AU2007201633A1 - Promotion of somatic embryogenesis in plants by wuschel gene expression - Google Patents

Description

0 0 c m,

C-^

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT Applicant: THE ROCKEFELLER UNIVERSITY Invention Title: PROMOTION OF SOMATIC EMBRYOGENESIS IN PLANTS BY WUSCHEL GENE EXPRESSION The following statement is a full description of this invention, including the best method of performing it known to us:

-IA-

0 0 TITLE OF THE INVENTION PROMOTION OF SOMATIC EMBRYOGENESIS IN PLANTS BY WUSCHEL GENE EXPRESSION SBACKGROUND OF THE INVENTION Somatic embryogenesis is a unique pathway for asexual propagation or somatic C cloning in plants. The developmental process of somatic embryogenesis shares Cc€ O considerable similarity with that of zygotic embryogenesis (Zimmerman, 1993; O Mordhorst et al., 1997) and this is likely due to the conservation in the underpinning 17 cellular and molecular mechanisms between the two processes. Therefore, somatic O 10 embryogenesis provides an attractive model system for studying zygotic embryogenesis, particularly because zygotic embryos are encased by maternal tissues and difficult to access by biochemical and molecular tools. Moreover, in biotechnological applications, most economically important crop as well as non-crop plants are regenerated via somatic embryogenesis.

In contrast to organogenesis, which requires a high cytokinin to auxin ratio (Skoog and Miller, 1957; Sugiyama, 1999; Sugiyama, 2000), somatic embryogenesis does not require any external cytokinins, but rather is dependent on high concentrations of 2,4-D (Zimmerman, 1993; Mordhorst et al., 1997; Sugiyama, 2000), a synthetic chemical that has long been used as a functional analog of auxin. It is generally believed that somatic embryogenesis is mediated by a signaling cascade triggered by external auxin or 2,4-D (Zimmerman, 1993; Mordhorst et al., 1997; Schmidt et al., 1997). However, very little is known about the signal transduction pathway, particularly the molecular mechanism involved in the transition of a vegetative cell to an embryogenic competent cell.

During the last two decades, considerable efforts have been made to identify genes with altered expression patterns during somatic embryogenesis (Schmidt et al., 1997; Lin et al., 1996; Thomas, 1993). Most of these genes, however, are up-regulated only in late developmental stages, suggesting that they do not play a direct role in die vegetative-toembryogenic transition. Thus far, the only exception is the carrot Somatic Embiyogenesis Receptor-like Kinase (SERK) gene the expression of which appears to mark the vegetative-to-embryogenic transition; however, its function remains unclear (Schmidt et al., 1997). An additional molecular approach was attempted by manipulating certain 0 O 2 Ci1 embryo-specific genes. The Arabidopsis Leafy cotyledon 1 (LEC1) gene, encoding a Csubunit of the HAP heterotrimeric transcription factor complex (HAP3), has been Sproposed as a key regulator for embryonic identity (Lotan et al., 1998). Mutations in the e¢n SLECI locus result in defective embryo maturation as well as the conversion of cotyledons into true-leaf-like structures (Lotan et al., 1998; Meinke, 1992; Meinke et al., 1994).

en Constitutive overexpression of LECI leads to severely abnormal plant growth and IND development with occasional formation of somatic embryo-like structures (Lotan et al., O 1998). The developmental fate of these embryo-like structures, however, remained Sunknown due to the lethality of LEC1 overexpression.

The publications and other materials used herein to illuminate the background of the invention, and in particular, cases to provide additional details respecting the practice, are incorporated herein by reference, and for convenience, are referenced by author and date in the text and respectively grouped in the appended List of References.

SUMMARY OF THE INVENTION One aspect of the present invention is a method to promote somatic embryogenesis from a tissue or organ of a plant, said method comprising overexpressing a Wuschel gene in said tissue or organ.

A second aspect of the invention is a method to generate somatic plant embryos wherein said method comprises overexpressing a Wuschel gene in a tissue or organ of a plant.

Another aspect of the invention is a method for generating shoots from a tissue or organ of a plant, said method comprising overexpressing a Wuschel gene in said tissue or organ.

Yet another aspect of the invention is a method of selecting plants transformed with a vector comprising a silent selectable marker wherein the marker is a Wuschel gene.

Another object of the invention is a method of producing an apomictic plant line.

Another object of the invention is a method of producing haploid plants.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram of the XVE activation tagging vector pER16.

Figures 2A-F illustrate the pga6 gain-of-function mutant phenotype. Root explants derived from pga6 seedlings were cultured on the non-inductive SCM (SCM minus 17-P-estradiol) for 20 days (Figure 2A); or on the inductive SCM for 10 days (Figure 2B), 20 days (Figure 2C), or 30 days (Figure 2D). Figure 2E shows an enlarged view of a germinating somatic embryo isolated from the explant shown in (Figure 2D).

Figure 2F shows a germinating seedling derived from a somatic embryo grown on MS medium (45 days). Scale bar, 100 gmn for Figures 2A and 2E; 1 mm, for Figures 2B, 2C, 2D and 2F.

Figures 3A-D are electron microscopic analyses showing somatic embryogenesis in pga6 mutant explants (culturing conditions were identical to those shown in Figures 2A-F). Figure 3A shows a pre-embryo stage before the first embryonic cell division (arrows) and a two-cell stage after the first asymmetric division with a smaller apical cell and a larger basal cell. Figure 3B shows embryos at the globular and the early heart stages. Figure 3C shows a germinating embryo. C: cotyledon; H: hypocotyl. Figure 3D shows an abnormal somatic embryo with three cotyledons (C) anchored on the hypocotyl Scale bar, 10 pmr for Figure 3A; 100 pm for Figures 3B-

D.

Figures 4A-H illustrate phytohomione-independent somatic embryo formation caused by the pga6 gain-of-function mutation. Figure 4A is an overview of pga6 mutant seedlings germinated and grown on MS medium (first seedling from die left) or the inductive MS medium (MST: 5 pM 17-p-estradiol) for 7 days. Figures 4B-D show pga6 seedlings that were cultured on the inductive MS medium for 10 days (Figure 4B), 14 days (Figure 4C) or 30 days (Figure 4D). Figures 4E-F show seven-day-old pga6 seedlings germinated and grown on MS medium which were transferred onto an inductive MS medium and cultured for 5 (Figure 4E) or 10 (Figure 4F) days. Figure 4G shows pga6 root explants which were cultured on the inductive MS medium for 20 days. Figure 4H is an enlarged view of Figure 4G. Scale bar, 1 mm.

1- 84 C l Figures 5A-D show that the pga6 phenotype is due to the inducer-dependent Ooverexpression of WUS. Figure 5A is a schematic diagram illustrating the insertion site of the T-DNA upstream of the WUS gene (not shown to scale). Arrows indicate the Sdirections of transcription. Figure 5B shows pga6 seeds (T2, homozygous) which were germinated and grown on MS medium supplemented with various concentrations of the en inducer as indicated. Ten-day old seedlings are shown. The scale bar represents 1 mm.

SFigure 5Cshows the expression of PGA6/WUS induced by different concentrations of the 0 inducer. Ten-day-old pga6 seedlings were germinated and grown on MS medium and o transferred to an MS medium containing various concentrations of 17-j3-estradiol as indicated and were cultured for an additional 16 hours before total RNA extraction. Five lig total RNA were used for Northern blot analysis using a WUS cDNA fragment as a probe. Positions of two RNA molecular weight markers are indicated at the right (GIBCO BRL, catalog number, 15620-016). Figure 5D shows ethidium bromide staining of the gel as a control for RNA loading and transfer.

Figures 6A-H are photographs showing that 35S- or XVE-controlled overexpression of WUS cDNA phenocopies the pga6 phenotype. Figure 6A shows embryogenic callus and Figure 6B shows somatic embryo formation from root tips of XVE-WUS cDNA T2 seedlings grown for 15 days in A medium supplemented with 17-3estradiol (10 pM). Figures 6C-H show 15 day-old TI 35S::WUS seedling phenotypes.

Figure 6C shows die tips of the roots are enlarged and show an embryo-like structure.

Figure 6D shows the adventitious root tip. Figure 6E shows that WUS overexpression induces both organogenesis and embryogenesis from the root. Figure 6F shows detail of early embryo structure formation. Figure 6G shows the shoot apical meristem is dramatically altered and, besides forming lateral organs with altered shaped, givers rise to adventitious shoots and somatic embryos. Figure 6H- shows the entire shoot apical meristem expands and lateral organs transform into meristematic tissues. Scale bar is 1 mm.

Figures 7A-C are Northern blots of RNA from root explants prepared from pga6 seedlings cultured on the screening medium (SCM) for different times as indicated. On day 28, when somatic embryos were apparent, cultures were transferred onto a freshlyprepared SCM or control medium (SCM without the inducer) and incubated for an additional day (28 1) or two days (28 Five micrograms of total RNA, prepared from the frozen materials, were analyzed by Northern blotting using WUS (Figure 7A) and LEC1 (Figure 7B) cDNA as probes. The blot was rehybridized with an actin cDNA probe (Figure 7C) to ensure that equal amounts of RNA were loaded.

Figures 8A-B illustrate formation of somatic embryos from isolated zygotic embryos ofPGA6 transgenic plants grown in the presence (Figure 8A) or absence (Figure 8B) of an inducer of PGA6.

DETAILED DESCRIPTION OF THE INVENTION To dissect the signaling pathway during somatic embryogenesis, we have employed a genetic approach to identify gain-of-ftmction mutations that can promote embryogenic callus fonnation from Arabidopsis root explants. Arabidopsis thaliana is known to be a species difficult for somatic embryogenesis. Thus far, embryogenic calli could only be induced from immature embryos of wild-type (WT) plants or from the primordia timing (pt) mutant plant (Wu et al., 1992; Mordhorst et al., 1998; and references therein). Therefore, Arabidopsis vegetative explants appear to be reliable materials for screening for genetic mutations involved in the vegetative-to-embryonic transition. Herein we disclose the identification of the Plant Growth Activator 6 (PGA6) gene by a novel genetic screen. Overexpression of PGA6 promotes the formation of somatic embryos from various vegetative tissues as well as from zygotic embryos independently of any external plant hormones. These somatic embryos, following a developmental process remarkably similar to that of zygotic embryogenesis, are able to germinate and grow into healthy, fertile plants, suggesting that PGA6 is involved in the maintenance of embryonic cell identity. PGA6 was found to be identical to the Wuschel gene (WUS), a homeodomain protein that was previously characterized as a key regulator for specification of meristem cell fate (Laux et al., 1996; see, also, WO 01/23572 regarding WUS homologs). The nucleic acid and protein sequences of Wuschel are those as shown by GenBank Accession No. AJ012310. Our results reveal an additional 0 S6 Cl function of WUS/PGA6 during embryogenesis, and also open a new avenue in Sbiotechnological applications.

In this study, we identified a genetic locus PGA6 by a novel functional screen e¢n aimed at elucidating the molecular mechanism of somatic embryogenesis. A gain-offunction mutation at this locus causes a rapid transition from vegetative or somatic to ^e embryonic cells, leading to somatic embryo development from various tissues and organs.

The pga6 mutation-dependent cell fate reprogramming can occur either in the presence or 'l absence of external plant hormones, although the local concentration of endogenous o growth regulators might play an important role in the vegetative-to-embryonic transition.

10 In addition, the hormone-independent somatic embryogenesis in pga6 strikingly resembles zygotic embryo development. These observations suggest that PGA6 plays a critical regulatory role during embryogenesis, likely involved in maintaining embryonic cell identity. Molecular and genetic analyses indicate that pga6 is a gain-of-function allele of the previously characterized wus loss-of-function mutation (Mayer et al., 1998).

In addition to causing vegetative tissues or organs to become embryogenic, inducing overexpression of PGA6 in zygotic embryos also results in the generation of somatic embryos at very high frequency, whereas no somatic embryos or embryogenic calli were observed in the absence of the inducer. Wu et al. (1992) previously reported the generation of somatic embryos from isolated Arabidopsis zygotic embryos at very low efficiency, which involved tedious multiple subculturing and hormone treatments. Our finding that the simple manipulation of a single gene (WUS) was able to generate somatic embryos at very high frequency is a major advance in plant biotechnology.

Loss-of-fuction mutations in WUS have been shown to cause impaired development of shoot and floral meristems in Arabidopsis, resulting in the absence of die shoot and floral meristems in all developmental stages of wus embryos and adult plants (Laux et al., 1996; Mayer et al., 1998). Genetic studies revealed that WUS interacts with CLA VA TA (CL and the WUSICL V self-regulatory loop, in which CLV presumably acts upstream of WUS (Clark, 2001) appears to be critical for the maintenance of stem cell identity (Schoofet al., 2000; Brand et al., 2000; Waites and Simon, 2000). On the other hand, ectopic expression of WUS results in enlarged meristems (Schoof et al., 2000).

Collectively, these observations suggested an instructive role of WUS for the

O

o 7 C specification of meristematic stem cell fate. Interestingly, the WUS gene is not expressed in dte stem or meristem cells; rather its expression is restricted to a small group of cells <1 underneath the stem cells during die entire embryogenesis and post-embryogenesis stages (Mayer et al., 1998; Schoof et al., 2000). The cluster of WUS expressing cells was termed as the organizing center (Mayer et al., 1998). The unexpected expression pattern ^f led to postulations that WUS promotes and/or maintains the stem cell fate by a diffusion IND mechanism or acted in a non-cell-autonomous manner (Mayer et al., 1998; Waites and O Simon, 2000).

Our observation that WUS is capable of promoting vegetative-to-embryonic O 10 transition and eventually somatic embryo formation uncovers an additional critical function of this homeodomain protein during embryogenesis. Presumably, the highly restrictive expression of WUS hallmarks the putative embryonic organizing center which, in turn, may give rise to stem cells during embryogenesis and later development.

Therefore, WUS is involved in promoting and maintaining the identity of embryonic cells from which stem cells are derived. Because WUS-expressing cells have not been morphologically and finctionally characterized, it remains of interest to determine whether this cluster of cells indeed represents a functional organizing center similar to Spemann's organizer discovered nearly 80 years ago in Xenopus embryos (Spemann and Mangold, 1924).

Interestingly, the LEC1 transcript was barely detectable in the organizing center or the WUS expressing domain, albeit LEC1 was found to express throughout embryogenesis as well as in seeds (Lotan et al., 1998). Consistent with these observations, somatic embryo expression of LEC1, presumably resembling that in zygotic embryos, was found to be promptly repressed by the WUS activity. In addition, the LEC1 function appeared to require unidentified embryo- and or seed-specific cofactors, since inducible overexpression of LEC1 by the XVE system (Zuo, et al., 2000b) during postgermination stages did not result in any detectable phenotype. By contrast, WUS appears to be a key player in promoting embryonic potential as its activity does not appear to require any developmentally specific factors under our tested conditions. Taken together, these observations further suggest that WUS plays a predominant role in inducing the embryonic potential, whereas LEC1 is likely involved in promoting differentiation of 0 8 embryonic cells at later developmental stages. A reasonable assumption would be that the SLECI activity, a driving force for embryo cell differentiation, must be excluded in order to fully maintain the embryonic potential in the putative organizing center.

Recently, Stone et al. (2001) showed that LEC2 encodes a transcription factor containing a B3 domain unique to several other plant transcription factors including e ABI3/VP1 and FUS3. Overexpression of LEC2 leads to formation of somatic embryos as well as the formation of callus, cotyledon-like and leaf-like structures, a phenotype O similar to that ofpga6 mutant, suggesting that LEC2 might be functionally close to WUS.

It will be interesting to determine if the LEC2 activity is also dependent on embryo- O 10 and/or seed-specific cofactors as in the case for LEC 1.

Systems for hormone-dependent somatic embryo formation have been well established in several model species, and rapidly extended to other species (Zimmerman, 1993; Mordhorst et al., 1997). In all these in vitro systems, phytohormones, particularly auxin or 2,4-D, are essential for induction of somatic embryo formation. Arabidopsis has been known to be a species difficult for somatic embryogenesis, even though with limited success by using wild-type immature zygotic embryos Wu et al., 1992) or certain vegetative tissues of the pt mutant plants (Mordhorst et al., 1998). Overexpression of Leafy cotyledon I (Lecl) causes severe developmental abnormality and growth arrest, a phenotype similar to that of the pga6 mutant (Lotan et al., 1998). Formation of somatic embryos is occasionally observed in the Lecl overexpression plants (Lotan et al., 1998), but these embryos never germinate or develop into normal adult plants. The finding that the pga6 gain-of-function mutation or overexpression of WUS results in hormoneindependent somatic embryo formation at a high frequency will have significant impact on plant biotechnology, and provides a convenient and attractive model system for many aspects of plant biological research.

In another embodiment of the invention, embryogenesis is induced in haploid cells, such as pollen cells, to produce haploid plants. This can be achieved by stably transfonning a plant cell or tissue with a WUS gene under the control of a tissue specific promoter that is active in a haploid cell or tissue, and expressing the WUS gene therein, or by introducing the WUS gene into a plant tissue or cell under the control of an inducible promoter and applying the inducer to cause expression of the WUS gene therein. In a 9 l preferred embodiment, the WUS gene is under the control of a promoter that is both haploid-tissue specific and inducible. In a preferred embodiment, a promoter is used that is both inducible and tissue-specific, giving greater control over the process. In a most preferred embodiment, a WUS gene linked to an inducible pollen-specific promoter is used to induce somatic embryogenesis in pollen cells.

c Expression of the gene in the haploid tissue or cell (for example, by application of N the inducer specific for the inducible promoter) results in the formation of haploid Ssomatic embryos, which can be grown into haploid plants using standard techniques.

o When an inducible promoter is used (whether tissue specific or not), a preferred method o 10 comprises exposing excised transgenic tissue containing the haploid cells pollen or ovules) to the inducer specific for the inducible promoter for a time sufficient to induce the formation of a somatic embryo, withdrawing the inducer, and growing the somatic embryo into a transgenic haploid plant in the absence of tie inducer.

Diploidization of the haploid plants to form dihaploids, either spontaneously or by treatment with the appropriate chemical colchicine) will significantly expedite the process of obtaining homozygous plants as compared to a method of conventional genetic segregation. This technology will not only be beneficial for breeding purposes but also for basic research such as studies of mutagenesis and other genetic studies, because dihaploids are truly homozygous down to the DNA level, containing two identical copies of each gene.

Additionally, WUS genes will be useful for inducing apomixis into plants.

Apomixis and methods of conferring apomixis into plants are discussed in several patents (see, U.S. Patent Nos. 5,710,367; 5,811,636; 6,028,185; 6,229,064; and 6,239,327 as well as WO 00/24914 which are incorporated herein by reference). Reproduction in plants is ordinarily classified as sexual or asexual. The term apomixis is generally accepted as the replacement of sexual reproduction by various forms of asexual reproduction (Rieger et al., IN Glossary of Genetics and Cytogenetics, Springer-Verlag, New York, 1976). In general the initiation of cell proliferation in the embryo and endosperm are uncoupled from fertilization. Apomixis is a genetically controlled method of reproduction in plants where the embryo is formed without union of an egg and a sperm. There are three basic types of apomictic reproduction: 1) apospory-embryo 0 O cI develops from a chromosomally unreduced egg in an embryo sac derived from a somatic Scell in the nucellus, 2) diplospory-embryo develops from an unreduced egg in an embryo sac derived from the megaspore mother cell, and 3) adventitious embryony-embryo develops directly from a somatic cell. In most forms of apomixis, pseudogamy or fertilization of the polar nuclei to produce endosperm is necessary for seed viability.

M These types of apomixis have economic potential because they can cause any genotype, I regardless of how heterozygous, to breed true. It is a reproductive process that bypasses O female meiosis and syngamy to produce embryos genetically identical to the maternal parent. With apomictic reproduction, progeny of specially adaptive or hybrid genotypes would maintain their genetic fidelity throughout repeated life cycles. In addition to fixing hybrid vigor, apomixis can make possible commercial hybrid production in crops where efficient male sterility or fertility restoration systems for producing hybrids are not known or developed. Apomixis can make hybrid development more efficient. It also simplifies hybrid production and increases genetic diversity in plant species with good male sterility.

It would be ideal to find genes controlling obligate or a high level of apomixis in the cultivated species and be able to readily hybridize cross-compatible sexual x apomictic genotypes to produce true-breeding F, hybrids. In reality, most desirable genes controlling apomixis are found in the wild species which are distantly related to the cultivated species. Although interspecific crosses may be possible between the cultivated and wild species, chromosome pairing between genomes is usually low or nonexistent.

Although apomixis is effectively used in Citrus to produce uniform and diseaseand virus-free rootstock (Parlevliet et al, 1959) and in buffelgrass (Bashaw, 1980) and Poa (Pepin et al, 1971) to produce improved cultivars, it has not been successfully transferred to a cultivated crop plant. The transfer of apomixis to important crops would make possible development of true-breeding hybrids and commercial production of hybrids without a need for cytoplasmic-nuclear male sterility and high cost, labor-intensive production processes. An obligately apomictic F, hybrid would breed true through the seed indefinitely and could be considered a vegetative or clonal method of reproduction through the seed. The development of apomictically reproducing cultivated crops would also provide a major contribution toward the food security in developing nations (Wilson et al, 1992).

O

o 11 (N The present invention is further detailed in the following Examples, which are offered by way of illustration and are not intended to limit the invention in any manner.

<1 Standard techniques well known in the art or the techniques specifically described below are utilized.

SExample 1 N Plant materials, growth conditions and plant transformation SThe Wassilewskija, Columbia and Landsberg ecotypes of A. thaliana were used.

Plants were grown under continuous white light at 22 0 C on solid A medium (1 X MS salts O 10 (Murashige and Skoog, 1962), 3% sucrose, 0.8% agar) supplemented with appropriate antibiotics and/or the inducer 17-0-estradiol. Unless indicated otherwise, 5 IM 17-pestradiol was used for induction. In planta transformation of Arabidopsis plants (the Columbia ecotype) was performed as described by Bechtold et al. (1993).

Transformation of root explants was carried out according to Koncz et al. (1989).

Light and electron microscope analyses were carried out as previously described (Zuo et al., 2000b).

Example 2 Screening ofpga mutants Agrobacteria ABI cells carrying pERI 6 were used to transform Arabidopsis (the Wassilewskija ecotype) root explants. Infected root explants were cultured on the screening medium (SCM; 1 X MS salts, 1% sucrose, 0.5 g/L MES morpholino]ethanesulfonic acid), 0.15 mg/L IAA (indole acetic acid), 5 p.M 17-0estradiol, 50 mg/L kanamycin, 100 mg/L carbenicillin and 0.2% phytagel, pH 5.7) at 22°C under a 16-hour white light/8-hour dark cycle. Putative pga mutants, which appeared as rapidly growing green-yellowish or green cell clumps or calli upon culturing on the SCM for 10-15 days, were transferred onto a non-inductive shoot induction medium (SIM; for green calli; 1 X MS salts, 1% sucrose, 0.5 g/L MES, 1 mg/L 2-IP (N 6 52isopentenyladenine), 0.15 mg/L IAA, 50 mg/L kanamycin, 100 mg/L carbenicillin and 0.2% phytagel, pH 5.7) to recover mutant shoots. The green-yellowish calli were transferred onto the callus induction medium (CIM; 1 X B5 salts (Sigma), 2% glucose,

O

o 12 N 0.5 g/L MES, 0.5 mg/L 2,4-D, 0.05 mg/L kinetin, 50 mg/L kanamycin, 100 mg/L carbenicillin and 0.2% phytagel, pH After culturing on the CIM for 7-10 days, the amplified calli were then transferred onto a SIM to regenerate shoots. Regenerated Sshoots, usually formed after culturing on the SIM for 2-3 weeks, were then transferred to a root induction medium (RIM; identical to SIM but without 2-IP) to promote root en formation. Whereas most putative pga mutant plantlets, including pga6, were able to set IN seeds after transferring to soil, a portion of roots were excised and placed on the CIM to O reinduce callus formation, followed by repeating the above-described screening procedure to confirm the inducer-dependent pga phenotype. The pga6 mutant was backcrossed with O 10 wild-type (Wassilewskija) plants twice for further genetic and phenotypic analyses.

Example 3 Molecular manipulations Molecular manipulations were performed as specifically stated or by the methods as taught by Sambrook et al. (1989). The XVE inducible expression vector pERIO is identical to pER8 (Zuo et al., 2000a) except that the hygromycin selectable marker of pER8 was replaced with a kanamycin selectable marker. To construct the mutagenesis vector pER16, pERlO was digested with Spel and Asp7181 followed by Klenow enzyme fill-in reaction and religation. The resulting pER16 vector lacked the rbcsS3A polyA addition sequence of the OLexA-46.:-T3A expression cassette (see Figure 1 of Zuo et al.

(2000a)).

pER16 is shown in Figure 1. Only the region between die Right Border (RB) and Left Border (LB) is shown (not to scale). Two transcription units and the OLexA- 4 6 promoter are located between the RB and LB. In the first transcription unit, the GO0-90 promoter (Ishige et al., 1999) drives the XVE fusion gene terminated by the rbcs E9 polyA addition sequence. The second transcription unit consists of the Nopaline Synthase (NOS) gene promoter, the coding sequence of the Neomycin Phosphotransferase II (NPT If) gene and the NOS polyadenylation sequence. The OLexA-46 promoter consists of 8 copies of the LexA operator sequence fused to the -46 CaMV35S promoter. Upon integration into the plant genome, the oLexA- 4 6 promoter can activate the transcription of sequences fused downstream from the promoter in a 17-p-estradiol-dependent fashion.

0 13 C1 The pER16-tagged genomic sequences were recovered by Tail-PCR (Liu et al., 1995), and the purified PCR fragments were directly subjected to DNA sequencing analysis.

The WUS cDNA was amplified from flower cDNA by polymerase chain reaction (PCR), using the primers WusUp CTTATTTACCGTTAACTTGTGAACA

(SEQ

en ID NO: 1) and WusLow CACATAACGAGAGATAACTAGTTAAC (SEQ ID SNO:2). The 1062-bp long PCR fragment, harboring the entire protein coding region plus O part of the 3' UTR, was first cloned into the pPCR-Script Amp vector (Stratagene) and then cut with XhoI (vector polylinker site) and SpeI (site in the 3' UTR of the WUS O 10 cDNA) for the subsequent cloning into the corresponding sites of both the pER10 vector (17-3-estradiol inducible) and the pBA002 vector (constitutive 35S promoter) for the WUS expression. The correct sequence of the WUS cDNA was confirmed by DNA sequencing analysis.

Genomic DNA Southern and RNA Northern blotting analyses were carried out as previously described (Zuo et al., 2000a; Zuo et al., 2001).

Example 4 Screening of the plant growth activator mutants Explants derived from Arabidopsis vegetative tissues are known to be incapable of forming somatic embryos or embryogenic calli promoted by external plant hormones.

We presumed that external hormones alone were incapable of activating key regulators of Arabidopsis necessary for vegetative-to-embryogenic transition. With appropriate hormone treatments, gain-of-function mutations in these regulatory genes may activate the vegetative-to-embryonic transition. Such gain-of-function mutations, on the other hand, may also cause severe defects during subsequent plant growth and development.

Therefore, if the expression of the mutated gene and/or the biological activity of related gene products is not appropriately controlled, it will be difficult to maintain the identified mutations. An example is the constitutive overexpression of the LECI gene (Lotan et al., 1998). As a consequence, we carried out the screen by using a previously developed chemical-inducible XVE system, which has been demonstrated to be stringently controlled and to be highly responsive to the inducer 17-p-estradiol, a mammalian S14 Cl hormone with no apparent physiological effects on plant growmn ana aevelopment (Luo et al., 2000a). The use of an inducible promoter thus allows us to recover normal mutant plants by withdrawal of the inducer, even in the case that the gain-of-function mutation is lethal.

Arabidopsis root explants were transformed with Agrobacteria (Koncz et al., en3 r 1989) carrying an XVE vector pER16 (Figure Transformed explants were cultured on Sa screening medium (SCM containing auxin, 17-P-estradiol and kanamycin but without 0 cytokinin). Note that mutations functionally analogous to cytokinin independent 1 (ckil; SKakimoto, 1996) should also be recovered under the screening conditions. In the primary 0 10 screen, we isolated 35 putative mutants by interrogating approximately 38,000 independent transformation events. As expected, most of these mutants (33 out of showed a ckil-like phenotype, shoot regeneration independent of cytokinin. The remaining two mutants gave rise to green-yellowish embryogenic calli. We collectively named these two classes of mutations as pga for plant growth activator. Here, we disclose a detailed characterization of one of these mutants which is named pga6.

Example The pga 6 gain-of-finction mutation promotes the vegetative-to-embryonic transition The pga6 mutant was initially identified by its ability to form embryogenic calli on SCM. The embryogenic calli were transferred onto a shoot induction medium containing both auxin and cytokinin but without the chemical inducer 17-0-estradiol.

After 2-3 weeks, shoots were regenerated from the isolated calli. Explants derived from different organs of the regenerated shoots were cultured on SCM as described before; a portion of the excised explants was cultured in SCM without the inducer to serve as controls. After culturing for 7-10 days, only slowly growing calli were occasionally observed in the absence of inducer (Figure 2A). In the presence of the inducer, however, the explants produced numerous, rapidly growing, yellowish embryogenic calli (Figure 2B), which subsequently developed into distinctive somatic embryos (Figure 2C). After being transferred onto a non-inductive medium, all these somatic embryos were able to germinate (Figures 2D and 2E) and develop into fertile adult plants, most of which were morphologically indistinguishable from wild-type plants (Figure 2F).

O Cl To further confirm the above observations.

e somatic embryo-derived plants, as shown in Figure 2F, were germinated on MS medium in the absence of the inducer. The inducer-dependent somatic embryo formation was Sreproducibly observed in pga6 explants prepared from different organs/tissues of previously uninduced TI plants. Similar to that observed in the TO explants, the highest en frequency of somatic embryo formation was observed from root explants, followed by N leaf petioles, stems and leaves. Isolated zygotic embryos had a frequency similar to that O of root explants. The above results suggested that thepga6 gain-of-function mutation was Sable to promote vegetative-to-embryonic transition under our experimental conditions, 0 10 and that the PGA6 gene was most likely tagged by the inducer-responsive OLexA- 4 6 promoter.

In addition to the formation of somatic embryos, which are characterized by the presence of cotyledons lacking trichomes on the surfaces, we also observed approximately 10% of the pga6 calli developing into shoots, suggesting that PGA6 is also involved in organogenesis.

Example 6 The development of pga6 somatic morphologically and temporally resembles that of zygotic embryos To more closely follow the somatic embryogenesis process of the pga6 mutant, we performed a scanning electron microscopic analysis. Somatic embryogenesis in the inducer-treated pga6 explants highly resembled zygotic embryogenesis. The process was initiated from an asymmetric division of a single cell, giving rise to a smaller apical cell and a larger basal cell (Figure 3A). Subsequently, embryo-like structures equivalent to that of zygotic embryos at the early globular stage were easily recognizable (Figure 3B).

These somatic embryos underwent a typical embryogenesis process, including the heart stage, the torpedo stage, and the cotyledon stage (Figures 3B and 3C), and eventually germinated into healthy seedlings (Figure 2F). In addition to the relatively normal growth and development of the somatic embryo-derived mutant plants, we observed that a small fraction (approximately 10%) of somatic embryos generated seedlings with three cotyledons (Figure 3D). This abnormality is presumably caused by different expression o 16 fCfK levels of the PGA6 gene or by abiotic effects as occasionally observed in seedlings p germinated from wild-type seeds. Nevertheless, these above results demonstrated that a <1 conditional gain-of- function mutation in the pga6 locus promoted vegetative-toen3 embryonic transition, leading to the initiation of somatic embryogenesis.

Ce Example 7 Plant hormone-independent somatic embryo formation o The data described above were obtained under tissue culture conditions. To F; investigate the effects of the pga6 mutation on normal plant growth and development, T1 O 10 seeds were germinated on MS media (Murashige and Skoog, 1962) with or without the inducer. No detectable abnormality in growth and development (phenotype) was observed in pga6 plants germinated and grown in the absence of tie inducer. On the inductive MS medium, progeny with the mutant and wild-type phenotype segregated in die ratio of 3:1, characteristic of a dominant mutation in a single locus. Compared to wild-type seeds, thepga6 mutant seeds germinated substantially later, with a delay of days, suggesting an inhibitory effect of the mutation on plant growth and development.

Whereas approximately one third of the mutant seedlings stopped further growth after germination and eventually died, dithe remaining two thirds mutant seedlings rapidly turned into green calli upon germination. These two distinctive mutant phenotypes, with an approximately 2:1 ratio, were presumably due to segregations (heterozygous or homozygous) for the pga6 gain-of-function mutation locus, leading to different PGA6 expression levels in homozygous and heterozygous seedlings. This notion was supported by subsequent genetic analysis of the T2 generation derived from T1 plants grown under non-inductive conditions. Whereas all progenies of the putative homozygous T2 families (5 out of 16) showed only the "lethal" phenotype on the inductive medium, the remaining 11 heterozygous families gave rise to a typical 2:1:1 segregation for embiyogenic calli, "lethal" and wild-type phenotypes. In addition, the strength of the mutant phenotype was dependent on the inducer concentrations and the induced PGA6 expression levels per se (see Figures In the absence of any external plant hormones, somatic embryos were formed from green embryogenic calli (Figures 4B and 4C), which were able to germinate and

O

o 17 grow normally into fertile adult plants (Figure 4D). Interestingly, a considerably higher Sfrequency of embryogenic callus formation was observed from excised pga6 roots cultured on the inductive MS medium (Figures 4G and 4H), suggesting the presence of an antagonistic activity to PGA6 in intact plants, which may promote post-embryogenesis growth and development.

en Overexpression of LEC1 leads to abnormal plant growth and development as well as the occasional formation of embryo-like structures (Lotan et al., 1998). The LECI 0g gain-of-function phenotype, however, appeared to be strictly restricted to developmental o stages prior to germination. After seedling germination, overexpression of LEC1, controlled by the XVE inducible expression system (Zuo et al., 2000a), did not produce any apparent abnormality in plant growth and development, although the LEC1 transgene was highly responsive to the inducer during post-germination stages. On the other hand, the same transgenic line showed a strong phenotype if germinated directly in the presence of the inducer and the LEC1 transgene was highly responsive to the inducer during postgermination stages. These observations suggest that embryo- or seed-specific co-factor(s) are required for LEC1 function. To examine if PGA6 function is also dependent on embryo- or seed-specific co-factors, pga6 mutant seedlings at different growth stages, germinated and grown on the non-inductive MS medium, were transferred onto an inductive MS medium. At these developmental stages, the fate of both root and shoot stem cells has already been highly specified in wild-type plants. The pga6 mutation, however, appeared to reverse the developmental program, causing both root and shoot meristems to transform into embryogenic calli (Figure 4D). Similar to that shown before, these embryogenic calli were eventually capable of forming somatic embryos, which were able to germinate and grow into morphologically normal adult plants. The above results strongly suggest that PGA6 plays a key role in specifying and maintaining embryonic cell identity, independent of any embryo- or seed-specific co-factors.

Example 8 PGA6 is identical to the homeodomain protein WUS Based on the segregation of the kanamycin selectable marker, the pga6 mutant genome appears to contain a single transgenic locus. However, molecular analysis

O

O 18 C indicated the presence of the OLexA- 4 6 promoter in two independent, loci. One OLexA- L 46 promoter was found to fuse to the WUS gene in chromosome II (Mayer et al., 1998), approximately 1 kilobase-pair (Kb) upstream from the putative translation initiation codon (Figure 5A). The second OLexA- 46 promoter fused immediately upstream of the putative translation initiation codon of an open reading frame (ORF) in chromosome V, c encoding a putative basic-helix-loop-helix type transcription factor (deigned ORFI). To IN verify the identity of the PGA6 gene(s), cDNA fragments containing both WUS and the O putative ORF1 were cloned into an XVE vector, and the resulting constructs were used to transform wild-type plants (Bechtold et al., 1993). Explants derived from XVE-ORFI T1 O 10 transgenic plants did not show any apparent inducer-dependent phenotype. In addition, ORFI expression did not appear to be up-regulated by the chemical inducer in pga6 plants, presumably due to the instability of the ORF1 transcript lacking the entire untranslated region (UTR). By contrast, all pga6 mutant phenotypes as described before were observed in the XVE-WUS T2 transgenic plants (Figures 6A and 6B) in an inducerdependent manner (see Example 11 for details).

WUS expression was strictly dependent on different concentrations of the inducer in pga6 mutant plants (Figure 5B). Figure 5C shows expression of PGA6/WUS induced by different concentrations of the inducer. Ten-day-old pga6 seedlings germinated and grown on the MS medium were transferred to an MS medium containing various concentrations of 17-3-estradiol as indicated and cultured for an additional 16 hours before total RNA extraction. Five pg total RNA were used for Northern blot analysis using a WUS cDNA fragment as a probe. Consistent with the inducer concentrationdependent WUS expression, pga6 plants also showed various penetrations of the mutant phenotype in an inducer concentration-dependent fashion (Figure 5B), thus providing a series of conditional mutant alleles for further functional studies.

We also used a 35S-WUS construct for transformation of Arabidopsis thaliana Columbia as well as Landsberg ecotype. Most of the transformants recovered from selection showed strong alteration in seedling development. The hypocotyl was swollen and root tips often enlarged to give rise to shoot-like or embryo-like structures which were unable to further develop (Figures 6C-G). Leaf development, when observed, was also compromised indicating strong alteration of the shoot apical meristem (SAM) which,

O

o 19 CI besides lateral organs, could also form adventitious shoots or embryo-like structures (Figures 6G-H). Overall, all the tissues of the transformants would give rise to organs or embryo-like structures whose further development, however, was impaired. This Cen observation supports the validity of our strategy of using an inducible system to isolate genes involved in the switch to embryo development as the continuous overexpression of e:n regulatory proteins would prevent recovery of the mutants.

I In summary, the above results indicated that tie LexA-4 6 promoter-tagged WUS O in the mutant genome represents PGA6. We will refer to the PGA6 gene/protein as WUS in the future, but use pga6 for the mutation/mutant identified in this study due to different 0 10 properties between the loss- and gain-of-function mutations.

Because the OLeA-4 6 promoter fused approximately 1 Kb upstream from the putative translation start codon (Figure 5A), the WUS promoter presumably remains functional in the mutant genome, leading to no apparent loss-of-function phenotype for the mutation. Nevertheless, the WUS gene was strongly inducible, giving rise to two transcripts, approximately 1.3 and 2.3 Kb (Figure 5C). The shorter transcript was presumably generated from the native transcription initiation site of the WUS gene, in which case the LexA operator sequence might serve as an enhancer to the WUS promoter.

On the other hand, the longer transcript might represent transcription from the OLeXA- 46 promoter. This suggests that the OLexA- 46 sequence can serve as a functional promoter, as well as a transcriptional enhancer for activation tagging.

Example 9 WUS represses LECI expression during embryogenesis The above data indicate that WUS, in addition to its meristem function described previously (Laux et al., 1996; Mayer et al., 1998), also plays a critical role in promoting or maintaining the embryonic potential. We have investigated expression of several embryo- or seed-specific genes in pga6 embryogenic callus and somatic embryos. Root explants derived from pga6 mutant plants were cultured on an MS medium supplemented with the inducer. Under such conditions, whereas embryogenic calli appeared after 10-15 days, somatic embryos and germinating seedlings were generated after 20 days (see Figures 2A-F, 3A-D and 4A-H). Due to the transient expression nature of the XVE 0 O (N inducible system (Zuo et al., 2000a), WUS expression gradually declined upon continued culture on the inductive medium; however, the expression can be strongly re-induced by adding freshly-prepared inducer (Figure 7A).

The LEC1 gene, normally expressed only in embryos and seeds (Lotan et al., 1998), was highly expressed in 20-30-day-old explants, a stage when somatic embryos ec and derived seedlings were generated. Figure 7B shows LEC1 expression from 14 days IN until 28 days when pga6-dependent somatic embryogenesis takes places. LEC1 O expression, however, was dramatically decreased upon reactivation of WUS expression (Figure 7B). No alteration of LEC 1 expression was detected when the explants/calli were O 10 transferred onto the control medium for an additional two days, suggesting that the LEC1 repression was a specific response to the 17-0-estradiol induced WUS expression.

These observations suggest that LECI expression in pga6 explants was not a direct response to WUS overexpression but rather a consequence of the pga6 somatic embryo development. On the other hand, a developmental path redefined by WUS overexpression leads to the repression of LEC1, a gene presumably involved in embryo maturation.

Example Enhanced frequency of somatic embryo formation by addition of 2,4-D Although we were able to generate somatic embryos from various pga6 tissues/organs in the absence of any external hormone, the frequency of somatic embryo fonnation appears to be lower compared to that observed in our original screening conditions, under which a 2,4-D pretreatment was included. To test the effects of 2,4-D on somatic embryogenesis, pga6 root explants were cultured on MS medium with mg/L 2,4-D for 5 days prior to being transferred to an MS medium with or without 10 tM 17-B-estradiol. No somatic embryo formation was observed in the medium without the inducer, whereas numerous somatic embryos were generated after 2-3 weeks culturing in the presence of the inducer. As shown above (see Example 7 and Figures 4A-H) and by the results disclosed in this Example, the absence of 2,4-D pretreatment resulted in a substantially decreased efficiency for somatic embryo formation, indicating that the 2,4-D treatment is able to significantly increase the efficiency of somatic embryo fonnation.

O

O 21 N Collectively, the above results suggest that 2,4-D was able to enhance yet unidentified components in the signaling network, which play a key role in promoting the vegetative- <1 to-embryonic transition. In addition, these results further demonstrated the validity of our working hypothesis as well as the screening strategy, that external hormones alone were incapable of activating key components for the vegetative-to-embryonic transition in Mn Arabidopsis, and that the appropriate external hormone treatment in combination with IND gain-of-function mutations in key regulatory genes is fully capable of promoting the 0 vegetative-to-embryonic transition or somatic embryogenesis.

O

o 22 Cl Example 11 SSomatic embryo formation of explants derived from transgenic plants carrying a pER 10- WUS transgene The XVE vector pER10 is identical to pER 8 except that the hygronycin resistance marker is replaced with a kanamycin-resistance marker (Zuo et al., 2000(a)).

en Full length WUS cDNA was placed under the control of the XVE system in pER 10. Stem IN segments derived from the pER10-US transgenic plants were pre-cultured on the MS 0 medium with 0.5 mg/L 2,4-D for 5 days and then transferred to an MS medium with or without 10 LM 17-P-estradiol. No somatic embryo was observed in medium in the O 10 absence of the inducer, whereas with the inducer, many somatic embryos were generated after 2-3 weeks of culture.

Similar to that of pga6 mutant plants, we found that the 2,4-D pre-treatment substantially increased the frequency of somatic embryo formation in pERIO-WUS transgenic plants.

The method of placing WUS in pER10 or similar vectors which can be regulated enables one to perform the same or similar experiments in plants other than Arabidopsis.

Example 12 Placing W'US under the control of a tissue specific promoter The G10-90 promoter in the XVE vector can be replaced with a tissue-specific promoter a pollen-, root- stem- or leaf-specific promoter). A variety of tissue specific promoters are well known to those of skill in the art. Because expression of a transgene is activated by the chimeric XVE gene which is controlled by a tissue-specific promoter in this Example, the OlexA-4 6 promoter controlling the WUS transgene is therefore tissue-specific in an inducer-dependent manner. This means that WUS will be induced only in the presence of an inducer and only in the specific tissue corresponding to the tissue specific promoter. Appropriate tissues or cell types, can then be collected from the transgenic plants and used for induction of somatic embryos as described in Examples and 11.

Particularly when pollen derived from transgenic plants carrying a pollen-specific promoter-XVE/OlexA-46-WUS vector is used, progeny plants generated from pollen-

O

o 23 c, derived somatic embryos should be haploid instead of diploid (see, Twell et al., 1989 and Twell et al., 1990 for pollen specific promoters). In this embodiment of the <1 invention, a transgenic plant having in its genome a Wuschel (WUS) gene under the control of an inducible, pollen-specific promoter would not normally express the gene.

Pollen from such a plant can be cultured in the presence of the inducer until somatic ¢C embryogenesis occurs, after which the inducer is removed and the haploid embryos are Spermitted to develop into haploid clones according to standard techniques.

O

Example 13 Use of the pERO1-WUS as a silent marker for transformation.

The pERIO-WUS vector can be used directly for transformation of explants without the use of an antibiotic resistance marker. Somatic embryos that formed in the presence of an inducer but in the absence of cytokinin should be transformants, because under such conditions non-transformants will be incapable of forming somatic embryos nor shoots due to the lack of induced WUS gene expression. Upon inducer removal, the embryos and shoots will develop into normal and fertile plants. The vector can include any gene or genes which are desired to be present in the transformed plants and these can be under the control of a desired promoter. The plants selected as a result of selecting for inducible WUS expression-dependent somatic embryos or shoots will contain the desired gene or genes.

If desired, the WUS transgene can be placed into a vector comprising a means of removing the WUS transgene as well as other portions of the vector which are no longer desired, using the XVE-Cre/lox system (Zuo et al., 2001). Such methods are also disclosed in U.S. Patent Application Serial No. 09/439,534, filed 12 November 1999, which is incorporated herein by reference.

O

o 24 Example 14 S....uction of formation of somatic embryos from isolated zygotic embryos ofPGA6 transgenic plants Zygote embryos at late heart stage were isolated from young siliques of PGA6 plants and transferred to either non-inductive medium (MICK: MS salts, 30 g/L sucrose, S0.15 mg/L IAA, 100 mg/L carbenicillin, 50 mg/L kanamycin, and 0.25% phytagel, pH N 5.7) or inductive medium (MICK plus 10.0 4M 17-p-estradiol). After two weeks of 0 culture, 10 zygotic embryos formed embryogenic calli on the inductive medium, but only o 3 zygotic embryos started to form embryogenic calli on the non-inductive MICK medium.

10 After 35 days of culture on the inductive medium, 10 zygotic embryos formed somatic embryos with a regeneration index (the number of somatic embryos per zygote embryo explant) of about 70-100 (Figure 8A). By contrast, on MICK medium without estradiol, only two zygotic embryos regenerated somatic embryos and the somatic embryo regeneration index was about 10-20 (Figure 8B).

Example Generating an apomictic plant Apomixis can be induced by introducing WUS into a plant cell in such a manner that the WUS gene is expressed in the appropriate tissues nucellus tissue). This can be by means of, but is not limited to, placing the WUS gene under the control of a tissuespecific promoter a nucellus-specific promoter), an inducible promoter, or a promoter that is both inducible and tissue-specific. Inducing expression of the WUS gene, e.g. in the nucellus, produces apomixis leading to an apomictic plant. This plant may then be used to establish a true-breeding plant line. Additionally, the vector utilized to transfer WUS into the plant cell can include any other desired heterologous gene in addition to WUS, including but not limited to, a marker gene or a gene to confer a desirable trait upon the plant, a gene resulting in larger plants, faster growth, resistance to stress, etc.

This would lead to the development of an apomictic line with the desired trait.

In a variation of the scheme, plant expression cassettes, including but not limited to monocot or dicot expression cassettes, directing WUS expression to the inner integument or nucellus can easily be constructed. An expression cassette directing

O

O N expression of the WUS DNA sequences to the nucellus was made using the barley Nuc 2 promoter (Doan et al., 1996). The expression was used for plant transformation. Other genes which confer desirable traits can also be included in the cassette.

It is anticipated that transgenic plants carrying the expression cassette will then be capable of producing de novo embryos from WUS expressing nucellar cells. In the case of maize, this is complemented by pollinating the ears to promote normal central cell IND fertilization and endosperm development. In another variation of this scheme, Nucl-:W.US O transformations could be done using afie (fertility-independent endosperm)-null genetic background which would promote both de novo embryo development and endosperm O 10 development without fertilization (Ohad et al., 1999). Upon microscopic examination of the developing embryos it will be apparent that apomixis has occurred by the presence of embryos budding off the nucellus. In yet another variation of this scheme the WUS DNA sequences could be delivered as described above into a homozygous zygotic-embryolethal genotype. Only the adventive embryos produced from somatic nucellus tissue would develop in tie seed.

While the invention has been disclosed by reference to the details of preferred embodiments of the invention, it is to be understood that the disclosure is intended in an illustrative rather than in a limiting sense, as it is contemplated that modifications will readily occur to those skilled in the art, within the spirit of the invention and the scope of the appended claims.

LIST OF REFERENCES Bashaw, Crop Science, 1980. 20: p. 112.

Bechtold, Ellis, and Pelletier, In planta Agr by infiltration of adult Arabidopsis thaliana plants. C. R. Acad. Sci. Ser. III Sci. Vie, 1993. 3 1 6 p. 119441199.

Brand, Fletcher, Hobe, Meyerowitz, and Simon, Dependence of stemn cell fate in Arabidopsis on a feedback loop regulated by CL V'3 activity. Science, 2000. 289: p. 617-619.

Clark, Cell signalling at the shoot mierist em. Nat. Rev. Mol. Cell Biol., 2001. 2: p.

276-284.

Doan, Linnestad, C. anid Olsen, Isolation of inolecular inarlers fr-om the barley endosperin coenocyte and the surrounding nucellus cell layers. Plant Mo. Bio., 1996. 31: p. 8 77-886 Ishige, Takaichi, Foster, Chua, and Oeda, A G-box inotif rGCGACGTGCC) let ramer confers high-lev~el constitutive expression in dicot and inonocot plants. Plant 1999. 18:443-448.

Kakimoto, CKII, a histidine kinase horntolog implicated1 in cv)tokin in signal transduction. Science, 1996. 274: p. 982-985.

Koncz, Martini, Mayerhofer, Koncz-Kalman, Korber, PRedei, and Schell, High-frequency T-DAA-mzedia ted gene tagging in plants. Proc. Natl. Acad. Sci.

USA, 1939. 86(2 p. 8467-847 1.

Laux, Mayer, Berger, and Jurgens, The WUSCHEL genie is required for shoot and floral Pneristem integrity in Arabidopsis. Development, 1996. 122: p. 87-96.

Lin, Hwang, G.J, and Zimmerman, Isolation and characterization of a diverse set ofgenesfroni carrot somatic einbiyos. Plant Physiol., 1996. 112: p. 1365-1374.

Liu, Mitsukawa, Oosurni, and Whittier, Efficient isolation and mnapping of Arabidopsis thaliana T-DNA4 imLwert junctions by thermal asymnmetric interlaced PCR. Plant 1995. 8: p. 457-463.

Lotan, Ohto, Yee, West, Lo, Kwong, Yamagishi, K., Fischer, Goldberg, and Harada, JTJ, Arabidopsis LEAFY COTYLEDON! is sufficient to induce embryo development in vegetative cells. Cell, 1998. 93: p. 1195-1205.

Mayer, Scboof, 1-aecker, Lenhard, Jurgens, Laux, Role of WUSCHEL in regulating stemt cellfate in the Arabidopsis shoot mieristem. Cell, 1998. p. 805-815.

w Meinke, A homoeotic mutant of Arabidops Science, 1992. 258: p. 1647-1650.

5 Meinke, Franzmann, Nickle, and Yeung, Leafy cotyledon mutants ofArabidopsis. Plant Cell, 1994. 6: p. 1049-1064.

en Mordhorst, Toonen, and de Vries, Plant embryogenesis. Crit. Rev.

,n Plant Sci., 1997. 16: p. 535-576.

N

O Mordhorst, Voerman, Hartog, Meijer, van Went, Koornneef, c and de Vries, Somatic embiyogenesis in Arabidopsis thaliana is facilitated by O mutations in genes repressing meristematic cell divisions. Genetics, 1998. 149: p. 549o 563.

CN Murashige, and Skoog, A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol. Plant, 1962. 15: p. 473-497.

Ohad, Yadegari, Margossian, Hannon, Michaeli, Harada, J.J., Goldberg, and Fischer, Mutations in FIE, a WD polycomb group gene, allow endosperm development without fertilization. The Plant Cell, 1999. 11: p. 407-416.

Parlevliet JE et al. Citrus. Proc. Am. Soc. Hort. Sci., 1959. 74: p. 252-260.

Pepin et al, Crop Science, 1971. 11: p. 445-448.

Rieger et al., in Glossary of Genetics and Cytogenetics, Springer-Verlag, New York, 1976 Sambrook, Fritsch, and Maniatis, Molecular cloning: A Laboratory Manual.

1989, Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.

Schmidt, Guzzo, Toonen, and de Vries, A leucine-rich repeat containing receptor-like kinase marks somatic plant cells competent to form embryos.

Development, 1997. 124: p. 2049-2062.

Schoof, Lenhard, Haecker, Mayer, Jurgens, and Laux, The stem cell population of Arabidopsis shoot meristems in maintained by a regulatory loop between the CLAVATA and WUSCHEL genes. Cell, 2000. 100: p. 635-644.

Skoog, and Miller, Chemical regulation of growth and organ fonnation in plant tissues cultured in vitro. Symp. Soc. Exp. Biol., 1957. 11: p. 118-131.

Spemann, and Mangold, Uber Induktion von Emblyonalanlagen durch Implantation artfremder Organisatoren. Archiv fur Mikroskopische Anatomie und Entwicklungsmechanik, 1924. 100: p. 599-638.

Stone, Kwong, LW. Yee, Pelletier, Lepiniec, Fisher, Goldberg, and H-arada, LEAFY COTYLEDON 2 enlcodes a B3 doinain transcription factor that induces einbryo developmnent. Proc. Nati, Acad. Sci. 2001. 98:11806-1 1811.

Sugiyamia, Organogenesis in vitro. Curr. Opin. Plant Biol., 1999. 2: p. 61-64.

Sugiyamia, Genetic analysis of plant moiphogenesis in vitro. Int. Rev. Cytol., 2000.

196: p. 67-84.

Thomas, Gene expression during plant ibryogenesis and germination- an overview. Plant Cell, 1993. 5: p. 140 1-14 Twel], Wing, R. Yamaguchi, J. and McCormick S. Isolation and expression of an anther-specific gene from tomnato. Mol. Gen. Genet., 1989. 217: p. 240-245.

Twel], D. Yamnaguchi, J. anid McCormiick, S. Pollen-specifc gene expression inl transgenic plants. coordinate regulation of two different tomnato gene promnoters during nicrosporogenesis. Development, 1990, 109: p. 705-713.

Waites, and Simon, Signaling cell fate in plant meristenis: Three clubs on one tousle, Cell, 2000. 103: p. 835-838.

Wilson et al, Proceedings of the International Workshop on Apomixis in Rice, Changsha, People's Republic of China, Jan. 13-Jan. 15, 1992. Hunan Hybrid Rice Research Center, Changsha, People's Republic of China.

Wu, H-aberland, Zhou, and Koop, Somatic em bryo genesis, formation of 11norp ho genetic callus and normal development in zygotic emnblyos of Arabidopsis tlialiana in vitro. Protoplasma, 1992. 169: p. 89-96.

Zimmerman, Soinatic enibryogenesis: A miodel for early development inl higher plants. Plant Cell, 1993. 5: p. 1411-1423.

Zuo, Niu, and Chua, Anl estrogen receptor-based transactivator X"VE mnediates highly inducible gene expression in transgenic plants. Plant 2 0 00a. 24: p.

265-273.

Zuo, Niu, Nishizawa, Wu, Kost, and Chua, KOPJJGAN, an Arabidopsis endo-], 4-b-glucanase, localizes to the cell plate by polarized targeting and is essential for cytokinesis. Plant Cel], 2000b. 12(7): p. 1137-1152.

Zuo, Niu, Moller, and Chua, Chemzical-regulated, site-specific DNA exrcision in transgenic plants. Nat. Biotechno]., 2001. 19. p. 157-161.

Patents and Patent Applications

O

o 29 Pu Q Published Patent Application WO 00/24914, published 4 May 2000.

e¢3 Published Patent Application WO 01/23575, published April 5, 2001.

U.S. Patent No. 5,710,367 c U.S. Patent No. 5,811,636

O

U.S. Patent No. 6,028,185

O

U.S. Patent No. 6,229,064 0 0U.S. Patent No. 6,239,327

Claims (69)

1. A method to promote somatic embryogenesis from a tissue or organ of a plant, said method comprising overexpressing a Wuschel gene in said tissue or organ.

2. The method of claim I wherein said method is performed in the absence of external plant hormones. IN

3. The method of claim I wherein said method comprises a step of culturing in the opresence of an auxin.

4. The method of claim 1 wherein said tissue or organ comprises a promoter controlling expression of said Wuschel gene.

The method of claim 4 wherein said tissue or organ is transgenic for said promoter.

6. The method of claim 4 wherein said promoter is inducible.

7. The method of claim I wherein said Wuschel gene which is overexpressed is in a vector.

8. The method of claim 7 wherein said vector comprises an inducible promoter controlling expression of said Wuschel gene.

9. The method of claim 8 wherein said inducible promoter is inducible by an estrogen.

The method of claim 9 wherein said estTogen is 17-f3-estTadiol.

11. The method of claim 7 wherein said vector comprises a tissue specific promoter. 0 o 31 'l

12. The method of claim 1 wherein said plant is selected from the group consisting of trees and crops. en3

13. A method to generate plant somatic embryos wherein said method comprises overexpressing a Wuschel gene in a tissue or organ of a plant. en3 en3 IN

14. The method of claim 13 wherein the method is performed in the absence of 0extemal plant hormones. Cl 10

15. The method of claim 13 wherein said method comprises a step of culturing in the C'-l presence of an auxin.

16. The method of claim 13 wherein said tissue or organ comprises a promoter controlling expression of said Wuschel gene.

17. The method of claim 16 wherein said tissue or organ is transgenic for said promoter.

18. The method of claim 16 wherein said promoter is inducible.

19. The method of claim 13 wherein said Wuschel gene which is overexpressed is in a vector.

The method of claim 19 wherein said vector comprises an inducible promoter controlling expression of said J'uschel gene.

21. The method of claim 20 wherein said inducible promoter is inducible by an estrogen.

22. The method of claim 21 wherein said estrogen is 17-[3-estradiol. S32

23. The method of claim 19 wherein said vector comprises a tissue specific promoter.

24. The method of claim 13 wherein said plant is selected from the group consisting en3 of trees and crops. en

25. A method for generating shoots from a tissue or organ of a plant, said method IND comprising overexpressing a Wuschel gene in said tissue or organ. 1

26. The method of claim 25 wherein said method is performed in the absence of O 10 external plant hormones.

27. The method of claim 25 wherein said method comprises a step of culturing in the presence of an auxin.

28. The method of claim 25 wherein said vegetative tissue or organ comprises a promoter controlling expression of said Wuschel gene.

29. The method of claim 28 wherein said vegetative tissue or organ is transgenic for said promoter.

The method of claim 28 wherein said promoter is inducible.

31. The method of claim 25 wherein said Wuschel gene which is overexpressed is in a vector.

32. The method of claim 31 wherein said vector comprises an inducible promoter controlling expression of said Wuschel gene.

33. The method of claim 32 wherein said inducible promoter is inducible by an estrogen. 0 0 33 C,

34. The method of claim 33 wherein said estrogen is 17-p-estradiol.

The method of claim 31 wherein said vector comprises a tissue specific promoter.

36. The method of claim 25 wherein said plant is selected from the group consisting M of trees and crops. 0

37. A method of selecting plants transformed with a vector comprising a silent o selectable marker, said silent selectable marker being a Wuschel gene, said method comprising a) transforming a plant, plant tissue or plant organ with a vector comprising said Wuschel gene under the control of an inducible promoter; b) culturing said plant, plant tissue or plant organ in the presence of an inducer of said inducible promoter; c) selecting somatic embryos which form; and d) growing said somatic embryos in the absence of said inducer to develop plants, wherein said plants which develop are transformed with said vector.

38. The method of claim 37 wherein method is performed in the absence of external plant honrmones.

39. The method of claim 37 wherein said method comprises a step of culturing in the presence of an auxin.

The method of claim 37 wherein said inducible promoter is inducible by an estrogen.

41. The method of claim 40 wherein said estrogen is 17-p-estradiol.

42. The method of claim 37 wherein said vector comprises a tissue specific promoter. O 34

43. The method of claim 37 wherein said vector comprises a gene of interest.

44. The method of claim 43 wherein said gene of interest is under the control of a promoter. IND

45. The method of claim 44 wherein said promoter controlling the gene of interest is a 0 constitutive promoter or an inducible promoter. O 10

46. The method of claim 37 wherein said plant, plant tissue or plant organ is selected from the group consisting of trees or crops.

47. The method of claim 37 wherein said vector comprises means for deleting said Wfuschel gene.

48. A method for inducing apomixis in a plant cell comprising overexpressing Wuschel in said plant cell.

49. The method of claim 48 wherein said plant cell is transformed with a vector comprising Wuschel.

The method of claim 49 wherein Wuschel is under the control of an inducible promoter.

51. The method of claim 49 wherein said vector comprises a second gene in addition to said Wuschel.

52. The method of claim 51 wherein said second gene encodes a marker.

53. The method of claim 51 wherein said second gene confers a desired trait to said plant. 0 0

54. An apomictic plant comprising Wuschel under the control of an inducible promoter. en3

55. A method for conferring apomixis to a plant comprising en3 M a. stably transforming a responsive plant cell with a DNA sequence comprising Ste at least one Wuschel coding sequence under the control of a promoter to 0 produce a transgenic plant cell, said promoter being selected from the group o consisting of a tissue-specific promoter, an inducible promoter and a promoter that is both tissue-specific and inducible, b. regenerating a transgenic plant from said transgenic plant cell, whereby overexpression of a Wuschel polypeptide from the Wuschel coding sequence in the transgenic plant results in the formation of an apomictic embryo.

56. The method of claim 55 wherein the promoter is a tissue-specific promoter.

57. The method of claim 55 wherein the promoter is an inducible promoter.

58. The method of claim 55 wherein the promoter is both tissue-specific and inducible.

59. The method of claim 56 wherein the promoter is a nucellus-specific promoter.

The method of claim 56 wherein the transgenic plant cell comprises at least one additional heterologous DNA sequence in its genome.

61. The method of claim 55 wherein the Wuschel gene is PGA6.

62. A plant produced by the method of claim

63. A method for producing a haploid plant comprising O o 36 C a. stably transforming a plant cell with a DNA sequence comprising at least one Wuschel coding sequence under the control of a promoter to produce a <1 transgenic plant cell, wherein the promoter is selected from the group consisting of a haploid tissue specific promoter, an inducible promoter and a promoter that is both haploid-tissue specific and inducible, en b. generating a transgenic plant from said transgenic plant cell, N c. overexpressing the Wuschel coding sequence in a haploid tissue of said 0 transgenic plant to produce a haploid somatic embryo, d. growing said embryo into a haploid plant.

64. The method of claim 63 wherein the promoter is a haploid tissue specific promoter.

The method of claim 64 wherein the promoter is a pollen-specific promoter, and the haploid tissue of the transgenic plant is pollen.

66. The method of claim 64 wherein the promoter is an ovule-specific promoter, and the haploid tissue of the transgenic plant is ovule tissue.

67. The method of claim 63 wherein the promoter is an inducible promoter.

68. The method of claim 63 wherein the promoter is both haploid-tissue specific and inducible.

69. The method of claim 67 wherein the promoter is pollen-specific, and the haploid plant cell is a pollen cell. The method of claim 66 wherein the overexpressing step c. is achieved in excised haploid tissue cultured in the presence of the inducer specific for the inducible promoter, for a time sufficient to induce formation of the haploid somatic embryo, 0 0 37 Cl followed by withdrawal of the inducer, and further wherein the growing step d. is carried out in the absence of the inducer. I-- C, 1^, 0O