CA2215335C

CA2215335C - Cauliflower floral meristem identity genes and methods of using same

Info

Publication number: CA2215335C
Application number: CA002215335A
Authority: CA
Inventors: Martin F. Yanofsky
Original assignee: University of California
Current assignee: University of California
Priority date: 1996-01-26
Filing date: 1996-01-26
Publication date: 2009-01-06
Anticipated expiration: 2016-01-26
Also published as: WO1997027287A1; CA2215335A1

Abstract

The present invention provides a nucleic acid molecule encoding a CAULIFLOWE R (CAL) gene product such as a nucleic acid molecule encoding Arabidopsis thatiana CAL and a nucleic acid molecule encoding Brassica oleracea CAL (BoCAL). The invention also provides a nucleic acid molecule encoding a truncated CAL gene product such as a nucleic acid molecule encoding Brassica oleracea var. botrytis CAL (BobCAL). The invention also provides a nucleic acid containing the Arabidopsis thaliana CAL gene, a nucleic acid molecule containing the Brassica oleracea CAL gene and a nucleic acid molecule containing the Brassica oleracea var. botritis CAL gene. The invention further provides a kit for converting shoot meristem to floral meristem and a kit for promoting early flowering in an angiosperm. The invention provides a CAL polypeptide and an antibody that specifically binds CAL polypeptides. In addition, the invention provides the truncated BobCAL polypeptide and an antibody that specifically binds truncated BobCAL polypeptide. The invention further provides a method of identifying a Brassica having a modified CAL CAL allele by detecting a polymorphism associated with a CAL CAL locus, where the CAL CAL locus comprises a modified CAL CAL allele that does not encode an active CAL gene product.

Description

CAULIFLOWER FLORAL MERISTEM IDENTITY GENES

AND METHODS OF USING SAME

This work was supported by grant DCB-9018749 awarded by the National Science Foundation. The United States Government has certain rights in this invention.
BACICGROUND OF THE INVENTION

FIELD OF THT+'. INVENTION

This invention relates generally to the field of plant flowering and more specifically to genes involved in the regulation of floweri.ng.

BACKGROUND INFORMATION

A flower is the reproductive structure of a -flowering plant. Following fertilization, the ovary of the flower becomes a fruit and bears seeds. As a practical consequence, production of fruit and seed-derived crops such as grapes, beans, corn, wheat and rice is dependent upon flowering.

Early in the plant life cycle, vegetative growth occurs, and roots, stems and leaves are formed.
During the later period of reproductive growth, flowers as well as new shoots or branches develop. However, the factors responsible for the transition from vegetative to Vb reproductive growth, and the onset of flowering, are poorly understood.
r A variety of external signals, such as length of daylight and temperature, affect the time of flowering. The time of flowering also is subject to genetic controls that prevent young plants from flowering .
prematurely. Thus, the pattern of genes expressed in a plant is an important determinant of the time of flowering.

Given these external signals and genetic controls, a relatively fixed period of vegetative growth precedes flowering in a particular plant species. The length of time required for a crop to mature to flowering limits the geographic location in which it can be grown and can be an important determinant of yield. In addition, since the time of flowering determines when a plant is reproductively mature, the pace of a plant breeding program also depends upon the length of time required for a plant to flower.

Traditionally, plant breeding involves generating hybrids of existing plants, which are examined for improved yield_or quality. The improvement of existing plant crops through plant breeding is central to increasing the amount of food grown in the world since the amount of land suitable for agriculture is limited.
For example, the development of new strains of wheat, corn and rice through plant breeding has increased the yield of these crops grown in underdeveloped countries such as Mexico, India and Pakistan. Unfortunately, plant breeding is inherently a slow process since plants must be reproductively mature before selective breeding can proceed.

YI

For some plant species, the length of time needed to mature to flowering is so long that selective breeding, which requires several rounds of backcrossing progeny plants with their parents, is impractical. For example, perennial trees such as walnut, hickory, oak, maple and cherry do not flower for several years after planting. As a result, breeding of such plant species for insect or disease-resistance or to produce improved wood or fruit, for example, would require many years, even if only a few rounds of selection were performed.

Methods of promoting early flowering can make breeding of long generation plants such as trees practical for the first time. Methods of promoting early flowering also would be useful for shortening growth periods, thereby broadening the geographic range in which a crop such as rice, corn or coffee can be grown.
Unfortunately, methods for promoting early flowering in a plant have not yet been described. Thus, there is a need for methods that promote early flowering. The present invention satisfies this need and provides related advantages as well.

~

STTMMARY OF THE INVENTION

The present invention provides a nucleic acid molecule encoding a CAULIFLOWER (CAL) gene product. For example, the invention "
provides a nucleic acid molecule encoding Arabidopsis thaliana CAL and a nucleic acid molecule encoding Brassica oleracea CAL.

The invention also provides a nucleic acid molecule encoding a truncated CAL gene product. For example, the invention provides a nucleic acid molecule encoding the truncated Brassica oleracea var. botrytis CAL gene product. The invention also provides a nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule encoding a CAL gene product, a truncated CAL gene product, or a complementary sequence thereto.

The invention further provides the Arabidopsis thaliana CAL gene, Brassica oleracea CAL gene and Brassica oleracea var. botrytis CAL gene. In addition, the invention provides a nucleotide sequence that hybridizes under relatively stringent conditions to the Arabidopsis thaliana CAL gene, Brassica oleracea CAL gene or Brassica oleracea var. botrytis CAL gene, or a complementary sequence thereto.

~

The invention also provides vectors, including expression vectors, containing a nucleic acid molecule encoding a CAL gene product. The invention further provides a kit for converting shoot meristem to floral = 5 meristem in an angiosperm and a kit for promoting early flowering in an angiosperm.

In addition, the invention provides a CAL
polypeptide, such as the Arabidopsis thaliana CAL
polypeptide or the Brassica oleracea CAL polypeptide, as well as an antibody that specifically binds a CAL
polypeptide. The invention further provides the truncated Brassica oleracea var. botrytis CAL polypeptide and an antibody that specifically binds the truncated Brassica oleracea var. botrytis CAL polypeptide.

The invention further provides a method of identifying a Brassica having a modified CAL allele by detecting a polymorphism associated with a CAL locus, where the CAL locus comprises a modified CAL allele that does not encode an active CAL gene product. For example, the polymorphism can be a restriction fragment length polymorphism and the modified CAL allele can be the Brassica oleracea var. botrytis CAL allele.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates the nucleotide (SEQ ID
NO: 1) and amino acid (SEQ ID NO: 2) sequence of the Arabidopsis thaliana AP3 cDNA.

Figure 2 illustrates the nucleotide (SEQ ID

NO: 3) and amino acid (SEQ ID NO: 4) sequence of the Brassica oleracea API cDNA.

Figure 3 illustrates the nucleotide (SEQ ID =
NO: 5) and amino acid (SEQ ID NO: 6) sequence of the Brassica oleracea var. botrytis AP1 cDNA.

Figure 4 illustrates the nucleotide (SEQ ID
NO: 7) and amino acid (SEQ ID NO: 8) sequence of the Zea mays AP1 cDNA. The GenBank accession number is L46400.

Figure 5 illustrates the nucleotide (SEQ ID
NO: 9) and amino acid (SEQ ID NO: 10) sequence of the Arabidopsis thaliana CAL cDNA.

Figure 6 illustrates the nucleotide (SEQ ID
NO: 11) and amino acid (SEQ ID NO: 12) sequence of the Brassica oleracea CAL cDNA.

Figure 7 illustrates the nucleotide (SEQ ID
NO: 13) and amino acid (SEQ ID NO: 14) sequence of the Brassica oleracea var. botrytis CAL cDNA.

Figure 8 illustrates CAL gene structure and provides a comparison of various CAL amino acid sequences.

Figure 8A. Exon-intron structure of Arabidopsis CAL gene. Exons are shown as boxes and introns as a solid line. Sizes (in base pairs) are indicated above. Locations of changes resulting in mutant alleles are indicated by arrows. MADS and K
domains are hatched.

Figure 8B. An alignment of three deduced amino acid sequences of CAL cDNAs. The complete Arabidopsis thaliana CAL amino acid sequence is displayed. The Brassica oleracea CAL (BoCAL) and Brassica oleracea var.
botrytis CAL (BobCAL) amino acid sequences are shown directly below the Arabidopsis sequence where the sequences differ. The API amino acid sequence is shown for comparison. The MADS domain is amino acids 1-57 and the K domain is underlined. GenBank accession numbers are as follows: Arabidopsis thaliana CAL (L36925);
Brassica oleracea CAL (L36926) and Brassica oleracea var.
botrytis CAL (L36927).

Figure 9 illustrates the nucleotide (SEQ ID
NO: 15) and amino acid (SEQ ID NO: 16) sequence of the Arabidopsis thaliana LEAFY (LFY) cDNA.

Figure 10 illustrates the genomic sequence of Arabidopsis thaliana APi (SEQ ID NO: 17).

Figure 11 illustrates the genomic sequence of Brassica oleracea AP1 (SEQ ID NO: 18).

Figure 12 illustrates the genomic sequence of Brassica oleracea var. botrytis AP1 (SEQ ID NO: 19).

Figure 13 illustrates the genomic sequence of Arabidopsis thaliana CAL (SEQ ID NO: 20).

Figure 14 illustrates the genomic sequence of Brassica oleracea CAL (SEQ ID NO: 21).

Figure 15 illustrates the genomic sequence of Brassica oleracea var. botrytis CAL (SEQ ID NO: 22).
Figure 16 illustrates the nucleotide (SEQ ID

NO: 23) and amino acid (SEQ ID NO: 24) sequence of the rat glucocorticoid receptor ligand binding domain.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a nucleic acid molecule encoding a CAULIFLOWER (CAL) gene product, which is a floral meristem identity gene product involved in the conversion of shoot meristem to floral meristem. For example, the invention provides a nucleic acid molecule encoding Arabidopsis thaliana CAL and a nucleic acid molecule encoding Brassica oleracea CAL (BoCAL) (Kempin et al., Science, 267:522-525 (1995)).

As disclosed herein, a CAL gene product can be expressed in an angiosperm, thereby converting shoot meristem to floral meristem in the angiosperm or promoting early flowering in the ang:~osperm. The invention also provides a nucleic acid' molecule encoding a truncated CAL gene product such as a nucleic acid molecule encoding Brassica oleracea var.
boLrytis CAL (BobCAL). The invention also provides a nucleic acid molecule containing the Arabidopsis thaliana CAL gene, a nucleic acid molecule containing the Brassica oleracea CAL gene and a nucleic acid molecule containing the Brassica oleracea var. botrytis CAL gene. The invention further provides a kit for converting shoot meristem to floral meristem and a kit for promoting early flowering in an angiosperm. The invention provides a CAL
polypeptide and an antibody that specifically binds CAL
polypeptide. In addition, the invention provides the truncated BobCAL polypeptide and an antibody that specifically binds the truncated BobCAL polypeptide. The invention further provides a method of identifying a Brassica having a modified CAL allele by detecting a polymorphism associated with a CAL locus, where the CAL

locus comprises a modified CAL allele that does not encode an active CAL gene product.

The present invention provides a non-naturally occurring angiosperm containing a first ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product. For example, the invention provides a transgenic angiosperm containing a first ectopically expressible floral meristem identity gene product such as APETALAl (APl), CAULIFLOWER (CAL) or LEAFY (LFY). Such a transgenic angiosperm can be, for example, a cereal plant, leguminous plant, oilseed plant, tree, fruit-bearing plant or ornamental flower.

A flower, like a leaf or shoot, is derived from the shoot apical meristem, which is a collection of ' 30 undifferentiated cells set aside during embryogenesis.

The production of vegetative structures, such as leaves or shoots, and of reproductive structures, such as flowers, is temporally segregated, such that a leaf or shoot arises early in a plant life cycle, while a flower 5 develops later- The transition from vegetative to reproductive development is the consecquence of a process termed floral induction (Yanofsky, Ann. Rev. Plant Phvsiol. Plant Mol, Biol. 46:167-188 (1995)).

Once induced, shoot apical meristem either 10 persists and produces floral meristem, which gives rise to flowers, and lateral meristem, which gives rise to branches, or is itself converted to floral meristem. The fate of floral meristem is to differentiate into a single flower having a fixed number of floral organs in a whorled arrangement. Dicots, for example, contain four whorls (concentric rings) in which sepals (first whorl) and petals (second whorl) surround stamens (third whorl) and carpels (fourth whorl).

Although shoot meristem and floral meristem both consist of meristemic tissue, shoot meristem is distinguishable from the more specialized floral meristem. Shoot meristem generally is indeterminate and gives rise to an unspecified number of floral and lateral meristems. In contrast, floral meristem is determinate and gives rise to the fixed number of floral organs that comprise a flower.

By convention herein, a wild-type gene sequence is represented in upper case italic letters (for example, `

APE2'ALAI), and a wild-type gene product is represented in upper case non-italic letters (APETALAl). Further, a mutant gene allele is represented in lower case italic letters (api), and a mutant gene product is represented in lower case non-italic letters (apl).

Genetic studies have identified a number of genes involved in regulating flower development. These genes can be classified into different groups depending on their function. Flowering time genes, for example, are involved in floral induction and regulate the transition from vegetative to reproductive growth. In comparison, the floral meristem identity genes, which are the subject matter of the present invention as disclosed herein, encode proteins that promote the conversion of shoot meristem to floral meristem. In addition, floral organ identity genes encode proteins that determine whether sepals, petals, stamens or carpels are formed (Yanofsky, supra, 1995; Weigel, Ann. Rev. Genetics 29:19-39 (1995)). Some of the floral meristem identity gene products also have a role in specifying organ identity.

Floral meristem identity genes have been identified by characterizing genetic mutations that prevent or alter floral meristem formation. Among floral meristem identity gene mutations in Arabidopsis thalzana, those in the gene LEAFY (LFY) generally have the strongest effect on floral meristem identity. Mutations in LFY completely transform the basal-most flowers into secondary shoots and have variable effects on later-arising (apical) flowers. In comparison, mutations in the floral meristem identity gene APETALA1 _(AP1) result in replacement of a few basal flowers by =
inflorescence shoots that are not subtended by leaves.
An apical flower produced in an api mutant has an indeterminate structure in which a flower arises within a flower. These mutant phenotypes indicate that both API
and LFY contribute to establishing the identity of the floral meristem although neither gene is absolutely required. The phenotype of 1fy api double mutants, in which structures with flower-like characteristics are very rare, indicates that LFY and API encode partially redundant activities.

In addition to the LFY and AP1 genes, a third locus that greatly enhances the api mutant phenotype has been identified in Arabidopsis. This locus, designated CAULXFLOWER (CAL), derives its name from the resulting "cauliflower" phenotype, which is strikingly similar to the common garden variety of cauliflower. In an apl cal double mutant, floral meristem that develops behaves as shoot meristem in that there is a massive proliferation of meristems in the position that normally would be occupied by a single flower. However, a plant homozygous for a particular cal mutation (cal-1) has a normal phenotype, indicating that AP1 can substitute for the loss of CAL in these plants. In addition, because floral meristem that forms in an api mutant behaves as shoot meristem in an api cal double mutant, CAL can largely =
substitute for AP1 in specifying floral meristem. These genetic data indicate that CAL and AP1 encode activities that are partially redundant in converting shoot meristem to floral meristem.

Other genetic loci play at least minor roles in specifying floral meristem identity. For example, although a mutation in APETALA2 (AP2) alone does not result in altered inflorescence characteristics, ap2 apl double mutants have indeterminate flowers (fiowers with shoot-like characteristics) (Bowman et al., Development 119:721-743 (1993)). Also, mutations in the CLAVATAI
(CLV1) gene result in an enlarged meristem and lead to a variety of phenotypes (Clark et al., Development 119:397-418 (1993)). In a c1 vl api double mutant, formation of flowers is initiated, but the center of each --flower often develops as an indeterminate inflorescence.

Thus, mutations in CLAVATAI result in the loss of floral meristem identity in the center of wild-type flowers.
Genetic evidence also indicates that the gene product of UNUSUAL FLORAL ORGANS (UFO) plays a role in determining the identity of floral meristem. Additional floral meristem identity genes associated with altered floral meristem formation remain to be isolated.

Mutations in another locus, designated TERMINAL
FLOWER (TFL), produce phenotypes that generally are reversed as compared to mutations in the floral meristem identity genes. For example, tf1 mutants flower early, and the indeterminate apical and lateral meristems develop as determinate floral meristems (Alvarez et al., Plant J. 2:103-116 (1992)). These characteristics indicate that the TFL promotes maintenance of shoot meristem. TFL also acts directly or indirectly to negatively regulate APl and LFY expression in shoot meristem since APi and LFY are ectopically expressed in the shoot meristem of tf1 mutants (Gustafson-Brown et al., Cell 76:131-143 (1994); Weigel et al., Cell 69:843-859 (1992)). It is recognized that a plant having a mutation in TFL can have a phenotype similar to a non-naturally occurring angiosperm of the invention.
Such tfl mutants, however, are explicitly excluded from the scope of the present invention.

The results of such genetic studies indicate that several floral meristem identity gene products, including AP1, CAL and LFY, act redundantly to convert shoot meristem to floral meristem and that TFL acts directly or indirectly to negatively regulate expression of the floral meristem identity genes. As disclosed herein, ectopic expression of a single floral meristem identity gene product such as AP1, CAL or LFY is sufficient to convert shoot meristem to floral meristem.
Thus, the present invention provides a non-naturally occurring angiosperm that contains an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product, provided that such ectopic expression is not due to a mutation in an endogenous TERMINAL FLOWER gene.

As disclosed herein, an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product can be, for example, a transgene encoding a floral meristem identity gene product under control of a heterologous gene regulatory element. In addition, such an ectopically expressible nucleic acid molecule can be an endogenous floral meristem identity gene coding sequence that is placed under control of a heterologous 5 gene regulatory element. The ectopically expressible nucleic acid molecule also can be, for example, an endogenous floral meristem identity gene having a modified gene regulatory element such that the endogenous floral meristem identity gene is no longer subject to 10 negative regulation by TFL.

The term "ectopically expressible" is used herein to refer to a gene transcript or gene product that can be expressed in a tissue other than a tissue in which it normally is produced. The actual ectopic expression 15 thereof -is dependent on various factors and can be constitutive or inducible expression. As disclosed herein, AP1, which normally is expressed in floral meristem, is ectopically expressible in shoot meristem.
As disclosed herein, when a floral meristem identity gene product such as AP1, CAL or LFY is ectopically expressed in shoot meristem, the shoot meristem is converted to floral meristem and early flowering can occur (see Examples II, IV and V).

In particular, an ectopically expressible nucleic acid molecule encpding a floral meristem identity gene product can be expressed prior to the developmental time at which the corresponding endogenous gene normally is expressed. For example, an Arabidopsis plant grown , under continuous light conditions expresses API just prior to day 18, when normal flowering begins. However, as disclosed herein, AP1 can be ectopically expressed in shoot meristem earlier than day 18, resulting in early conversion of shoot meristem to floral meristem and early flowering. As shown in Example IID, a transgenic Arabidopsis plant that ectopically expresses AP1 in shoot meristem under control of a constitutive promoter flowers earlier than the corresponding non-transgenic plant (day as compared to day 18).

10 As used herein, the term "floral meristem identity gene product" means a gene product that promotes conversion of shoot meristem to floral meristem. As disclosed herein, expression of a floral meristem identity gene product such as APi, CAL or LFY in shoot meristem can convert shoot meristem to floral meristem.
Furthermore, expression of a floral meristem identity gene product in shoot meristem also can promote early flowering (Examples IID, IVA and V). A floral meristem identity gene product is distinguishable from a late flowering gene product or an early flowering gene product, which are not encompassed within the present invention. In addition, reference is made herein to an "inactive" floral meristem identity gene product, as exemplified by BobCAL (see below). Expression of an inactive floral meristem identity gene product in an angiosperm does not result in the conversion of shoot meristem to floral meristem in the angiosperm.

A floral meristem identity gene product can be, for example, an AP1 gene product such as Arabidopsis AP1, which is a 256 amino acid gene product encoded by the API
cDNA sequence isolated from Arabidopsis thaliana (Figure 5, SEQ ID NO: 2). The Arabidopsis AP1 cDNA
encodes a highly conserved MADS domain, which can function as a DNA-binding domain, and a K domain, which is structurally similar to the coiled-coil domain of keratins and can be involved in_protein-protein interactions.

In Arabidopsis, AP1 RNA is expressed in flowers but is not detectable in roots, stems or leaves (Mandel et al., Nature 360:273-277 (1992), which is incorporated herein by reference). The earliest detectable expression of API RNA is in young floral meristem at the time it initially forms on the flanks of shoot meristem.
Expression of AP1 increases as the floral meristem increases in size; no AP1 expression is detectable in shoot meristem. In later stages of development, API
expression ceases in cells that will give rise to reproductive organs (stamens and carpels), but is maintained in cells that will give rise to non-reproductive organs (sepals and petals; Mandel, supra, 1992) .

As used herein, the term "APETALAI" or "AP1"
means a floral meristem identity gene product that is characterized, in part, by having an amino acid sequence that is related to the Arabidopsis APi amino acid sequence shown in Figure 1(SEQ ID NO: 2) or to the Zea mays AP1 amino acid sequence shown in Figure 4 (SEQ ID
NO: 8). In nature, APi is expressed in floral meristem.

CAULIFLOWER (CAL) is another example of a floral meristem identity gene product. As used herein, the term "CAULIFLOWER" or "CAL" means a floral meristem identity gene product that is characterized in part by having an amino acid sequence that has at-least about 70 percent identity with the amino acid sequence shown in Figure 5 (SEQ ID NO: 10) in the region from amino acid 1 to amino acid 160 or with the amino acid sequence shown in Figure 6 (SEQ ID NO: 12) in the region from amino acid 1 to amino acid 160. In nature, CAL is expressed in floral meristem.

The present invention provides a nucleic acid molecule encoding a CAL, including, for example, the Arabidopsis CAL cDNA sequence shown in Figure 5 (SEQ ID

NO: 9). As disclosed herein, CAL, like AP1, contains a MADS domain and a K domain. The MADS domains of. CAL and APi differ in only five of 56 amino acid residues, where four of the five differences represent conservative amino acid replacements. Over the entire sequence, the Arabidopsis CAL and Arabidopsis APi sequences (SEQ ID
NOS: 10 and 2) are 76% identical and are 88% similar if conservative amino acid substitutions are allowed.

Similar to the expression pattern of AP1, CAL
RNA is expressed in young floral meristem in Arabidopsis.
However, in contrast to API expression, which is high throughout sepal and petal development, CAL expression is low in these organs.

LEAFY (LFY) is yet another example of a floral meristem identity gene product. As used herein, the term "LEAFY" or "LFY" means a floral meristem identity gene product that is characterized in part by having an amino acid sequence that is related to the amino acid sequence shown in Figure 9 (SEQ ID NO: 16) In nature, LFY is expressed in floral meristem as well as during vegetative development. As disclosed herein, ectopic expression of floral meristem identity gene products, which normally are expressed in floral meristem, such as AP1 or CAL or LFY or combinations thereof, in shoot meristem can convert shoot meristem to floral meristem and promote early flowering.

Flower development in Arabidopsis is recognized in the art as a model for flower development in angiosperms in general. Gene orthologs corresponding to the Arabidopsis genes involved in the early steps of flower formation have been identified in distantly related plant species, and these gene orthologs show remarkably similar RNA expression patterns. Mutations in these genes also result in phenotypes that correspond to the phenotype produced by a similar mutation in Arabidopsis. For example, orthologs of the Arabidopsis floral meristem identity genes API and LFY and the Arabidopsis organ identity genes AGAMOUS, APETALA3 and PISTILLATA have been isolated from monocots such as maize and, where characterized, reveal the anticipated RNA
expression patterns and related mutant phenotypes.
(Schmidt et al., Plant Cell 5:729-737 (7-993); and Veit et al., Plant Cell 5:1205-1215 (1993)).

Furthermore, a gene ortholog can be functionally interchangeable in that it can function across distantly related species boundaries (Mandel et al., Cell 71:133-143 (1992)).

5 Taken together, these data suggest that the underlying mechanisms controlling the initiation and proper development of flowers are conserved across distantly related dicot and monocot boundaries. Therefore, results obtained using 10 Arabidopsis can be predictive of results that can be expected in other angiosperms.

Floral meristem identity genes in particular are conserved throughout the plant kingdom. For example, a gene ortholog of Arabidopsis API has been isolated from 15 Antirrhinum majus (snapdragon; Huijser et al., EMBO J.
11:1239-1249 (1992)).

As disclosed herein, an ortholog of Arabidopsis API also has been isolated from Zea Mays (maize; see Example IA). Similarly, gene orthologs of 20 Arabidopsis LFY have been isolated from Antirrhinum majus, tobacco and poplar tree (Coen et al., Cell, 63:1311-1322 (1990); Kelly et al., Plant Cell 7:225-234 (1995); and Strauss et al., Molec. Breed 1:5-26 (1995)).

In addition, a mutatioii in the Antirrhinum AP1 ortholog results in a phenotype ::imilar to the Arabidopsis api mutant phenotype described above (Huijser et al., supra, 1992). Similarly, a mutation in the P.ntirrninum LFY
ortholog results in a phenotype similar to the Arabidopsis lfy mutant phenotype (Coen et al., supra, 1995). These studies indicate that AP1 and LFY function similarly in distantly related angiosperms.

A floral meristem identity gene product also can function across species boundaries. For example, Arabidopsis LFY can convert shoot meristem to floral meristem when expressed in aspen trees (Weigel and Nilsson, Nature 377:495-500 (1995)).

As disclosed herein, a nucleic acid molecule encoding an Arabidopsis AP1 or CAL gene product (SEQ ID NOS: 1 and 9), for example, also can be used to convert shoot meristem to floral meristem in an angiosperm. Thus, a nucleic acid molecule encoding an Arabidopsis AP1 gene product (SEQ ID NO: 1) or an Arabidopsis CAL gene product (SEQ ID NO: 9) can be introduced into an angiosperm such as corn, wheat or rice and, upon expression, can convert shoot meristem to floral meristem in the transgenic angiosperm.
Furthermore, as disclosed herein, the conserved nature of an AP1 or CAL or LFY gene among diverse angiosperms, allows a nucleic acid molecule encoding a floral meristem identity gene product from essentially any angiosperm to be introduced into essentially any other angiosperm, wherein the expression of the nucleic acid molecule in shoot meristem can convert shoot meristem to floral meristem.

If desired, a novel API, CAL or LFY seauence can be isolated from an angiosperm using a nucleotide sequence as a probe and methods well known in the art of molecular biology (Sambrook et al. (eds.), Molecular c'loning= A Laboratory Manual (Second Edition), Plainview, NY: Cold Spring Harbor Laboratory Press (198 ))=
As exemplified herein and discussed in detail below (see S Example IA), the API ortholog from Zea Mays (maize; SEQ
ID NO: 7) was isolated using the Arabidopsis API cDNA as a probe (SEQ ID NO: 1).

In one embodiment, the invention provides a non-naturally occurring angiosperm that contains an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product and that is characterized by early flowering. As used herein, the term "characterized by early flowering," when used in reference to a non-naturally occurring angiosperm of the invention, means a non-naturally occurring angiosperm that forms flowers sooner than flowers would form on a corresponding naturally occurring angiosperm that does not ectopically express a floral meristem identity gene product, grown under the same conditions. Flowering times for naturally occurring angiosperms are well known in the art and depend, in part, on genetic factors and on the environmental conditions, such as day length. Thus, given a defined set of environmental conditions, a naturally occurring plant will flower at a relatively predictable time.

It is recognized that various transgenic p_ants that are characterized by early flowering have been described. Such transgenic plants are described herein and are readily distinguishable or explicitly excluded from the present invention. For example, a product of a "late-flowering gene" can promote early flowering but does not specify the conversion of shoot meristem to floral meristem. Therefore, a transgenic plant expressing a late-flowering gene product is distinguishable from a non-naturally occurring angiosperm of the invention. For example, a transgenic plant expressing the late-flowering gene, CONSTANS (CO), flowers earlier than a corresponding wild type plant (Putterill et al., Cell 80:847-857 (1995)). However, expression of exogenous CONSTANS does not convert shoot meristem to floral meristem.

Early flowering also has been observed in a transgenic tobacco plant expressing an exogenous rice MAIDS domain gene. Although the product of this gene promotes early flowering, it does not specify the identity of floral meristem and, thus, cannot convert shoot meristem to floral meristem (Chung et al., Plant Mol. Biol. 26:657-665 (1994)). Therefore, the early-flowering CO and rice MADS domain gene transgenic plants are distinguishable from the early-flowering non-naturally occurring angiosperms of the invention.

Mutations in a class of genes known as "early-flowering genes" also result in plants that flower prematurely. Such early flowering genes include, for example, EARLY FLOWERING 1-3 (ELF1, ELF2, ELF3);
EMBRYONIC FLOWER 1,2 (EMFZ, EMF2); LONG HYPOCOTYL 1,2 (HYI, HY2 ) ; PHYTOCHROME B (PHYB ) , SPINDLY (SPY) and TERMINAL FLOWER (TFL) (Weigel, supra, 1995). However, 'WO 97/27287 PCT/US96/01041 the wild type product of an early flowering gene retards flowering and is distinguishable from a floral meristem identity gene product in that it does not promote conversion of shoot meristem to floral meristem.

An Arabidopsis plant having a mutation in the TERMINAL FLOWER (TFL) gene flowers early and is characterized by the conversion of shoots to flowers (Alvarez et al., Plant J. 2:103-116 (1992)).

However, TFL is not a floral meristem identity gene product, as defined herein.
Specifically, it is the loss of TFL that promotes conversion of shoot meristem to floral meristem. Since the function of TFL is to antagonize formation of floral meristem, a tfl mutant, which has lost this antagonist function, permits conversion of shoot meristem to floral meristem. Although TFL is not a floral meristem identity gene product and does not itself convert shoot meristem to floral meristem, the loss of TFL can result in a nlant with an ectopically expressed floral meristem identity gene product. Such tfl mutants, in which a mutation in TFL results in conversion of shoot meristem to floral meristem, are explicitly excluded from the present invention.

As used herein, the term "non-naturally occurring angiosperm" means an angiosperm that contains a genome that has been modified by man. A transgenic angiosperm, for example, contains an exogenous nucleic acid molecule and, therefore, contains a genome tha-~ has been modified by man. Furthermore, an angiosperm that contains, for example, a mutation in an endogenous floral meristem identity gene regulatory element as a result of exposure to a mutagenic agent by man also contains a genome that has been modified by man. In contrast, a plant containing a spontaneous or naturally occurring mutation is not a "non-naturally occurring angiosperm"
and, therefore, is not encompassed within the invention.

As used herein, the term "transgenic" refers to an angiosperm that contains in its genome an exogenous nucleic acid molecule, which can be derived from the same or a different species. The exogenous nucleic acid molecule that is introduced into the angiosperm can be a gene regulatory element such as a promoter or other regulatory element or can be a coding sequence, which can be linked to a heterologous gene regulatory element.

As used herein, the term "angiosperm" means a flowering plant. Angiosperms are well known and produce a variety of useful products including materials such as lumber, rubber, and paper; fibers such as cotton and linen; herbs and medicines such as quinine and vinblastine; ornamental flowers such as roses and orchids; and foodstuffs such as grains, oils, fruits and vegetables.

Angiosperms are divided into two broad classes based on the number of cotyledons, which are seed leaves that generally store or absorb food. Thus, a monocotyledonous angiosperm is an angiosperm having a single cotyledon, and a dicotyledonous angiosperm is an angiosperm having two cotyledons.

.
Angiosperms encompass a variety of flowering plants, including, for example, cereal plants, leguminous plants, oilseed plants, trees, fruit-bearing plants and ornamental flowers, which general classes are not necessarily exclusive. Such angiosperms include for example, a cereal plant, which produces an edible grain cereal. Such cereal plants include, for example, corn, rice, wheat, barley, oat, rye, orchardgrass, guinea grass, sorghum and turfgrass. In addition, a leguminous plant is an angiosperm that is a member of the pea family (Fabaceae) and produces a characteristic fruit knownas a legume. Examples of leguminous plants include, for example, soybean, pea, chickpea, moth bean, broad bean, kidney bean, lima bean, lentil, cowpea, dry bean, and peanut. Examples of legumes further also include alfalfa, birdsfoot trefoil, clover and sainfoin.
Furthermore, an oilseed plant is an angiosperm that has seeds useful as a source of oil. Examples of oilseed plants include soybean, sunflower, rapeseed and cottonseed.

A tree is an angiosperm and is a perennial woody plant, generally with a single stem (trunk).
Examples of trees include alder, ash, aspen, basswood (linden), beech, birch, cherry, cottonwood, elm, eucalyptus, hickory, locust, maple, oak, persimmon, poplar, sycamore, walnut and willows. Such trees are used for pulp, paper, and structural material, as well as providing a major source of fuel.
~

A fruit-bearing plant also is an angiosperm and produces a mature, ripened ovary (usually containing seeds) that is suitable for human or animal consumption.
Examples of fruit-bearing plants include grape, orange, lemon, grapefruit, avocado, date, peach, cherry, olive, plum, coconut, apple and pear trees and blackberry, blueberry, raspberry, strawberry, pineapple, tomato, cucumber and eggplant plants. An ornamental flower is an angiosperm cultivated for its decorative flower.
Examples of ornamental flowers include rose, orchid, lily, tulip and chrysanthemum, snapdragon, camelia, carnation and petunia. The skilled artisan will recognize that the invention can be practiced on these or other angiosperms, as desired.

In various embodiments, the present invention provides a non-naturally occurring angiosperm having an ectopically expressible first nucleic acid molecule encoding a first floral meristem identity gene product, provided the first nucleic acid molecule is not ectopically expressed due to a mutation in an endogenous TFL gene. If desired, a non-naturally occurring angiosperm of the invention can contain an ectopically expressible second nucleic acid molecule encoding a second floral meristem identity gene product, which is different from the first floral meristem identity gene product.

An ectopically expressible nucleic acid molecule can be expressed, as desired, either constitutively or inducibly. Such an ectopically =
expressible nucleic acid molecule can be an endogenous nucleic acid molecule and can contain, for example, a mutation in its endogenous gene regulatory element or can contain an exogenous, heterologous gene regulatory element that is linked to and directs expression of the endogenous nucleic acid molecule. In addition, an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product can be an exogenous nucleic acid molecule encoding a floral meristem identity gene product and containing a heterologous gene regulatory element.

The invention provides, for example, a non-naturally occurring angiosperm containing a first ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product. If desired, a non-naturally occurring angiosperm of the invention can contain a floral meristem identity gene having a modified gene regulatory element and also can contain a second ectopically expressible nucleic acid molecule encoding a second floral meristem identity gene product, provided that neither the first nor second ectopically expressible nucleic acid molecule is ectopically expressed due to a mutation in an endogenous TERMINAL FLOWER gene.

As used herein, the term "modified gene regulatory element" means a regulatory element having a mutation that results in ectopic expression in shoot meristem of the floral meristem identity gene regulated by the gene regulatory element. Such a gene regulatory element can be, for example, a promoter or enhancer element and can be positioned 5' or 3' to the coding sequence or within an intronic sequence of the floral meristem identity gene. Such a modification can be, for example, a nucleotide insertion, deletion or substitution and can be produced by chemical mutagenesis using a mutagen such as ethylmethane sulfonate (see Example IIIA}
or by insertional mutagenesis using a transposable element. For example, a modified gene regulatory element can be a functionally inactivated binding site for TFL or a gene product regulated by TFL, such that modification of the gene regulatory element results in ectopic 15. expression of the floral meristem identity gene product in shoot meristem.

The invention also provides a transgenic angiosperm containing a first exogenous gene promoter that regulates a first ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product and a second exogenous gene promoter that regulates a second ectopically expressible nucleic acid molecule encoding a second floral meristem identity gene product.

The invention also provides a transgenic angiosperm containing a first exogenous ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product and a second exogenous gene promoter that regulates a second ectopically WO 97/27287 PCTiUS96/01041 expressible nucleic acid molecule encoding a second floral meristem identity gene product, provided that the first nucleic acid molecule is not ectopically expressed due to a mutation in an endogenous TERMINAL FLOWER gene.

5 The invention also provides a transgenic angiosperm containing a first exogenous ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product and a second exogenous ectopically expressible nucleic acid molecule encoding a 10 second floral meristem identity gene product, where the first floral meristem identity gene product is different from the second floral meristem identity gene product and provided that neither nucleic acid molecule is ectopically expressed due to a mutation in an endogenous 15 TERMINAL FLOWER gene.

The ectopic expression of first and second floral meristem identity gene products can be particularly useful. For example, ectopic expression of APl and LFY in a plant promotes flowering earlier than 20 ectopic expression of AP1 alone or ectopic expression of LFY alone. Thus, plant breeding, for example, can be further accelerated, if desired.

First and second floral meristem identity gene products can be, for example, AP1 and CAL, or can be APl 25 and LFY or can be CAL and LFY. It should be recognized that where a transgenic angiosperm of the invention contains two exogenous nucleic acid molecules, the order of introducing such a first and a second nucleic acid molecule is not important for purposes of the present invention. Thus, a transgenic angiosperm of the invention having, for example, APl as the first floral meristem identity gene product and CAL as the second floral meristem identity gene product is equivalent to a transgenic angiosperm having CAL as the first floral meristem identity gene product and APl as the second floral meristem identity gene product.

The invention also provides methods of converting shoot meristem to floral meristem in an angiosperm by ectopically expressing an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product in the angiosperm. Thus, the invention provides, for example, methods of converting shoot meristem to floral meristem in an angiosperm by introducing an exogenous ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product into the angiosperm, thereby producing a transgenic angiosperm. A floral meristem identity gene product such as AP1, CAL or LFY, or a chimeric protein containing, in part, a floral meristem identity gene product (see below) is useful in the methods of the invention.

As used herein, the term "introducing," when used in reference to an angiosperm, means transferring an exogenous nucleic acid molecule into the angiosperm. For example, an exogenous nucleic acid molecule can be introduced into an angiosperm by methods such as Agrobacterium-mediated transformation or direct gene transfer methods including microprojectile-mediated transformation (Klein et al., Nature 327:70-73 (1987)).
These and other methods of introducing a nucleic acid molecule into an angiosperm are well known in the art (Bowman et al.
(ed.), Arabidogsis: An Atlas of Morphology and Development, New York: Springer (1994); Valvekens et al., Proc. Natl. Acad. Sci., USA 85:5536-5540 (1988); and Wang et al., Transformation of Plants and Soil Microorganisms, Cambridge, UK: University Press (1995))=
As used herein, the term "converting shoot meristem to floral meristem" means promoting the formation of flower progenitor tissue where shoot progenitor tissue would normally be formed. As a result of the conversion of shoot meristem to floral meristem, flowers form in an angiosperm where shoots normally would form. The conversion of shoot meristem to floral meristem can be identified using well known methods, such as scanning electron microscopy, light microscopy or visual inspection.

The invention also provides methods of converting shoot meristem to floral meristem in an angiosperm by introducing a first ectopically expressible nucleic acid molecule encoding a first floral meristem identity gene product and a second ectopically expressible nucleic acid molecule encoding a second floral meristem identity gene product into the angiosperm. As discussed above, first and second floral meristem identity gene products useful in the invention can be, for example, AP1 and CAL or AP1 and LFY or CAL
and LFY.

The invention also provides methods of promoting early flowering in an angiosperm by ectopically expressing a nucleic acid molecule encoding a floral meristem identity gene product in the angiosperm, provided that the nucleic acid molecule is not ectopically expressed due to a mutation in an endogenous TERMINAL FLOWER gene. For example, the invention provides methods of promoting early flowering in an angiosperm by introducing an ectopically expressible nucleic acid molecule encoding a floral meristem identity gene product into the angiosperm, thus producing a transgenic angiosperm. A floral meristem identity gene product such as APl, CAL or LFY, or a chimeric protein containing, in part, a floral meristem identity gene product (see below) is useful in methods of promoting early flowering.

The present invention further provides nucleic acid molecules encoding floral meristem identity gene products. For example, the invention provides a nucleic acid molecule encoding CAL, having at least about 70 percent amino acid identity with amino acids 1 to 160 of SEQ ID NO: 10 or SEQ ID NO: 11. The invention also provides a nucleic acid molecule encoding Arabidopsis thaliana CAL having the amino acid sequence shown in Figure 5 (SEQ ID NO: 10) and a nucleic acid molecule encoding Brassica oleracea CAL having the amino acid sequence shown in Figure 6 (SEQ ID NO: 12). In addition, the invention provides a nucleic acid molecule encoding Brassica oZeracea AP1 having the amino acid sequence shown in Figure 2 (SEQ ID NO: 4) and a nucleic acid molecule encoding Brassica oleracea var. botrytis AP1 having the amino acid sequence shown in Figure 3 (SEQ ID
NO: 6). The invention also provides a nucleic acid molecule encoding Zea mays AP1 having the amino acid sequence shown in Figure 4 (SEQ ID NO: 8).

As disclosed herein, CAL is highly conserved among different angiosperms. Forexample, Arabidopsis CAL (SEQ ID NO: 10) and Brassica oleracea CAL (SEQ ID NO:
12) share about 80 percent amino acid identity. In the region from amino acid 1 to amino acid 160, Arabidopsis CAL and Brassica oleracea CAL are about 89 percent identical at the amino acid level. Using a nucleotide sequence derived from a conserved region of SEQ ID NO: 9 or SEQ ID NO: 11, a nucleic acid molecule encoding a novel CAL ortholog can be isolated from other angiosperms. Using methods sizch as those described by Purugganan et al. (Genetics 40: 345-356 (1995)), one can readily confirm that the newly isolated molecule is a CAL
ortholog. Thus, a nucleic acid molecule encoding CAL, which has at least about 70 percent amino acid identity with Arabidopsis CAL (SEQ ID NO: 10) or Brassica oleracea CAL (SEQ ID NO: 12), can be isolated and identified using well known methods.

The invention also provides a nucleic acid molecule encoding a truncated CAL gene product. For example, the invention provides a nucleic acid molecule encoding the Brassica oleracea var. botrytis CAL gene product (BobCAL). BobCAL contains 150 amino acids of the approximately 255 amino acids encoded by a full-length 5 CAL cDNA (see Figure 7; SEQ ID NO: 14; see, also, Figure 8B).

The invention also provides a nucleic acid containing the Arabidopsis thaliana APi gene (Figure 10;
SEQ ID NO: 17), a nucleic acid molecule containing the 10 Brassica oleracea AP1gene (Figure 11; SEQ ID NO: 18) and a nucleic acid molecule containing the Brassica oleracea var. botrytis AP1 gene (Figure 12; SEQ ID NO: 19). In addition, the invention also provides a nucleic acid containing the Arabidopsis thaliana CAL gene (Figure 13;

15 SEQ ID NO: 20) and a nucleic acid molecule containing the Brassica oleracea CAL gene (Figure 11; SEQ ID NO: 21).

In addition, the invention provides a nucleic acid molecule containing the Brassica oleracea var. botrytis CAL gene (Figure 15; SEQ ID NO: 22).

The invention further provides a nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule encoding a CAL, or a complementary sequence thereof. In particular, such a nucleotide sequence can hybridize under relatively stringent conditions to a nucleic acid molecule encoding Arabidopsis CAL (SEQ ID NO: 9) or Brassica oleracea CAL
(SEQ ID NO: 11), or a complementary sequence thereof.
Similarly, the present invention provides a nucleotide sequencethat hybridizes under relatively stringent conditions to a nucleic acid molecule encoding Zea mays AP1 (SEQ ID NO: 7), or a complementary sequence thereof.

In general, a nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule is a single-stranded nucleic acid sequence that can range in size from about 10 nucleotides to the full-length of a gene or a cDNA. Such a nucleotide sequence can be chemically synthesized, using routine methods or can be purchased from a commercial source. Ih addition, such nucleotide sequences can be obtained by enzymatic methods such as random priming methods, the polymerase chain reaction (PCR) or by standard restriction endonuclease digestion, followed by denaturation (Sambrook et al., supra, 1989).

A nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule can be used, for example, as a primer for PCR
(Innis et al. (ed.) PCR Protocols: A Guide to Methods and A.pplications, San Diego, CA: Academic Press, Inc.

(1990)). Such a nucleotide sequence generally contains about 10 to about 50 nucleotides.

A nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule also can be used to screen a cDNA or genomic library to obtain a related nucleotide sequence. For example, a cDNA library that is prepared from rice or wheat can be screened with a nucleotide sequence derived from the Zea mays AP1 sequence in order to isolate a rice or wheat ortholog of AP1. Generally, such a nucleotide sequence contains at least about 14-16 nucleotides depending, for example, on the hybridization conditions to be used.

A nucleotide sequence derived from a nucleic acid molecule encoding Zea mays APi (SEQ ID NO: 7) also can be used to screen a Zea mays cDNA library to isolate a sequence that is related to but distinct from APi.
Furthermore, such a hybridizing nucleotide sequence can be used to analyze RNA levels or patterns of expression, as by northern blotting or by in situ hybridization to a tissue section. Such a nucleotide sequence also can be used in Southern blot analysis to evaluate gene structure and identify the presence of related gene sequences.

One skilled in the art would select a particular nucleotide sequence that hybridizes under relatively stringent conditions to a nucleic acid molecule encoding a floral meristem identity gene product based on the application for which the sequence will be used. For example, in order to isolate an ortholog of AP1, one can choose a region of AP1 that is highly conserved among known API sequences such as Arabidopsis API (SEQ ID NO: 1) and Zea mays AP1 (GenBank accession number L46400; SEQ ID NO: 7). Similarly, in order to isolate an ortholog of CAL, one can choose a region of CAL that is highly conserved among known CAL cDNAs, such as Arabidopsis CAL (SEQ ID NO: 9) and Brassica CAL (SEQ
ID NO: 11). It further would be recognized, for example, that the region encoding the MADS domain, which is common to a number of genes, can be excluded from the nucleotide sequence. In addition, one can use a full-length Arabidopsis AP1 or CAL cDNA nucleotide sequence (SEQ ID NO: 1 or SEQ ID NO: 9) to isolate an ortholog of API or CAL.

For example, the Arabidopsis API cDNA shown in Figure 1 (SEQ ID NO: 1) can be used as a probe to identify and isolate a novel APZ ortholog. Similarly, the Arabidopsis CAL cDNA shown in Figure 5 (SEQ ID NO: 9) can be used to identify and isolate a novel CAL ortholog (see Examples IA and IIIC, respectively). In order to identify related MADS domain genes, a nucleotide sequence derived from the MADS domain of API or CAL, for example, also can be useful to isolate a related gene sequence encoding this DNA-binding motif.

Hybridization utilizing a nucleotide sequence of the invention requires that hybridization be performed under relatively stringent conditions such that non-specific hybridization is minimized. Appropriate hybridization conditions can be determined empirically, or can be estimated based, for example, on the relative G+C content of the probe and the number of mismatches between the probe and target sequence, if known.
Hybridization conditions can be adjusted as desired by varying, for example, the temperature of hybridizing or the salt concentration (Sambrook, supra, 1989).

The invention also provides a vector containing a nucleic acid molecule encoding a CAL gene product. In addition, the invention provides a vector containing a nucleic acid molecule encoding the Zea mays AP1 gene product. A vector can be a cloning vector or an expression vector and provides a means to transfer an exogenous nucleic acid molecule into a host cell, which can be a prokaryotic or eukaryotic cell. Such vectors are well known and include plasmids, phage vectors and viral vectors. Various vectors and methods for introducing such vectors into a cell are described, for example, by Sambrook et al., supra, 1989, and by Glick and Thompson (eds.), Methods in Plant Molecular Bioloqy and Biotechnology, Boca Raton, FL: CRC Press (1993)).

The invention also provides an expression vector containing a nucleic acid molecule encoding a floral meristem identity gene product such as CAL, AP1 or LFY. Expression vectors are well known in the art and provide a means to transfer and express an exogenous nucleic acid molecule into a host cell. Thus, an expression vector contains, for example, transcription start and stop sites such as a TATA sequence and a poly-A
signal sequence, as well as a translation start site such as a ribosome binding site and a stop codon, if not present in the coding sequence.

An expression vector can contain, for example, a constitutive regulatory element useful for promoting expression of an exogenous nucleic acid molecule in a plant cell. The use of a constitutive regulatory element can be particularly advantageous because expression from 'WO 97/27287 PCT/US96/01041 the element is relatively independent of developmentally regulated or tissue-specific factors. For example, the cauliflower mosaic virus 35S promoter (CaMV35S) is a well-characterized constitutive regulatory element that 5 produces a high level of expression in all plant tissues (Odell et al., Nature 313:810-812 (198-5)).

The CaMV35S promoter is particularly useful because it is active in numerous different angiosperms (Benfey and Chua, Science 10 250:959-966 (1990);

Odell et al., supra, 1985). Other constitutive regulatory elements useful for expression in an angiosperm include, for example, the nopaline synthase (nos) gene promoter (An, Plant Physiol. 81:86 (1986)).

In addition, an expression vector of the invention can contain a regulated gene regulatory element such as a promoter or enhancer element. A particularly useful regulated promoter is a tissue-specific promoter such as the shoot meristem-specific CDC2 promoter (Hemerly et al., Plant Cell 5:1711-1723 (1993)), or the AGLB promoter, which is active in the apical shoot meristem immediately after the transition to f:owering (Mandel and Yanofsky, Plant Cell 7:1763-1771 (1995)).

An expression vector of the invention also can contain an inducible regulatory element, which has conditional activity dependent upon the presence of a particular regulatory factor. Useful inducible regulatory elements include, for example, a heat-shock promoter (Ainley and Key, Plant Mol. Biol. 14:949 (1990)) or a nitrate-inducible promoter derived from the spinach nitrite reductase gene (Back et al., Plant Mol.

Bi . 17:9 (1991)).

A hormone-inducible element (Yamaguchi-Shinozaki et al., Plant Mol. Biol. 15:905 (1990) and Kares et al., Plant Mol. Biol. 15:225 (1990)) or a light-inducible promoter, such as that associated with the small subunit of RuBP carboxylase or the LHCP gene families (Feinbaum et al., MQl. Gen. Genet. 226:449 (1991) and Lam and Chua, Science 248:471 (1990)) also can be useful in an expression vectcr of the invention. A human glucocorticoid response element also can be used to achieve steroid hormone-dependent gene expression in plants (Schena et al., Proc. Natl. Acad. Sci. USA
88:10421 (1992.)).

An appropriate gene regulatory element such as a promotor is selected depending on the desired pattern or level of expression of a nucleic acid molecule linked thereto. For example, a constitutive promoter, which is active in all tissues, would be appropriate to express a desired gene product in all cells containing the vector.
In addition, it can be desirable to restrict expression of a nucleic acid molecule to a particular tissue or during a particular stage of development. A
developmentally regulated or tissue-specific expression can be useful for this purpose and can avoid potential undesirable side-effects that can accompany unregulated expression. Inducible expression also can be particularly useful to manipulate the timing of gene expression such that, for example, a population of transgenic angiosperms of the invention that contain an expression vector comprising a floral meristem identity gene linked to an inducible promoter can be induced to flower essentially at the same time. Such timing of flowering can be useful, for example, for manipulating the time of crop harvest.

The invention also provides a kit containing an expression vector having a nucleic acid molecule encoding a floral meristem identity gene product. Such a kit is useful for converting shoot meristem to floral meristem in an angiosperm or for promoting early flowering in an angiosperm. If desired, such a kit can contain appropriate reagents, which can allow relatively high efficiency of transformation of an angiosperm with the vector. Furthermore, a control plasmid lacking the floral meristem identity gene can be included in the kit to determine, for example, the efficiency of transformation.

The invention further provides a host cell containing a vector comprising a nucleic acid molecule encoding CAL. A host cell can be prokaryotic or eukaryotic and can be, for example, a bacterial cell, WO 97/27287 PCT/iJS96/01041 yeast cell, insect cell, xenopus cell, mammalian cell or plant cell.

The invention also provides a transgenic garden } variety cauliflower plant containing an exogenous nucleic acid molecule selected from the group consisting of a nucleic acid molecule encoding a CAL gene product and a nucleic acid molecule encoding an APi gene product. Such a transgenic cauliflower plant can produce an edible flower in place of the typical cauliflower vegetable.

A nucleic acid encoding CAL has been isolated from a Brassica oleracea line that produces wild-type flowers (BoCAL) and from the common garden variety of cauliflower, Brassica oleracea var. botrytis (BobCAL) which lacks flowers. The Brassica oleracea CAL cDNA (SEQ

ID NO: 10) is highly similar to the Arabidopsis CAL cDNA
(SEQ ID NO: 12; and see Figure 8). In contrast, the Brassica oleracea var. botrytis CAL cDNA contains a stop codon, predicting that the BobCAL protein will be truncated after amino acid 150 (SEQ ID NO: 14 and see Figure 8). The correlation of full-length Arabidopsis and Brassica oleracea CAL gene products with a flowering phenotype indicates that transformation of non-flowering garden varieties of cauliflower such as Brassica oleracea var. botrytis with a full-length CAL cDNA can induce flowering in the transgenic cauliflower plant.

As used herein, the term "CAL gene product"
means a full-length CAL gene product that does not terminate substantially before amino acid 255 and that, when ectopically expressed in shoot meristem, converts shoot meristem to floral meristem. A nucleic acid molecule encoding a CAULIFLOWER gene product can be, for example, a nucleic acid molecule encoding Arabidopsis CAL

shown in Figure 5 (SEQ ID NO: 9) or a nucleic acid molecule encoding Brassica oleracea CAL shown in Figure 6 (SEQ ID NO: 11). In comparison, a nucleic acid molecule encoding a truncated CAL gene product that terminates substantially before amino acid 255, such as the encoded truncated BobCAL gene product (SEQ ID NO: 13), is not a nucleic acid molecule encoding a CAL gene product as defined herein. Furthermore, ectopic expression of BobCAL in an angiosperm does not result in conversion of shoot meristem to floral meristem.

As used herein, the term "APi gene product"
means a full-length API gene product that does not terminate substantially before amino acid 256. A nucleic acid molecule encoding an APi gene product can be, for example, a nucleic acid molecule encoding Arabidopsis AP1 shown in Figure 1 (SEQ ID NO: 1), Brassica oleracea API
shown in Figure 2, (SEQ ID NO: 3), Brassica oleracea var.
botrytis API shown in Figure 3 (SEQ ID NO: 5) or Zea mays API shown in Figure 4 (SEQ ID NO: 7).

The invention provides a CAL polypeptide having at least about 70 percent amino acid identity with amino acids 1 to 160 of SEQ ID NO: 10 or SEQ ID NO: 12. For example, the Arabidopsis thaliana CAL polypeptide, having the amino acid sequence shown as amino acids 1 to 255 in Figure 5(SEQ ID NO: 10), and the Brassica oleracea CAL

polypeptide, having the amino acid sequence shown as amino acids 1 to 255 in Figure 6 (SEQ ID NO: 12) are provided by the invention.

The invention also provides the truncated 5 Brassica oleracea var. botrytis CAL polypeptide having the amino acid sequence shown as amino acids 1 to 150 in Figure 7 (SEQ ID NO: 14). The BobCAL polypeptide can be useful as an immunogen to produce an antibody that specifically binds the truncated BoCAL polypeptide, but 10 does not bind a full length CAL gene product. Such an antibody can be useful to distinguish between a full length CAL and truncated CAL.

The invention provides also provides a Zea mays AP1 polypeptide. As used herein, the term "polypeptide"
15 is used in its broadest sense to include proteins, polypeptides and peptides, which are related in that each consists of a sequence of amino acids joined by peptide bonds. For convenience, the terms "polypeptide,"
"protein" and "gene product" are used interchangeably.
20 While no specific attempt is made to distinguish the size limitations of a protein and a peptide, one skilled in the art would understand that proteins generally consist of at least about 50 to 100 amino acids and that peptides generally consist of at least two amino acids up to a few 25 dozen amino acids. The term polypeptide is used generally herein to include any such amino acid sequence.
The term polypeptide also includes an active fragment of a floral meristem identity gene product. As used herein, the term "active fragment," means a polypeptide portion of a floral meristem identity gene product that can convert shoot meristem to floral meristem or can provide early flowering. For example, an active fragment of a CAL polypeptide can consist of an amino acid sequence derived from a CAL protein as shown in Figure 5 or 6 (SEQ ID NOS: 10 and 12) and that has an activity of a CAL. An active fragment can be, for example, an amino terminal or carboxyl terminal truncated form of Arabidopsis thaliana CAL or Brassica oleracea CAL
(SEQ ID NOS: 10 or 12, respectively). Such anactive fragment can be produced using well known recombinant DNA
methods (Sambrook et al., supra, 1989). The product of the BobCAL gene, which is truncated at amino acid 150, lacks activity in converting shoot meristem to floral meristem and, therefore, is an example of a polypeptide portion of a CAL floral meristem identity gene product that is not an "active fragment."

An active fragment of a floral meristem identity gene product can convert shoot meristem to floral meristem and is readily identified using the methods described in Example II, below). Briefly, Arabidopsis can be transformed with a nucleic acid molecule encoding a portion of a-floral meristem identity gene product, in order to determine whether the fragment can convert shoot meristem to floral meristem or promote early flowering and, therefore, has an activity of a floral meristem identity gene product.

The invention further provides an antibody that specifically binds a CAL polypeptide, an antibody that specifically binds the truncated Brassica oleracea var.
botrytis CAL polypeptide, and an antibody that specifically binds the Zea mays AP1 polypeptide. As used herein, the term "antibody" is used in its broadest sense to include polyclonal and monoclonal antibodies, as well as polypeptide fragments of antibodies that retain a specific binding activity for CAL protein of at least about 1 x 105 M-1. One skilled in the art would know that anti-CAL antibody fragments such as Fab, F(ab')z and Fv fragments can retain specific binding activity for CAL
and, thus, are included within the definition of an antibody. In addition, the term "antibody" as used herein includes naturally occurring antibodies as well as non-naturally occurring antibodies and fragments that have binding activity such as chimeric antibodies or humanized antibodies. Such non-naturally occurring antibodies can be constructed using solid phase peptide synthesis, produced recombinantly or obtained, for example, by screening combinatorial libraries consisting of variable heavy chains and variable light chains as described by Huse et al., Science 246:1275-1281 (1989)).

An antibody "specific for" a polypeptide, or that "specifically binds" a polypeptide, binds with substantially higher affinity to that polypeptide than to an unrelated polypeptide. An antibody specific for a polypeptide also can have specificity for a related polypeptide. For example, an antibody specific for Arabidopsis CAL also can have specificity for Brassica oleracea CAL.

An anti-CAL antibody, for example, can be prepared using a CAL fusion protein or a synthetic peptide encoding a portion of Arabidopsis CAL or of Brassica oleracea CAL as an immunogen. One skilled in the art would know that purified CAL protein, which can be prepared from natural sources or produced recombinantly, or fragments of CAL, including a peptide portion of CAL such as a synthetic peptide, can be used as an immunogen. Non-immunogenic fragments or synthetic peptides of CAL can be made immunogenic by coupling the hapten to a carrier molecule such as bovine serum albumin (BSA) or keyhole limpet hemocyanin (KLH). In addition, various other carrier molecules and methods for coupling a hapten to a carrier molecule are well known in the art and described, for example, by Harlow and Lane, Antibodies; A laboratory manual (Cold Spring Harbor Laboratory Press, 1988).

An antibody that specifically binds the truncated Bob CAL polypeptide or an antibody that specifically binds the Zea mays APl polypeptide similarly can be produced using such methods. An antibody that specifically binds the truncated Brassica oleracea var.

botrvtis CAL polypeptide can be particularly useful to distinguish between full-length CAL polypeptide and truncated CAL polypeptide.

WO 97/27287 PCT[US96/01041 The invention provides a method of ideritifying a Brassica having a modified CAL allele by detecting a polymorphism associated with a CAL locus, where the CAL
locus comprises a modified CAL allele that does not encode an active CAL gene product. Such a method is useful for the genetic improvement of Brassica plants, a genus of_great economic value.

Brassica plants are a highly diverse group of crop plants useful as vegetables and as sources of condiment mustard, edible and industrial oil, animal fodder and green manure. Brassica crops encompass a variety of well known vegetables including cabbage, cauliflower, broccoli, collard, kale, mustard greens, Chinese cabbage and turnip, which can be interbred for crop improvement (see, for example, King, Euphytica 50:97-112 (1990) and Crisp and Tapsell, Genetic improvement of vecretable crops pp. 157-178 (199:)).

Breeding of Brassica crops is useful, for example, for improving the quality and early development of vegetables. In addition, such breeding can be useful to increase disease resistance, such as resistance, of a Brassica to clubroot disease or mildew; viral resistance, such as resistance to turnip mosaic virus and cauliflower mosaic virus; or pest resistance (King, supra, 1990).

The use of polymorphic molecular markers in the breeding of Brassicae is well recognized in the art (Crisp and Tapsell, supra, 1993). Identification of a polymorphic molecular marker that is associated with a desirable trait can vastly accelerate the time required to breed the desirable trait into a new Brassica species or variant. In particular, since many rounds of 5 backcrossing are required to breed a new trait into a different genetic background, early detection of a desirable trait by molecular methods can be performed prior to the time a plant is fully mature, thus accelerating the rate of crop breeding (see, for example, 10 Figidore et al., Euphytica 69: 33-44 (1993)).

A polymorphism associated with a CAL locus comprising a modified CAL allele that does not encode an active CAL gene product, is disclosed herein. Figure 6 15 shows the nucleotide (SEQ ID NO: 11) and amino acid (SEQ
ID NO: 12) sequence of Brassica oleracea CAL (BoCAL), and Figure 7 shows the nucleotide (SEQ ID NO: 13) and amino acid (SEQ ID NO: 14) sequence of Brassica oleracea var.
botrytis CAL (BobCAL). At amino acid 150, which is 20 glutamic acid (Glu) in BoCAL, a stop codon is present in BobCAL. This polymorphism results in a truncated BobCAL
gene product that is not active as a floral meristem identity gene product. The BoCAL nucleic acid sequence (ACGAGT) can be readily distinguished from the BobCAL

25 nucleic acid sequence (ACTAGT) using well known molecular methods. For example, the polymorphic ACTAGT BobCAL
sequence is recognized by a SpeI restriction endonuclease site, whereas the ACGAGT BoCAL sequence is not recognized by SpeI. Thus, a restriction fragment length 30 polymorphism (RFLP) in BobCAL provides a simple means for identifying a modified CAL allele (BobCAL) and, therefore, can serve as a marker to predict the inheritance of the "cauliflower" phenotype.

A modified CAL allele encoding a truncated CAL
gene product also can serve as a marker to predict the "cauliflower" phenotype in other cauliflower variants.
For example, nine romanesco variants of Brassica oleracea var. botrytis, which each have the "cauliflower"
phenotype, were examined for the presence of a stop codon at position 151 of the CAL coding sequence. All nine of the romanesco variants contained the SpeI site that indicates a stop codon and, thus, a truncated CAL gene product. In contrast, Brassica oleracea variants that lack the "cauliflower" phenotype (broccoli and brussels sprouts) were examined for the SpeI site. In every case, the broccoli and brussel sprout variants had a full-length CAL coding sequence, as indicated by the absence of the distinguishing Spel site. Thus, a truncated CAL gene product can be involved in the "cauliflower phenotype" in numerous different Brassica variants.

As used herein, the term "modified CAL allele"
means a CAL allele that does not encode a CAL gene product active in converting shoot meristem to floral meristem. A modified CAL allele can have a modification within a gene regulatory element such that a CAL gene product is not produced. In addition, a modified CAL
allele can have a modification such as a mutation, deletion or insertion in a CAL coding sequence which results in an inactive CAL gene product. For example, an inactive CAL gene product can result from a mutation creating a stop codon, such that a truncated, inactive CAL gene product lacking the ability to convert shoot meristem to floral meristem is produced.

As used herein, the term "associated" means closely linked and describes the tendency of'two genetic loci to be inherited together as a result of their proximity. If two genetic loci are associated and are polymorphic, one locus can serve as a marker for the inheritance of the second locus. Thus, a polymorphism associated with a CAL locus comprising a modified CAL
allele can serve as a marker for inheritance of the modified CAL allele. An associated polymorphism can be located in proximity to a CAL gene or can be located within a CAL gene.

A polymorphism in a nucleic acid sequence can be detected by a variety of methods. For example, if the polymorphism occurs in a particular restriction endonuclease site, the polymorphism can be detected by a difference in restriction fragment length observed following restriction with the particular restriction endonuclease and hybridization with a nucleotide sequence that is complementary to a nucleic acid sequence including a polymorphism.

The use of restriction fragment length polymorphism as an aid to breeding Brassicae is well known in the art (see, for example, Slocum et al., Theor.

Appl. Genet. 80:57-64 (1990); Kennard et al., Theor, Appl. Genet. 87:721-732 (1994); and Figidore et al., supra, 1993)).

A restriction endonuclease such as SpeI, which is useful for identifying the presence of a BobCAL
allele in an angiosperm, is readily available and can be purchased from a commercial source. Furthermore, a nucleotide sequence that is complementary to a nucleic acid sequence having a polymorphism associated with a CAL

locus comprising a modified CAL allele can be derived, for example, from the nucleic acid molecule encoding Brassica oleracea var. botrytis CAL shown in Figure 7 (SEQ ID NO: 13) or from the nucleic acid molecule encoding Brassica oleracea CAL shown in Figure 6 (SEQ ID
NO: 11).

In some cases, a polymorphism is not distinguishable by a RFLP, but nevertheless can be used to identify a Brassica having a modified CAL allele. For example, the polymerase chain reaction (PCR) can be used to detect a polymorphism associated with a CAL locus comprising a modified CAL allele. Specifically, a polymorphic region of a modified allele can be selectively amplified by using a primer that matches the nucleotide sequence of one allele of a polymorphic locus, but does not match the sequence of the second allele (Sobral and Honeycutt, The Polymerase Chain Reaction, pp.
304-319 (199,1)).

Other well-known approaches for analyzing a polymorphism using PCR include discriminant hybridization of PCR-amplified DNA to allele-specific oligonucleotides and denaturing gradient gel electrophoresis (see Innis et al., supra, 1990).

The invention further provides a nucleic acid molecule encoding a chimeric protein, comprising a nucleic acid molecule encoding a floral meristem identity gene product such as AP1, LFY or CAL operably linked to a nucleic acid molecule encoding a ligand binding domain.
Expression of a chimeric protein of the invention in an angiosperm is particularly useful because the ligand binding domain confers regulatable activity on a gene product such as a floral meristem identity gene product to which it is fused. Specifically, the floral meristem identity gene product component of the chimeric protein is inactive in the absence of the particular ligand, whereas, in the presence of ligand, the ligand binds the ligand binding domain, resulting in floral meristem identity gene product activity.

A nucleic acid molecule encoding a chimeric protein of the invention contains a nucleic acid molecule encoding a floral meristem identity gene product, such as a nucleic acid molecule encoding the amino acid sequence shown in Figure 1(SEQ ID NO: 2), in Figure 5(SEQ ID NO:
10), or in Figure 9 (SEQ ID NO: 10), either of which is operably linked to a nucleic acid molecule encoding a ligand binding domain. The expression of such a nucleic acid molecule results in the production of a chimeric protein comprising a floral meristem identity gene product fused to a ligand binding domain. Thus, the invention also provides a chimeric protein comprising a floral meristem identity gene product fused to a ligand binding domain.

A ligand binding domain useful in a chimeric protein of the invention can be a steroid binding domain 5 such as the ligand binding domain of a glucocorticoid receptor, estrogen receptor, progesterone receptor, androgen receptor, thyroid receptor, vitamin D receptor or retinoic acid receptor. A particularly useful ligand binding domain is a glucocorticoid receptor ligand 10 binding domain, encompassed, for example, within amino acids 512 to 795 of the rat glucocorticoid receptor as shown in Figure 16 (SEQ ID NO: 24; Miesfeld et al., Cell 46:389-399 (1986)).

15 A chimeric protein containing a ligand binding domain, such as the rat glucocorticoid receptor ligand binding domain, confers glucocorticoid-dependent activity on the chimeric protein. For example, the activity of chimeric proteins consisting of adenovirus ElA, c-myc, 20 c-fos, the HIV-1 Rev transactivator, MyoD or maize regulatory factor R fused to the rat glucocorticoid receptor ligand binding domain is regulated by glucocorticoid hormone (Eilers et al., Nature 340:66 (1989); Superti-Furga et al., Proc. Natl. Acad. Sci., 25 U.S.A. 88:5114 (1991); Hope et al., Proc. Natl. Acad.
Sc=., U.S.A. 87:7787 (1990); Hollenbera et al., Proc.
Nati Acad Sci U S A 90:8028 ,199;)).

Such a chimeric protein also can be regulated in plants. For example, a chimeric protein containing a heterologous protein fused to a rat glucocorticoid receptor ligand binding domain (amino acids 512 to 795) was expressed under the control of the constitutive cauliflower mosaic virus 35S promoter in Arabidopsis.
The activity of the chimeric protein was inducible; the chimeric protein was inactive in the absence of ligand, and became active upon treatment of transformed plants with a synthetic glucocorticoid, dexamethasone (Lloyd et al., Science 266:436-439 (1994)).

As disclosed herein, a ligand binding domain fused to a floral meristem identity gene product can confer ligand inducibility on the activity of a fused floral meristem identity gene product in plants such that, upon exposure to a particular ligand, the floral meristem identity gene product is active.

Methods for constructing a nucleic acid molecule encoding a chimeric protein are routine and well known in the art (Sambrook et al., supra, 1989). For example, the skilled artisan would recognize that a stop codon in the 5' nucleic acid molecule must be removed and that the two nucleic acid molecules must be linked such that the reading frame of the 3' nucleic acid molecule is preserved. Methods of transforming plants with nucleic acid molecules also are well known in the art (see, for example, Mohoney et al., U.S. patent number 5,463,174, and Barry et al., U.S. patent number 5,463,17`=).

As used herein, the term "operably linked,"
when used in reference to two nucleic acid molecules " comprising a nucleic acid molecule encoding a chimeric protein, means that the two nucleic acid molecules are linked in frame such that a full-lengthchimeric protein can be expressed. In particular, the 5' nucleic acid molecule, which encodes the amino-terminal portion of the chimeric protein, must be linked to the 3" nucleic acid molecule, which encodes the carboxyl-terminal portion of the chimeric protein, such that the carboxyl-terminal portion of the chimeric protein is produced in the correct reading frame.

The invention further provides a transgenic angiosperm containing a nucleic acid molecule encoding a chimeric protein, comprising a nucleic acid molecule encoding a floral meristem identity gene product such as AP1, CAL or LFY linked to a nucleic acid molecule encoding a ligand binding domain. Such a transgenic angiosperm is particularly useful because the angiosperm can be induced to flower by contacting the angiosperm with a ligand that binds the ligand binding domain.
Thus, the invention provides a method of promoting early flowering or of converting shoot meristem to floral meristem in a transgenic angiosperm containing a nucleic 25- acid molecule encoding a chimeric protein of the invention, comprising expressing the nucleic acid molecule encoding the chimeric protein in the angiosperm, and contacting the angiosperm with a ligand that binds the ligand binding domain, wherein binding of the ligand to the ligand binding domain activates the floral meristem identity gene product. In particular, the invention provides methods of promoting early flowering or of converting shoot meristem to floral meristem in a transgenic angiosperm containing a nucleic acid molecule encoding a chimeric protein that consists of a nucleic acid molecule encoding APi or CAL or LFY linked to a nucleic acid molecule encoding a glucocorticoid receptor ligand binding domain by contacting the transgenic angiosperm with a glucocorticoid such as dexamethasone.

As used herein, the term "ligand" means a naturally occurring or synthetic chemical or biological molecule such as a simple or complex organic molecule, a peptide, a protein or an oligonucleotide that specifically binds a ligand binding domain. A ligand of the invention can be used, alone, in solution or can be used in conjunction with an acceptable carrier that can serve to stabilize the ligand or promote absorption of the ligand by an angiosperm.

One skilled in the art can readily determine the optimum concentration of ligand needed to bind a ligand binding domain and render a floral meristem identity gene product active. Generally, a concentration of about 1 nM to l M dexamethasone is useful for activating floral meristem identity gene product activity in a chimeric protein comprising a floral meristem identity gene product and a glucocorticoid receptor ligand binding domain (Lloyd et al., supra, 1994).

A transgenic angiosperm expressing a chimeric protein of the invention can be contacted with ligand in a variety of manners including, for example, by spraying, injecting or immersing the angiosperm. Further, a plant may be contacted with a ligand by adding the ligand to the plant's water supply or to the soil, whereby the ligand is absorbed into the angiosperm.

The following examples are intended to illustrate but not limit the present invention.
EXAMPLE I

Identification and characterization of the Zea mays APETALA1 cDNA

This example describes the isolation and characterization of the Zea mays ZAP-i "gene", which is an ortholog of the Arabidopsis floral meristem identity gene, AP1.

A. Identification and characterization of a nucleic acid sequence encoding ZAP-I

The utility of using a cloned floral homeotic gene from Arabidopsis to identify the putative ortholog in maize has previously been demonstrated (Schmidt et al., supra, (1993)).

As described in Mena et al. (Plant J.
8(6):845-854 (1995)), the maize ortholog of the Arabidopsis AP1 floral meristem identity gene, was isolated by screening a Zea mays ear cDNA library using ,WO 97/27287 PCT/US96/01041 the Arabidopsis AP1 cDNA (SEQ ID NO: 1) as a probe. A
cDNA library was prepared from wild-type immature ears as described by Schmidt et al., supra, 1993, using an Arabidopsis AP1 cDNA sequence as a probe. The 5 Arabidopsis AP1 cDNA (SEQ ID NO: 1), which is shown in Figure 1 (SEQ ID NO 1), was used as the probe.
Low-stringency hybridizations with the AP1 probe were conducted as described previously for the isolation of ZAG1 using the AG cDNA as a probe (Schmidt et al., supra, 10 1993). Positive plaques were isolated and cDNAs were recovered in Bluescript by in vivo excision.
Double-stranded sequencing was performed using the Sequenase Version 2.0 kit (U.S. Biochemical, Cleveland, Ohio) according to the manufacturer's protocol.

15 The cDNA sequence and deduced amino acid sequence for ZAP1 are shown in Figure 4 (SEQ ID NOS: 7 and 8). The deduced amino acid sequence for ZAP1 shares 89o identity with Arabidopsis AP1 through the MADS domain (amino acids 1 to 57) and 70% identity through the first 20 160 amino acids, which includes the K domain. The high level of amino acid sequence identity between ZAP1 and AP1 (SEQ ID NOS: 8 and 2), as well as the expression pattern of ZAP1 in maize florets (see below), indicates that ZAP1 is the maize ortholog of Arabidopsis AP1.

25 B, RNA expression pattern of ZAP1 Total RNA was isolated from different maize tissues as described by Cone et al., Proc. Natl, Acad.
Sci., USA 83:9631-9635 (1986).

RNA was prepared from ears or tassels at early developing stages (approximately 2 cm in size), husk leaves from developing ear shoots, shoots and roots of germinated seedlings, leaves from 2 to 3 week old plants and endosperm, and embryos at 18 days after pollination. Mature floral organs were dissected from ears at the time of silk emergence or from tassels at several days pre-emergence. To study expression patterns in the mature female flower, carpels were isolated and the remaining sterile organs were pooled and analyzed together. In the same way, stamens were dissected and collected from male florets and the remaining organs (excluding the glumes) were pooled as one sample.

RNA concentration and purity was determined by absorbance at 260/280 nM, and equal amounts (10 /.cg) were fractionated on formaldehyde-agarose gels. Gels were stained in a solution of 0.125 g ml-1 acridine orange to confirm the integrity of the RNA samples and the uniformity of gel loading, then RNA was blotted on to Hybond-W) membranes (Amersham International, Arlington Heights, Illinois) according to the manufacturer's instructions. Prehybridization and hybridization solutions were prepared as previously described (Schmidt et al., Science 238:960-963 (1987)).

The probe for ZAP1 RNA expression studies was a 445 bp SacI-NsiI fragment from the 3' end of the cDNA. Southern blot analyses were conducted to establish conditions for specific hybridization of this probe. No cross-hybridization was detected with hybridization at 60 C in 50o formamide and washes at 65 C
in 0.1x SSC and 0.5% SDS.

The strong sequence similarity between ZAP1 and API indicated that ZAP1 was the ortholog of this Arabidopsis floral meristem identity gene. As a first approximation of whether the pattern of ZAP1 expression paralleled that of AP1, a blot of total RNA from vegetative and reproductive organs was hybridized with a gene-specific fragment of the ZAP1 cDNA (nucleotides 370 to 820 of SEQ ID NO: 7). ZAP1 RNA was detected only in male and female inflorescences and in the husk leaves that surround the developing ear. No ZAP1 RNA expression was detectable in RNA isolated from root, shoot, leaf, endosperm, or embryo tissue. The restriction of ZAP1 expression to terminal and axillary inflorescences is consistent with ZAP1 being the Arabidopsis AP1 ortholog.
Male and female florets were isolated from mature inflorescences, and the reproductive organs were separated from the remainder of the floret. RNA was isolated from the reproductive and the sterile portions of the florets. ZAP1 RNA expression was not detected in maize stamens or carpels, whereas high levels of ZAP1 RNA were present in developing ear and tassel florets from which the stamens and carpels had been removed.
Thus, the exclusion of ZAP1 expression in stamens and carpels and its inclusion in the RNA of the non-reproductive portions of the floret (lodicules, lemma and palea) is similar to the pattern of expression of API
in flowers of Arabidopsis.

EXAMPLE II

Conversion of shoot meristem to floral meristem in an APETALA1 transgenic plant This example describes methods for producing a transgenic Arabidopsis plant, in which shoot meristem is converted to floral meristem.

A. Ectopic expression of APETALA1 converts inflorescence shoots into flowers Transgenic plants that constitutively express AP1 from the cauliflower mosaic virus 35S (CaMV35S) promoter were produced to determine whether ectopic AP1 expression could convert shoot meristem to floral meristem. The AP1 coding sequence was placed under control of the cauliflower mosaic virus 35S promoter (Odell et al., supra, 1985) as follows. BamHI linkers were ligated to the HincIl site of the full-length AP1 complementary DNA (Mandel et al., supra, (1992)) in pAM116, and the resulting BamHI fragment was fused to the cauliflower mosaic virus 35S promoter (Jack et al., Cell 76:703-716 (1994)) in pCGN18 to create pAM563.

Transgenic AP1 Arabidopsis plants of the Columbia ecotype were generated by selecting kanamycin-resistant plants after Agrobacteriurn-mediated plant transformation using the in planta method (Bechtold et al., C.R. Acad. Sci. Paris 316:1194-1199 (1993)).

. All analyses were performed in subsequent generations. Approximately 120 independent transgenic lines that displayed the described phenotypes were obtained.

Remarkably, in 35S-AP1 transgenic plants, the normally indeterminate shoot apex ) prematurely terminated as a floral meristem and formed a terminal flower. In addition, all lateral meristems that normally would produce inflorescence shoots also were converted into solitary flowers. These results demonstrate that ectopic expression of AP1 in shoot meristem is sufficient to convert shoot meristem to floral meristem, even though AP1 normally is not absolutely required to specify floral meristem identity.

B LEAFY is not required for the conversion of inflorescence shoots to flowers in an APETALAI
transcrenic -plant To determine whether the 35S-AP1 transgene causes ectopic LFY activity, and whether ectopic LFY
activity is required for the conversion of shoot meristem to floral meristem, the 35S-AP1 transgene was introduced into Arabidopsis 1fy mutants. The 35S-AP1 transgene was crossed into the strong lfy-6 mutant background and the F, progeny were analyzed.

Lfy mutant plants containing the 35S-API
transgene displayed the same conversion of apical and lateral shoot meristem to floral meristem as was observed in transgenics containing wild type LFY. However, the 5 resulting flowers had the typical lfy mutant phenotype, in which floral organs developed as sepaloid and carpelloid structures, with an absence of petals and stamens. These results demonstrate that LFY is not required for the conversion of shoot meristem to floral 10 meristem in a transgenic angiosperm that ectopically expresses APl.

C. APETALAI is not sufficient to specify organ fate As well as being involved in the early step of specifying floral meristem identity, API also is involved 15 in specifying sepal and petal identity at a later stage in flower development. Although API RNA is initially expressed throughout the young flower primordium, it is later excluded from stamen and carpel primordia (Mandel et al., Nature 360:273-277 (1992)). Since the 20 cauliflower mosaic virus 35S promoter is active in all floral organs, 35S-API transgenic plants are likely to ectopically express AP1 in stamens and carpels. However, 35S-API transgenic plants had normal stamens and carpels, indicating that AP1 is not sufficient to specify sepal 25 and petal organ fate.

D, Ectopic expression of APETALA1 causes early flowering In addition to its ability to alter inflorescence meristem identity, ectopic expression of API also influences the vegetative phase of plant growth.

Wild-type plants have a vegetative phase during which a basal rosette of leaves is produced, followed by the transition to reproductive growth. The transition from vegetative to reproductive growth was measured both in terms of the number of days post-germination until the first visible flowers were observed, and by counting the number of leaves. Under continuous light, wild-type and 35S-AP1 transgenic plants flowered after producing 9.88 1.45 and 4.16 0.97 leaves, respectively. Under short-day growth conditions (8 hours light, 16 hours dark, 24 C), wild-type and 35S-AP1 transgenic plants flowered after producing 52.42 3.47 and 7.4 1.18 leaves, respectively.
In summary, under continuous light growth conditions, flowers appear on wild-type Arabidopsis plants after approximately 18 days, whereas the 35S-AP1 transgenic plants flowered after an average of only 10 days. Furthermore, under short-day growth conditions, flowering is delayed in wild-type plants until_ approximately 10 weeks after germination, whereas, 35S-AP1 transgenic plants flowered in less than 3 weeks.

Thus, ectopic AP1 activity significantly reduced the time to flowering and reduced the delay of flowering caused by short day growth conditions.

EXAMPLE III
zsolation and characterization of the Arabidopsis and Brassica oleracea CAULIFLOWER aenes ' This example describes methods for isolating and characterizing the Arabidopsis and Brassica oleracea CAL genes.

A. Isolation of the ArabidoDsis and Brassica oleracea [, AUL TFLOWER crenes Genetic evidence that CAL and APl proteins may be functionallyrelated indicated that these proteins may share similar DNA sequences. In addition, DNA blot hybridization revealed that the Arabidopsis genome contains a gene that is closely related to APi. The CAL
gene, which is closely related to APi, was isolated and identified as a member of the family of Arabidopsis MADS
domain genes known as the AGAMOUS-like (AGL) genes.
Hybridization with an AP1 probe was used to isolate a 4.8-kb Eco RI genomic fragment of CAL. The corresponding CAL complementary DNA (pBS85) was cloned by reverse transcription-polymerase chain reaction (RT-PCR) with the oligonucleotides AGLIO-1 (5'-GATCGTCGTTATCTCTCTTG-3'; SEQ ID NO: 25) and AGL10-12 (5'-GTAGTCTATTCAAGCGGCG-3'; SEQ ID NO: 26).

The Arabidopsis CAL cDNA encodes a putative 255 amino acid protein (Figure 5; SEQ ID NO: 10) having a calculated molecular weight of 30.1 kD and an isoelectric point of 8.78. The deduced amino acid sequence for CAL
contains a MADS domain which generally is present in a class of transcription factors. The MADS domains of CAL
and AP1 were markedly similar, differing in only 5 of 56 amino acid residues, 4 of which represent conservative replacements. Overall, the putative CAL protein is 76%
identical to AP1; with allowance for conservative amino acid substitutions, the two proteins are 88o similar.
These results indicate that CAL and APl may recognize similar target sequences and regulate many of the same genes involved in floral meristems identity.

CAL was mapped to the approximate location of the loci identified by classical genetic means for the cauliflower phenotype (Bowman et al., Development 119:721 (1993)).
Restriction fragment length polymorphism (RFLP) mapping filters were scored and the results analyzed with the Macintosh version of the Mapmaker program as described by Rieter et al., (Proc. Natl. Acad. Sci., USA, 89:1477 (1992)). The results localized CAL to the upper arm of chromosome 1, near marker X235.

A genomic fragment spanning the CAL gene was used to transform cal-1 apl-i plants. A 5850-bp Bam HI
fragment containing the entire coding region of the Arabidopsis CAL gene as well as 1860 bp upstream of the putative translational start site was inserted into the pBIN19 plant transformation vector (Clontech, Palo Alto, California) and used for transformation of root tissue 'WO 97/27287 PCT/US96/01041 from caI-I api-i plants as described by Valvekens et al.
(Proc. Natl. Acad. Sci., USA 85:5536 (1988))-Seeds were harvested from primary transformants, and all phenotypic analyses were performed in subsequent generations. Four independent lines transformed with CAL showed a complementation of the cauliflower (cal) phenotype and displayed a range of phenotypes similar to those exhibited by api mutants. These results demonstrated that CAL functions to convert shoot meristem to floral meristem.

In order to identify regions of functional importance in the CAL protein, cal mutants were generated and analyzed. The cal alleles were isolated by mutagenizing seeds homozygous for the api-I allele in Ler with 0.1% or 0.05% ethylmethane sulfonate (EMS) for 16 hours. Putative new cal alleles were crossed to ca1-I
api-i chlorina plants to verify allelism. Two sets of oligonucleotides were used to amplify and clone new alleles: oligos AGL10-1 (SEQ ID NO: 25) and AGL10-2 (5'-GATGGAGACCATTAAACAT-3; SEQ ID NO: 27) for the 5' portion and oligos AGL10-3 (5'-GGAGAAGGTACTAGAACG-3'; SEQ
ID NO: 28) and AGL10-4 (5'-GCCCTCTTCCATAGATCC-3'; SEQ ID
NO: 29) for the 3' portion of the gene. All coding reaions and intron-exon boundaries of the mutant alieles were sequenced.

Sequence analysis of the cal-i allele, which exists in the wild-type Wassilewskija (WS) ectoype, revealed a cluster of three amino acid differences in the seventh exon, relative to the wild-type gene product from Landsberg erecta (Ler) (Figure 8). One or more of these amino acid differences can be responsible for the cal phenotype, because the cal-i gene was expressed normally 5 and the transcribed RNA was correctly spliced in the WS
background. The three additional cal alleles that were isolated, designated caZ-2, caI-3, and cal-4, exhibited phenotypes similar to that of the cal-i allele.

Sequence analyses revealed a single missense 10 mutation for each (Figure 8). Since mutations in the cal-2 and cal-3 alleles lie in the MADS domain, these mutations can affect the ability of CAL to bind DNA and activate its target genes. Because the cal-4 allele contains a substitution in the K domain, a motif thought 15 to be involved in protein-protein interactions, this mutation can affect the ability of CAL to form homodimers or to interact with other proteins such as AP1.

B. RNA expression pattern of CAULIFLOWER

To characterize the temporal and spatial 20 pattern of CAL RNA accumulation, RNA in situ hybridizations were performed using a CAL-specific probe.
35S-labeled antisense CAL and BoCAL mRNA was synthesized from Sca 1-digested cDNA templates and hybridized to 8,um sections of Arabidopsis Ler or Brassica oleracea 25 inflorescences. The probes did not contain any MADS box sequences in order to avoid cross-hybridization with other MADS box genes. Hybridization conditions were as = WO 97/27287 PCT/US96/01041 previously described (Drews et al., Cell 65:991 (1991)).

As with AP1, CAL RNA accumulated in young flower primordia, consistent with the ability of CAL to substitute for APi in specifying floral meristems. In contrast to AP1 RNA, however, which accumulated at high levels throughout sepal and petal development, CAL RNA
was detected only at very low levels in these organs.
These results demonstrate that CAL was unable to substitute for AP1 in specifying sepals and petals, at least in part as a result of the relatively low levels of CAL RNA in these developing organs.

C. Molecular Basis of the cauliflower ghenotyge The cal phenotype in Arabidopsis is similar to the inflorescence structure that develops in the closely related species Brassica oleracea var. botrytis, the cultivated garden variety of cauliflower, indicating that the CAL gene can contribute to the cal phenotype of this agriculturally important species. Thus, CAL gene homologs were isolated from a Brassica oleracea line that produces wild-type flowers (BoCAL) and from the common garden variety of cauliflower Brassica oleracea var.

bo tryti s (BobCAL) .

The single-copy BobCAL gene (Snowball Y

Improved, NK Lawn & Garden, Minneapolis, MN) was isolated from a size-selected genomic library in XBlueStar (Novagen) on a 16-kbp BamHI fragment with the Arabidopsis CAL gene as a probe. The BoCAL gene was isolated from a.
rapid cycling line (Williams and Hill, Science 232:1385 (1986)) by PCR on both RNA and genomic DNA. The cDNA was isolated by RT-PCR using the oligonucleotides: Bobl 5(5'-TCTACGAGAAATGGGAAGG-3'; SEQ ID NO: 30) and Bob2 (5'-GTCGATATATGGCGAGTCC-3'; SEQ ID NO: 31). The 5' portion of the gene was obtained using oligonucleotides Bob 1 (SEQ ID NO: 30) and Bob4B
(5'-CCATTGACCAGTTCGTTTG-3'; SEQ ID NO: 32). The 3' portion was obtained using oligonucleotides Bob3 (5'-GCTCCAGACTCTCACGTC-3'; SEQ ID NO: 33) and Bob2 (SEQ
ID NO: 31).

RNA in situ hybridizations were performed to determine the expression pattern of BoCAL gene from Brassica oleracea. As in Arabidopsis, BoCAL RNA
accumulated uniformly in early floral primordia and later was excluded from the cells that give rise to stamens and carpels.

DNA sequence analyses revealed that the open reading frame of the BoCAL gene is intact, whereas that of the BobCAL gene is interrupted by a stop codon in exon 5(Figure 8). Translation of the resulting BobCAL
protein product is truncated after only 150 of the wild-type 255 amino acids. Because similar stop codon mutations in the fifth exon of the Arab%dopsis AP1 coding sequence result in plants having a severe api phenotype, the BobCAL protein likely is not functional. These results indicate that, as in Arabidopsis, the molecular basis for the cauliflower phenotype in Brassica oleracea var. botrytis is due, at least in part, to a mutation in the BobCAL gene.

EXAMPLE IV
Conversion of inflorescenc shoots into flowers in an CAULIFLOWER transgenic 81ant This example describes methods for producing a transgenic CAL plant.

A Ectopic expression of CAULTF WER converts inflorescence shoots to flowers Transgenic Arabidopsis plants that ectopically_ express CAL in shoot meristem were generated. The full-length CAL cDNA was inserted downstream of the 35S
cauliflower mosaic virus promoter in the EcoRI of pMON530 (Monsanto Co. Co., St. Louis, Missouri) This plasmid was introduced into Agrobacterium strain ASE (check) and used to transform the Columbia ecotype of Arabidopsis using a modified vacuum infiltration method described by Bechtold et al. (supra, 1993). The 96 lines generated that harbored the 35S-CAL construct had a range of weak to strong phenotypes. The transgenic plants with the strongest phenotypes (27 lines) closely resembled the tfl mutant.

35S-CAL transgenic plants had converted apical and lateral inflorescence shoots into flowers and showed an early flowering phenotype. These results demonstrate that CAL is sufficient for the conversion of shoots to flowers and for promoting early flowering.

EXAMPLE V

Conversion of shoots into flowers in a LEAFY transgenic plant This example describes methods for producing a transgenic LFY Arabidopsis and aspen.

A. Conversion of Arabidozasis shoots by LEAFY

Transgenic Arabidopsis plants were generated by transforming Arabidopsis with LFY under the control of the cauliflower mosaic virus 35S promoter (CaMV35S)(Odell et al., supra, (1985)). A LFY complementary cDNA (Weigel et al, Cell 69:843-859 (1992)) was inserted into a T-DNA

transformation vector containing a CaMV 35S promoter/3' nos cassette (Jack et al., supra, 1994). Transformed seedlings were selected for kanamycin resistance.
Several hundred transformants in three different genetic backgrounds (Nossen, Wassilewskija and Columbia) were recovered and several lines were characterized in detail.
High levels of LFY RNA expression were detected by northern blot analysis. In general, Nossen lines had weaker phenotypes, especially when grown in short days.
The 35S-LFY transgene of line DW151.117 (ecotype Wassilewskija) was introgressed into the erecta background by backcrossing to a Landsberg erecta strain.

Plants were grown under 16 hours light and 8 hours dark.
The 35S-LFY transgene provided at least as much LFY
activity as the endogenous gene and completely suppressed the 3fy mutant phenotype when crossedinto the background 5 of the lfy-6 null allele.

Most 35S-LFY transgenic plants lines demonstrated a very similar, dominant and he'ritable phenotype. Secondary shoots that arose in lateral positions were consistently replaced by solitary flowers, 10 and higher-order shoots were absent. Although the number of rosette leaves was unchanged from the wild type, 35S-LFY plants flowered earlier than wild type; the solitary flowers in the axils of the rosette leaves developed and opened precociously. In addition, the 15 primary shoot terminated with a flower. In the most extreme cases, a terminal flower was formed immediately above the rosette. This gain of function phenotype (conversion of shoots to flowers) is the opposite of the Ify loss of function phenotype (conversion of flowers to 20 shoots). These results demonstrate that LFY encodes a developmental switch that is both sufficient and necessary to convert shoot meristem to flower meristem.

The effects of constitutive LFY expression differ for primary and secondary shoot meristems.
25 Secondary meristems were transformed into flower meristem, apparently as soon as it developed, and produced only a single, solitary flower. In contrast, primary shoot meristem produced leaves and lateral flowers before being consumed in the formation of a WO 97/27287 PCTlUS96/01041 terminal flower. These developmental differences indicate that a meristem must acquire competence to respond to the activity of a floral meristem identity gene such as LFY.

B. Conversion of aspen shoots by LEAFY

Given that constitutive expression of LFY
induced precocious flowering during the vegetative phase of Arabidopsis, the effect of LFY on the flowering of other species was examined. The perennial tree, hybrid aspen, is derived from parental species that flower naturally only after 8-20 years of growth (Schopmeyer (ed.), USDA Aariculture Handbook 450: Seeds of Woody Plants in the United States, Washington DC, USA: US
Government Printing Office, pp. 645-655 (1974)). 35S-LFY

aspen plants were obtained by Agrobacterium-mediated transformation of stem segments and subsequent regeneration of transgenic shoots in tissue culture.

Hybrid aspen was transformed exactly as described by Nilsson et al. (Transcren. Res. 1:209-220 (1992)).

Levels of LFY RNA expression were similar to those of 35S-LFY Arabidopsis, as determined by northern blot analysis. The number of vegetative leaves varied between different regenerating shoots, and those with a higher number of vegetative leaves formed roots, allowing for transfer to the greenhouse. Individual flowers were removed either from primary transformants that had been transferred to the greenhouse, or from catkins collected in spring, 1995, at Carlshem, ITmea, Sweden) from a tree whose age was determined by counting the number of annual rings in a core extracted with an increment borer at 1.5 meters above ground level. Flowers were fixed in formaldehyde/acetic acid/ethanol and destained in ethanol before photography.

The overall phenotype of 35S-LFY aspen was similar to that of 35S-LFY Arabidopsis. In wild-type plants of both species, flowers normally are formed in lateral positions on inflorescence shoots. In aspen, these inflorescence shoots, called catkins, arise from the leaf axils of adult trees. In both 35S-LFY
Arabidopsis and 35S-LFY aspen, solitary flowers were formed instead of shoots in the axils of vegetative leaves. Moreover, as in Arabidopsis, the secondary shoots of trangenic aspen were more severely affected than the primary shoot.

Regenerating 35S-LFY aspen shoots initially produced solitary flowers in the axils of normal leaves.
However, the number of vegetative leaves was limited, and the shoot meristem was prematurely consumed in the formation of an aberrant terminal flower. Precocious flower development was specific to 35S-LFY transformants and was not observed in non-transgenic controls.
Furthermore, not a single instance of precocious flower development has been observed in more than 1,500 other lines of transgenic aspen generated with various constructs from 1989 to 1995 at the Swedish University of Agricultural Sciences. These results demonstrate that a heterologous floral meristem identity gene product is active in an angiosperm.

Although the invention has been described with reference to the examples above, it should be understood that various modifications can be made without departing from the spirit pf the invention. Accordingly, the invention is limited only by the following claims.

Claims

What is claimed is:

1. A nucleic acid molecule encoding a CAULIFLOWER (CAL) gene product having at least 89 percent amino acid identity with amino acids 1 to 160 of the sequence shown in Figure 5 (SEQ ID NO:10) or with amino acids 1 to 160 of the sequence shown in Figure 6 (SEQ ID NO:12 ), wherein the CAL gene product converts shoot meristem to floral meristem.

2. The nucleic acid molecule of claim 1, wherein said CAL gene product is selected from the group consisting of Arabidopsis thaliana CAL having the amino acid sequence shown in Figure 5 (SEQ ID NO:10) and Brassica oleracea CAL having the amino acid sequence shown in Figure 6 (SEQ ID NO:12).

3. A nucleic acid molecule selected from the group consisting of a nucleic acid molecule having the nucleic acid sequence shown in Figure 5 (SEQ ID NO:9) and a nucleic acid molecule having the nucleic acid sequence shown in Figure 6 (SEQ ID NO:11).

4. A nucleic acid molecule encoding a truncated CAULIFLOWER (CAL) gene product, wherein said truncated CAL
gene product is Brassica oleracea var. botrytis CAL having the amino acid sequence shown in Figure 7 (SEQ ID NO:14).

5. A nucleic acid molecule having the nucleic acid sequence shown in Figure 7 (SEQ ID NO:13).

6. A nucleic acid molecule that hybridizes under relatively stringent conditions to a nucleic acid molecule selected from the nucleic acid molecule of claim 3 or a nucleic acid molecule complementary thereto; and the nucleic acid molecule of claim 5 or a nucleic acid molecule complementary thereto, the relatively stringent conditions comprising hybridization at 60°C
in 50% formamide and washes at 65°C in 0.1 x SSC and 0.5% SDS, wherein the nucleic acid molecule has at least 89 percent identity to the nucleic acid molecule of claim 3 or claim 5 or complements thereof, wherein the nucleic acid molecule encodes a CAL gene product that converts shoot meristem to floral meristem.

7. A CAULIFLOWER (CAL) gene selected from the group consisting of an Arabidopsis thaliana CAL gene having the nucleotide sequence shown in Figure 13 (SEQ ID NO: 20), a Brassica oleracea CAL gene having the nucleotide sequence shown in Figure 14 (SEQ ID NO: 21) and a Brassica oleracea var. botrytis CAL gene having the nucleotide sequence shown in Figure 15 (SEQ ID NO: 22).

8. A vector, comprising the nucleic acid molecule of claim 1.

9. A vector, comprising the gene of claim 7.

10. A vector, comprising a nucleic acid molecule selected from the group consisting of the nucleic acid molecule of claim 2 and the nucleic acid molecule of claim 3.

11. A host cell, comprising the vector of claim 8.

12. The vector of claim 8, wherein said vector is an expression vector.

13. An expression vector, comprising a nucleic acid molecule selected from the group consisting of the nucleic acid molecule of claim 2 and the nucleic acid molecule of claim 3.

14. The expression vector of claim 12, further comprising a cauliflower mosaic virus 35s promoter.

15. The expression vector of claim 12, further comprising an inducible regulatory element.

16. A kit for converting shoot meristem to floral meristem in an angiosperm, comprising the expression vector of claim 12 and instructions for use.

17. A kit for promoting early flowering in an angiosperm, comprising the expression vector of claim 12 and instructions for use.

18. A CAULIFLOWER (CAL) polypeptide having at least 89 percent amino acid identity with amino acids 1 to 160 of the sequence shown in Figure 5 (SEQ ID NO:10) or with amino acids 1 to 160 of the sequence shown in Figure 6 (SEQ ID NO:12), wherein the CAL polypeptide converts shoot meristem to floral meristem.

19. The CAL polypeptide of claim 18, wherein said CAL
polypeptide is Arabidopsis thaliana CAL polypeptide having the amino acid sequence shown as amino acids 1 to 255 in Figure 5 (SEQ ID NO:10).

20. The CAL polypeptide of claim 18, wherein said CAL
polypeptide is Brassica oleracea CAL polypeptide having the amino acid sequence shown as amino acids 1 to 255 in Figure 6 (SEQ ID NO:12) .

21. An antibody that specifically binds the CAL
polypeptide of claim 18.

22. A truncated Brassica oleracea var. botrytis CAULIFLOWER (CAL) polypeptide having the amino acid sequence shown as amino acids 1 to 150 in Figure 7 (SEQ ID NO:14).

23. An antibody that specifically binds the truncated Brassica oleracea var. botrytis CAL polypeptide of claim 22.