlying callus and transferred to glass jars with selective medium without zeatin to form roots. The formation of roots in a medium containing kanamycin sulfate is regarded as a positive indication of a successful transformation. Transformation of soybean plants may be conducted using the methods found in, for example, U.S. Patent 5,563,055 (Townsend et al, issued October 8,1996), described in brief here. In this method soybean seed is surface sterilized by exposure to chlorine gas evolved in a glass bell jar. Seeds are germinated by plating on 1/10 strength agar solidified medium without plant growth regulators and culturing at 28° C with a 16 hour day length. After three or four days, seed may be prepared for cocultivation. The seedcoat is removed and the elongating radicle removed 3-4 mm below the cotyledons. Overnight cultures oϊ Agrobacterium tumefaciens harboring the expression vector comprising a polynucleotide of the invention are grown to log phase, pooled, and concentrated by centrifugation.

Inoculations are conducted in batches such that each plate of seed was treated with a newly resuspended pellet oϊ Agrobacterium. The pellets are resuspended in 20 ml inoculation medium. The inoculum is poured into a Petri dish containing prepared seed and the cotyledonary nodes are macerated with a surgical blade. After 30 minutes the explants are transferred to plates of the same medium which has been solidified. Explants are embedded with the adaxial side up and level with the surface of the medium and cultured at 22° C for three days under white fluorescent light. These plants may then be regenerated according to methods well established in the art, such as by moving the explants after three days to a liquid counter-selection medium (for example, as presented in U.S. Patent 5,563,055). The explants may then be picked, embedded and cultured in solidified selection medium. After one month on selective media transformed tissue becomes visible as green sectors of regenerating tissue against a background of bleached, less healthy tissue. Explants with green sectors are fransferred to an elongation medium. Culture is continued on this medium with transfers to fresh plates every two weeks. When shoots are 0.5 cm in length they may be excised at the base and placed in a rooting medium.

Example XVTI: Genes that Confer Significant Improvements to non-Arabidopsis species The function of specific orthologs of G867 may be analyzed through their ectopic overexpression in plants, using the CaMV 35 S or other appropriate promoter, identified above. These genes, which include polynucleotide sequences found in the Sequence Listing, Table 6 and Figure 3, encode members of the AP2 transcription factors, such as those found in Oryza sativa (SEQ ED NO: 20, 30, 32, 34, 36, 52, and 53), Arabidopsis thaliana (SEQ ID NO 2, 4, 6, 8), Glycine max (SEQ ID NO: 18, 22, 24, 26, 28), Zea mays (SEQ ID NO: 38, 40, 50), Triticum aestivum (SEQ ID NO: 48), Brassica oleracea (SEQ ID NO: 42), and Helianthus annuus (SEQ ID NO: 44 and 46). The polynucleotide and polypeptide sequences derived from monocots may be used to transform both monocot and dicot plants, and those derived from dicots may be used to transform either class, although some of these sequences will function best if the gene is transformed into a plant from the same group as that from which the sequence is derived. Seeds of these transgenic plants are subjected to germination assays to measure sucrose sensing. Sterile monocot seeds, including, but not limited to, com, rice, wheat, rye and sorghum, as well as dicots including, but not limited to soybean and alfalfa, are sown on 80% MS medium plus vitamins with 9.4% sucrose; confrol media lack sucrose. All assay plates are then incubated at 22° C under 24-hour light, 120- 130 μEin m²/s, in a growth chamber. Evaluation of germination and seedling vigor is then conducted three days after planting. Overexpressors of these genes may be found to be more tolerant to high sucrose by having better germination, longer radicles, and more cotyledon expansion. These results would indicate that overexpressors of members of the G867 clade of transcription factor polypeptides are involved in sucrose-specific sugar sensing. These results have thus been determined in plants overexpressing rice G3389, G3391, com G3432, or soy G3451, G3452, or G3455. Plants overexpressing these orthologs may also be subjected to soil-based drought assays to identify lines that are more tolerant to water deprivation than wild-type control plants. Generally, plants that overexpress a member of the G867 clade of transcription factor orthologs will appear significantly larger and greener, with less wilting or desiccation, than wild-type controls plants, particularly after a period of water deprivation is followed by rewatering and a subsequent incubation period. Examples VIE, IX and X demonstrate that not all plants transformed with expression vectors encoding G867 clade polypeptides have been shown to be more tolerant to abiotic stresses than control plants. Thus, the use of a screening step to identify the plants that are more tolerant to these stresses may be advantageous.

Example XVIII: Identification of Orthologous and Paralogous Sequences Orthologs to Arabidopsis genes may be identified by several methods, including hybridization, amplification, or bioinformatically. This example describes how one may identify homologs to the Arabidopsis AP2 family transcription factor CBFl that confers tolerance to abiotic stresses (Thomashow et al. (2002) US Patent No. 6,417,428), and an example to confirm the function of homologous sequences. In this example, orthologs to CBFl were found in canola (Brassica napus) using polymerase chain reaction (PCR). Degenerate primers (US Patent No. 6,417,428) were designed for regions of AP2 binding domain and outside of the AP2 (carboxyl terminal domain): Mol 368 (reverse) 5'- CAY CCN ATH TAY MGN GGN GT -3' Mol 378 (forward) 5'- GGN ARN ARC ATN CCY TCN GCC -3' (Y: C/T, N: A/C/G/T, H: A/C/T, M: A/C, R: A G ) Primer Mol 368 is in the AP2 binding domain of CBFl (amino acid sequence: His-Pro-Ile-Tyr-

Arg-Gly-Val) while primer Mol 378 is outside the AP2 domain (carboxyl terminal domain) (amino acid sequence: Met-Ala-Glu-Gly-Met-Leu-Leu-Pro). The genomic DNA isolated from B. napus was PCR-amplified by using these primers following these conditions: an initial denaturation step of 2 min at 93° C; 35 cycles of 93° C for 1 min, 55° C for 1 min, and 72° C for 1 min ; and a final incubation of 7 min at 72° C at the end of cycling. The PCR products were separated by elecfrophoresis on a 1.2% agarose gel and transferred to nylon membrane and hybridized with the AT CBFl probe prepared from Arabidopsis genomic DNA by PCR amplification. The hybridized products were visualized by colorimetric detection system (Boehringer Mannheim) and the corresponding bands from a similar agarose gel were isolated using the Qiagen Extraction Kit (Qiagen). The DNA fragments were ligated into the TA clone vector from TOPO TA Cloning Kit (Invitrogen) and transformed into E. coli strain TOP 10 (Invitrogen). Seven colonies were picked and the inserts were sequenced on an ABI 377 machine from both strands of sense and antisense after plasmid DNA isolation. The DNA sequence was edited by sequencer and aligned with the AtCBFl by GCG software and NCBI blast searching. The nucleic acid sequence and amino acid sequence of one canola ortholog found in this manner

(bnCBFl; US Patent No. 6,417,428) identified by this process is shown in the Sequence Listing. The aligned amino acid sequences show that the bnCBFl gene has 88% identity with the Arabidopsis sequence in the AP2 domain region and 85% identity with the Arabidopsis sequence outside the AP2 domain when aligned for two insertion sequences that are outside the AP2 domain. Similarly, paralogous sequences to Arabidopsis genes, such as CBFl, may also be identified. Two paralogs of CBFl from Arabidopsis thaliana: CBF2 and CBF3. CBF2 and CBF3 have been cloned and sequenced as described below. A lambda cDNA library prepared from RNA isolated from Arabidopsis thaliana ecotype Columbia (Lin and Thomashow (1992) Plant Physiol. 99: 519-525) was screened for recombinant clones that carried inserts related to the CBFl gene (Stockinger et al. (1997) Proc. Natl. Acad. Sci. 94: 1035- 1040). CBFl was ³²P-radiolabeled by random priming (Sambrook, supra) and used to screen the library by the plaque-lift technique using standard stringent hybridization and wash conditions (Hajela et al. (1990) Plant Physiol. 93:1246-1252; and Sambrook, supra) 6 X SSPE buffer, 60° C for hybridization and 0.1 X SSPE buffer and 60° C for washes). Twelve positively hybridizing clones were obtained and the DNA sequences of the cDNA inserts were determined. The results indicated that the clones fell into three classes. One class carried inserts corresponding to CBFl. The two other classes carried sequences corresponding to two different homologs of CBFl, designated CBF2 and CBF3. The nucleic acid sequences and predicted protein coding sequences for Arabidopsis CBFl, CBF2, CBF3 and for the Brassica napus CBF ortholog may be found in US Patent No. 6,417,428. A comparison of the nucleic acid sequences of Arabidopsis CBFl, CBF2 and CBF3 indicate that they are 83 to 85% identical as shown in Table 29.

^a Percent identity was determined using the Clustal algorithm from the MEGALIGN program (DNASTAR, Inc.). ^b Comparisons of the nucleic acid sequences of the open reading frames are shown.

Similarly, the amino acid sequences of the three CBF polypeptides range from 84 to 86% identity. An alignment of the three amino acid sequences reveals that most of the differences in amino acid sequence occur in the acidic C-terminal half of the polypeptide. This region of CBFl serves as an activation domain in both yeast and Arabidopsis (not shown). Residues 47 to 106 of CBFl correspond to the AP2 domain of the protein, a DNA binding motif that to date, has only been found in plant proteins. A comparison of the AP2 domains of CBFl, CBF2 and CBF3 indicates that there are a few differences in amino acid sequence. These differences in amino acid sequence might have an effect on DNA binding specificity.

Example XIX: Transformation of Canola with a Plasmid Containing CBFl, CBF2, or CBF3 After identifying homologous genes to CBFl, canola was transfonned with a plasmid containing Arabidopsis CBFl, CBF2, or CBF3 genes cloned into vector pGA643 (An (1987) Methods Enzymol. 253: 292). In these constructs the CBF genes were expressed constitutively under the CaMV 35S promoter. In addition, the CBFl gene was cloned under the control of the Arabidopsis COR15 promoter in the same vector pGA643. Each construct was transformed into Agrobacterium strain GV3101. Transformed Agrobacteria were grown for 2 days in minimal AB medium containing appropriate antibiotics. Spring canola (B. napus cv. Westar) was transformed using the protocol of Moloney et al. ((1989) Plant Cell Reports 8: 238) with some modifications as described. Briefly, seeds were sterilized and plated on half strength MS medium, containing 1% sucrose. Plates were incubated at 24° C under 60- 80 μE/m²s light using a 16 hour light/ 8 hour dark photoperiod. Cotyledons from 4-5 day old seedlings were collected, the petioles cut and dipped into the Agrobacterium solution. The dipped cotyledons were placed on co-cultivation medium at a density of 20 cotyledons/plate and incubated as described above for 3 days. Explants were transferred to the same media, but containing 300 mg/1 timentin (SmithKline Beecham, PA) and thinned to 10 cotyledons/plate. After 7 days explants were transferred to Selection/Regeneration medium. Transfers were continued every 2-3 weeks (2 or 3 times) until shoots had developed. Shoots were transferred to Shoot-Elongation medium every 2-3 weeks. Healthy looking shoots were fransferred to rooting medium. Once good roots had developed, the plants were placed into moist potting soil. The transformed plants were then analyzed for the presence of the NPTII gene/ kanamycin resistance by ELISA, using the ELISA NPTII kit from 5Prime-3Prime Inc. (Boulder, CO). Approximately 70% of the screened plants were NPTII positive. Only those plants were further analyzed. From Northern blot analysis of the plants that were transformed with the constitutively expressing constructs, showed expression of the CBF genes and all CBF genes were capable of inducing the Brassica napus cold-regulated gene BN115 (homolog oϊ the Arabidopsis COR15 gene). Most of the transgenic plants appear to exhibit a normal growth phenotype. As expected, the transgenic plants are more freezing tolerant than the wild-type plants. Using the electrolyte leakage of leaves test, the wild-type control showed a 50% leakage at -2 to -3° C. Spring canola transformed with either CBFl or CBF2 showed a 50% leakage at -6 to -7° C. Spring canola transformed with CBF3 shows a 50% leakage at about -10 to -15° C. Winter canola transformed with CBF3 may show a 50% leakage at about -16 to - 20° C. Furthermore, if the spring or winter canola are cold acclimated the transformed plants may exhibit a further increase in freezing tolerance of at least -2° C. To test salinity tolerance of the transformed plants, plants were watered with 150 mM NaCl. Plants overexpressing CBFl, CBF2 or CBF3 grew better compared with plants that had not been transformed with CBFl, CBF2 or CBF3. These results demonstrate that homologs oϊ Arabidopsis transcription factors can be identified and shown to confer similar functions in non-Arabidopsis plant species.

Example XX: Cloning of transcription factor promoters Promoters are isolated from transcription factor genes that have gene expression patterns useful for a range of applications, as determined by methods well known in the art (including transcript profile analysis with cDNA or oligonucleotide microarrays, Northern blot analysis, semi-quantitative or quantitative RT-PCR). Interesting gene expression profiles are revealed by determining transcript abundance for a selected transcription factor gene after exposure of plants to a range of different experimental conditions, and in a range of different tissue or organ types, or developmental stages. Experimental conditions to which plants are exposed for this puφose includes cold, heat, drought, osmotic challenge, and varied hormone concentrations (ABA, GA, auxin, cytokinin, salicylic acid, brassinosteroid). The tissue types and developmental stages include stem, root, flower, rosette leaves, cauline leaves, siliques, germinating seed, and meristematic tissue. The set of expression levels provides a pattern that is determined by the regulatory elements of the gene promoter. Transcription factor promoters for the genes disclosed herein are obtained by cloning 1.5 kb to 2.0 kb of genomic sequence immediately upstream of the translation start codon for the coding sequence of the encoded transcription factor protein. This region includes the 5 '-UTR of the transcription factor gene, which can comprise regulatory elements. The 1.5 kb to 2.0 kb region is cloned through PCR methods, using primers that include one in the 3' direction located at the translation start codon (including appropriate adaptor sequence), and one in the 5' direction located from 1.5 kb to 2.0 kb upstream of the translation start codon (including appropriate adaptor sequence). The desired fragments are PCR-amplified from Arabidopsis Col-0 genomic DNA using high-fidelity Taq DNA polymerase to minimize the incoφoration of point mutation(s). The cloning primers incoφorate two rare restriction sites, such as Notl and Sfil , found at low frequency throughout the Arabidopsis genome. Additional restriction sites are used in the instances where a Notl or Sfil restriction site is present within the promoter. The 1.5-2.0 kb fragment upstream from the translation start codon, including the 5 '-untranslated region of the transcription factor, is cloned in a binary transformation vector immediately upstream of a suitable reporter gene, or a transactivator gene that is capable of programming expression of a reporter gene in a second gene construct. Reporter genes used include green fluorescent protein (and related fluorescent protein color variants), β-glucuronidase, and luciferase. Suitable transactivator genes include LexA-GAL4, along with a transactivatable reporter in a second binary plasmid (as disclosed in US patent application 09/958,131 , incoφorated herein by reference). The binary plasmid(s) is transferred into Agrobacterium and the structure of the plasmid confirmed by PCR. These strains are introduced into Arabidopsis plants as described in other examples, and gene expression patterns determined according to standard methods know to one skilled in the art for monitoring GFP fluorescence, β-glucuronidase activity, or luminescence.

All publications and patent applications mentioned in this specification are herein incoφorated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incoφorated by reference. The present invention is not limited by the specific embodiments described herein. The invention now being fully described, it will be apparent to one of ordinary skill in the art that many changes and modifications can be made thereto without departing from the spirit or scope of the Claims. Modifications that become apparent from the foregoing description and accompanying figures fall within the scope of the following Claims.

Claims

What is claimed is:

1. A transgenic plant having greater abiotic stress tolerance than a control plant, wherein said transgenic plant comprises a recombinant polynucleotide that encodes a member of the G867 clade of transcription factor polypeptides; wherein the member of the G867 clade of transcription factor polypeptides comprises an AP2 domain, a DML motif, and a B3 domain.

2. The transgenic plant of Claim 1 , wherein the AP2 domain, the DML motif, and the B3 domain are at least 81 %, 71 %, and 78% identical to the AP2 domain, the DML motif, and the B3 domain of SEQ

ID NO: 2, respectively.

3. The transgenic plant of Claim 1 , wherein the abiotic stress is cold, heat, or an osmotic stress.

4. The transgenic plant of Claim 3, wherein the osmotic stress is selected from the group consisting of drought, desiccation, freezing, and high salt.

5. The transgenic plant of Claim 1, wherein the member of the G867 clade of transcription factor polypeptides comprises SEQ ID NO: 18.

6. The transgenic plant of Claim 1, wherein the recombinant polynucleotide comprises SEQ ID NO: 17.

7. The transgenic plant of Claim 1, wherein the member of the G867 clade of transcription factor polypeptides comprises SEQ ID NO: 55.

8. The transgenic plant of Claim 1, wherein the member of the G867 clade of transcription factor polypeptides is selected from the group consisting of SEQ ID NOs: 4, 6, 22, 24, 28, 32, 34, 36 and 38.

9. The transgenic plant of Claim 1 , wherein the recombinant polynucleotide is operably linked to at least one regulatory element capable of regulating expression of the recombinant polynucleotide when the recombinant polynucleotide is transformed into a plant.

10. The transgenic plant of Claim 9, wherein the regulatory element comprises an inducible or tissue-specific promoter.

11. The transgenic plant of Claim 10, wherein the inducible or tissue-specific promoter is selected from the group consisting of a SUC2 promoter, a CUTl promoter, an RBCS3 promoter, an ARSKl promoter, and an RD29A promoter.

12. Seed produced from the transgenic plant according to Claim 1.

13. A method for producing a transgenic plant having increased tolerance to an abiotic stress, the method steps comprising: (a) transforming a plant with an expression vector comprising a regulatory element operably linked to a polynucleotide sequence encoding an AP2 domain, a DML motif, and a B3 domain to produce a transgenic plant; and (b) identifying an abiotic stress tolerance trait of the transgenic plant.

14. The method of Claim 13, the method steps further comprising: (c) selftng or crossing the abiotic stress tolerant plant with itself or another plant, respectively, to produce seed.

15. The method of Claim 13, wherein said abiotic stress tolerance is cold, heat, or an osmotic stress.

16. The method of Claim 15, wherein said osmotic stress is selected from the group consisting of desiccation, drought, freezing, and high salt.

17. The method of Claim 13, wherein the AP2 domain, the DML motif, and the B3 domain are at least 81 %, 71 %, and 78% identical to the AP2 domain, the DML motif, and the B3 domain of SEQ ID

NO: 2, respectively.

18. The method of Claim 13, wherein the regulatory element is an inducible or tissue-specific promoter.

19. The method of Claim 18, wherein the inducible or tissue-specific promoter is selected from the group consisting of a CUTl promoter, a SUC2 promoter, an RBCS3 promoter, an ARSKl promoter, and an RD29A promoter.

20. The method of Claim 18, wherein the transformed plant has a moφhology that is substantially similar to a control plant.

21. The method of Claim 13, wherein the member of the G867 clade of transcription factor polypeptides comprises a polypeptide sequence selected from the group consisting of SEQ ID NO: 4, 6, 18, 22, 24, 28, 32, 34, 36 and 38.

22. Seed produced from the transformed plant produced by the method according to Claim 13.

23. A transgenic plant having greater abiotic stress tolerance than a control plant, wherein said transgenic plant has been transformed with an expression vector comprising a polynucleotide that encodes a polypeptide having an AP2 domain and a DML motif, from N-terminal to C-terminal, respectively, and wherein said expression vector does not encode a B3 domain.

24. The transgenic plant of Claim 23, wherein said AP2 domain and said DML motif are at least 81%, and 71% identical to the AP2 domain and the DML motif of SEQ ID NO: 2, respectively.

25. The transgenic plant of Claim 23, wherein said expression vector comprises SEQ ID NO: 56.

26. A transgenic plant having greater abiotic stress tolerance than a control plant, wherein said transgenic plant has been transformed with an expression vector comprising a polynucleotide that encodes a polypeptide having a B3 domain and wherein the polynucleotide does not encode an AP2 domain.

27. The transgenic plant of Claim 26, wherein said B3 domain is at least 78% identical to the B3 domain of SEQ ID NO: 2.

28. The transgenic plant of Claim 26, wherein said expression vector comprises SEQ ID NO: 57.

29. A transgenic plant having greater tolerance to an abiotic stress than a control plant, wherein said transgenic plant has been transformed with an expression vector comprising a polynucleotide that encodes a GAL4 activation domain and a member of the G867 clade of transcription factor polypeptides, in order from N-terminal to C-terminal, respectively.

30. The transgenic plant of Claim 29, wherein the expression vector comprises SEQ ID NO: 58.

31. A transgenic plant having greater tolerance to an abiotic stress than a control plant, wherein said transgenic plant has been transformed with an expression vector comprising a polynucleotide that encodes a member of the G867 clade of transcription factor polypeptides and a GAL4 activation domain, in order from N-terminal to C-terminal, respectively.

32. The transgenic plant of Claim 31, wherein the expression vector comprises SEQ ID NO: 59.

33. The transgenic plant of Claim 29 or Claim 31, wherein the transformed plant has a moφhology that is substantially similar to the control plant.

34. The transgenic plant of Claim 29 or Claim 31, wherein the abiotic stress is cold, heat, or an osmotic sfress.

35. The transgenic plant of Claim 34, wherein the osmotic stress is selected from the group consisting of drought, desiccation, freezing, and high salt.

36. A method for producing a transgenic plant having increased tolerance to an abiotic stress, the method steps comprising: (a) identifying in a polypeptide database a polypeptide sequence comprising a DML motif having at least 71% sequence identity to SEQ ID NO: 54; (b) providing a polynucleotide sequence that encodes the polypeptide sequence; (c) incoφorating the polynucleotide sequence into an expression vector, wherein the expression vector also comprises a regulatory element operably linked to the polynucleotide sequence; (d) transforming a plant with the expression vector to produce a transgenic plant; and (e) screening the transgenic plant for greater tolerance to abiotic stress than a control plant.

37. The method of Claim 36, wherein the abiotic stress is cold, heat, or an osmotic stress.

38. The method of Claim 37, wherein the osmotic stress is selected from the group consisting of drought, desiccation, freezing, and high salt.

39. The method of Claim 36, wherein the polypeptide sequence is selected from the group consisting of SEQ ID NOs: 4, 6, 8, 18, 22, 24, 28, 32, 34, 36, 38, 40, 42, 44, 46, 48, 50, 52, and 53.

40. A transgenic plant produced by the method of Claim 36.