AU2005253642A1 - Nucleic acid molecules and their use in plant male sterility - Google Patents

Description

WO 2005/122751 PCT/AU2005/000851 NUCLEIC ACID MOLECULES AND THEIR USE IN PLANT STERILITY The present invention relates to nucleic acid molecules. More particularly, the invention relates to genes involved in pollen 5 formation and genetic methods for producing male sterile plants using constructs which disrupt expression of such genes. The invention further relates to the constructs used and transgenic plants transformed by the constructs. 10 BACKGROUND OF THE INVENTION Heterosis is the term used to describe the superior performance of F1 hybrids over parental lines from which the F1 is derived. In crop plants heterosis often manifests in the production of larger 15 plants, tolerance to stress, disease resistance, uniformity and improved yield, which are collectively referred to as hybrid vigour. Heterosis has been observed and documented in many important crop 20 species and the development of hybrids by plant breeders is established practice. However, hybrids can only be used in crops when effective and economical means of pollination control exist to ensure cross pollination and prevent self-pollination. Pollination control mechanisms include mechanical, chemical and genetic means. 25 A mechanical means for hybrid plant production can be used if the plant of interest has spatially separate male and female flowers or separate male and female plants. For example, a maize plant has pollen-producing male flowers in an inflorescence at the apex of 30 the plant, and female flowers in the axiles of leaves along the stem. Outcrossing of maize is assured by mechanically detasseling the female parent to prevent selfing. Even though detasseling is currently used in hybrid seed production for plants such as maize, WO 2005/122751 PCT/AU2005/000851 2 the process is labor-intensive and costly, both in terms of the actual detasseling cost and yield loss as a result of detasseling the female parent. Further, most major crop plants have both functional male and female organs within the same flower, and 5 therefore emasculation is not a simple procedure. While it is possible to remove by hand the pollen forming organs before pollen is shed, this form of hybrid production is extremely labor intensive and expensive. 10 Chemical means of producing hybrid plants involves the use of chemicals that kill or block viable pollen formation. These chemicals, termed gametocides, are used to impart a transitory male-sterility. Commercial production of hybrid plants by use of gametocides is limited by the expense and availability of the 15 chemicals and the reliability and length of action of the applications. A serious limitation of gametocides is that they have phytotoxic effects, the severity of which are dependent on genotype. Other limitations include that these chemicals may not be effective for crops with an extended flowering period because new 20 flowers produced may not be affected. Consequently, repeated application of chemicals is required. Many current commercial hybrid plant production systems for field crops rely on a genetic means of pollination control. In such 25 systems, plants that are used as females either fail to make pollen, fail to shed pollen, or produce pollen that is biochemically unable to effect self-fertilization. Plants that are unable to self-fertilize are said to be "self-incompatible" (SI). Difficulties associated with the use of a self-incompatibility 30 system include availability and propagation of the self incompatible female line, and stability of the self-compatibility. In some instances, self-incompatibility may be overcome chemically, or immature buds can be pollinated by hand before the biochemical WO 2005/122751 PCT/AU2005/000851 3 mechanism that blocks pollen is activated. Self-incompatible systems that can be deactivated are often very vulnerable to stressful climatic conditions that break or reduce the effectiveness of the biochemical block to self-pollination. 5 Genetic systems involving cytoplasmic or nuclear genes may be used to generate male sterility. Cytoplasmic male sterility (CMS) is at present the most widely used mechanism of pollen control in crops. However, CMS has a number of disadvantages including increased 10 disease susceptibility, breakdown of sterility under certain conditions, undesirable characters linked to restorer genes (genes that can suppress the male-sterile effect of the cytoplasm and are incorporated into the male parent to restore pollen fertility in the Fl hybrid), unreliable restoration, etc. Crops in which such 15 problems occur are, for example, maize, oilseed rape and wheat. Consequently, plant breeders and seed producers require a versatile and durable male sterility system. The genetic engineering approach has a number of advantages over natural systems. The 20 disruption of the genotype of new male-sterile plants normally associated with the introduction of sterility by sexual means is completely avoided and sterility is not linked to particular cytoplasms as in cytoplasmic male sterility. 25 Nuclear-encoded male sterility (NMS) is caused by mutations in the nuclear genome. The advent of plant genetic engineering technology has now made it feasible to develop strategies to permit the use of NMS genes. 30 Plant breeders and seed producers require a versatile and durable male sterility system. The problems that must be overcome include the isolation of genes that induce male sterility, the production of 100% male-sterile progeny, the achievement of complete female WO 2005/122751 PCT/AU2005/000851 4 fertility and subsequent restoration of pollen fertility in the F1 hybrid. An aim of the present invention is to determine gene(s) involved in 5 pollen formation and develop methods of preventing pollen formation by disrupting the expression or activity of the gene(s) or their gene product(s). A further aim of the present invention is to develop a method for the production of transgenic male sterile plants and/or seeds and harvest plants and/or seeds so produced. 10 SUMMARY OF THE INVENTION Accordingly the present invention provides in a first aspect a method for disrupting pollen development in a plant, the method 15 comprising inhibiting the expression of an endogenous nucleic acid molecule which is, under normal conditions, detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding transcription factors. 20 Preferably the nucleic acid molecule whose expression is blocked encodes MYB 32 or MYB 103. Preferably MYB32 has the amino acid sequence provided in Figure 1 or is a homologue, orthologue or derivative thereof having at least 50% similarity to the amino acid 25 sequence of Figure 1 and which capable of being detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding transcription factors. 30 The invention extends to ortholgues of MYB32, including that from Brassica napus, where BnMYB32 has the amino acid sequence provided in Figure 1 (SEQ ID NO 1), Arabidopsis thaliana, where AtMYB32 has the amino acid sequence provided in Figure 4 (SEQ ID NO. 7), WO 2005/122751 PCT/AU2005/000851 5 tomato, where TmH27 has the amino acid sequence provided in Figure 5a (SEQ ID NO. 9), cotton, where GhMYB9 has the amino acid sequence provided in Figure Sb (SEQ ID NO 10), Zea mays, where ZmMYB32-1 has the amino acid sequence provided in Figure 7a (SEQ ID NO. 12) and 5 ZmMYB32-2 has the amino acid sequence provided in Figure 7b (SEQ ID NO. 13), Hordeum vulgare, where HvMYB32 has the amino acid sequence provided in Figure 7c (SEQ ID NO. 14), Nicotiana tabacum, where TaMYB32 has the amino acid sequence provided in Figure 7d (SEQ ID NO. 15) or Triticum aestivum, where TaMYB32 has the amino acid 10 sequence provided in Figure 36 (SEQ ID NO. 17). The invention also extends to orthologues of MYB103, including Brassica napus, where BnMYB 103-1 has the amino acid sequence provided in Figure 2 (SEQ ID NO. 3) and BnMYB103-2 has the amino 15 acid sequence provided in Figure 3 (SEQ ID NO. 5), Arabidopsis thaliana, where AtMYB103 has the amino acid sequence provided in Figure 11 (SEQ ID NO. 22), rice, where OsMYB103 has the amino acid sequence provided as Figure 6 (SEQ ID NO. 11) or Figure 37 (SEQ ID NO. 18), or Triticum aestivum, where TaMYB103 has the amino acid 20 sequence provided in Figure 38 (SEQ ID NO. 20). Expression of the endogenous nucleic acid molecule may be effected by sense or co-suppression technology, for example by antisense RNA interference technology. 25 The disruption of pollen formation according to the first aspect of the invention may induce male sterility in plants. In a preferred embodiment the induction of male sterility is inducible and reversible. 30 To bring about 100% male sterility it may be necessary to block expression of nucleic acid molecules encoding both MYB 32 and MYB 103. This is particularly the case for Arabidopsis thaliana.

WO 2005/122751 PCT/AU2005/000851 6 In a second aspect the invention provides an isolated nucleic acid molecule capable of blocking expression of a gene encoding MYB32 having the amino acid sequence provided in Figure 1 or an 5 orthologue thereof having at least 50% sequence homology to the amino acid sequence provided in Figure 1, which MYB32 gene or orthologue are detectably expressed in anther tissue during pollen formation. 10 In a preferred embodiment the MYB32 gene comprises the nucleotide sequence provided in Figure 1 (SEQ ID NO. 2), 4 (SEQ ID NO. 8), or Figure 36 (SEQ ID NO. 16) or a sequence complementary or antisense thereto, or a fragment of the nucleotide sequence or its complement, which nucleic molecule is capable of blocking 15 expression of the gene encoding MYB32. In a third aspect the invention provides an isolated nucleic acid molecule capable of blocking expression of a gene encoding MYB103 having the amino acid sequence provided in Figure 2 or Figure 3 or 20 an orthologue thereof having at least 50% sequence homology to the amino acid sequence provided in Figure 2 or Figure 3, which MYB103 gene or orthologue are detectably expressed in anther tissue during pollen formation. 25 In a preferred embodiment the MYB103 gene comprises the nucleotide sequence provided in Figure 2 (SEQ ID NO. 4), 3 (SEQ ID NO. 6), 37 (SEQ ID NO. 19) or Figure 38 (SEQ ID NO. 21) or a sequence complementary or antisense thereto, or a fragment of the nucleotide sequence or its complement, which nucleic 30 molecule is capable of blocking expression of the gene encoding MYB103.

WO 2005/122751 PCT/AU2005/000851 7 The MYB103 or MYB32 genes according to the second and, or third aspects of the invention may be provided in a nucleic acid construct. The construct may further comprise a promoter capable of targeting expression of the nucleic acid molecule in 5 the construct to the anther, preferably during pollen formation. The construct may comprise both a MYB32 and a MYB103 gene, or more than one copy of the, or each, gene. The construct may further comprise one or more additional genes involved in male 10 specific development, such as the AtMYB103 gene shown in Figure 12 (SEQ ID NO. 23). The construct may comprise an inducible promoter. In a fourth aspect the invention provides a plant cell transformed with the isolated nucleic acid molecule according to the second or 15 third aspect of the invention or a construct comprising such nucleic acid molecule. In a fifth aspect the invention provides a transgenic plant generated from the plant cells according to the fourth aspect of 20 the invention. Also encompassed are the progeny of such plants and seeds, tissues, roots, shoots, tubers or propagating material of such transgenic plants or their progeny. The nucleic acid molecules according to the second and third 25 aspects of the invention find particular utility in the production of male sterile plants. In a sixth aspect the invention provides a male sterile plant in which expression of an endogenous nucleic acid molecule is blocked, 30 which nucleic acid molecule is, under normal conditions, detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding transcription factors.

WO 2005/122751 PCT/AU2005/000851 8 In a preferred embodiment of the sixth aspect expression of one or more endogenous genes is blocked using a nucleic acid molecule according to the second or third aspect of the invention or a 5 nucleic acid construct comprising said nucleic acid molecule. The male sterile plant may be monocotyledonous or dicotyledonous. The invention is particularly suitable to legumes, crops, cereals, native grasses, fruiting plants or flowering plants. 10 Preferred plants include Brassicaceae or other Solanaceae species, including potato and cole vegetables, cabbage, kale, collards, turnips, rutabaga, kohlrabi, Brussels sprouts, broccoli and cauliflower, mustards and oilseeds, crucifers, broccoli, canola, 15 tomato, grain legumes, wheat, barley, maize, tobacco and rice. Reversible male sterility is critical for the production of hybrid plant varieties that exhibit a higher yield than either of the parents. In particular the present invention enables regulation of 20 plant genes by turning off the endogenous gene encoding MYB 32 or MYB 103 in a female parental line but turning on the endogenous equivalent of a the gene on the hybrid that results from crossing the female parental line with another line. 25 BRIEF DESCRIPTION OF THE FIGURES Figure 1 provides the nucleic acid sequence and proposed amino acid sequence of BnMYB32. 30 Figure 2 provides the nucleic acid sequence and proposed amino acid sequence of BnMYBl03-1. Conserved amino acids are indicated by underscore.

WO 2005/122751 PCT/AU2005/000851 9 Figure 3 provides the nucleic acid sequence and proposed amino acid sequence of BnMYBl03-2. Figure 4 provides the nucleic acid sequence and proposed amino acid 5 sequence of AtMYB32. No known plant transcription elements were identified in the promoter region of AtMYB32. The Bbsl sites are shown underlined, as is the putative TATA consensus sequence. The conserved amino acid sequences found in the carboxyl region of other plant MYB-related proteins are shown underlined. 10 Figure 5 provides the proposed amino acid sequences of (a) TmH27 and (b)GhMYB9. Figure 6 provides the proposed amino acid sequence of OsMYB103. 15 Figure 7 provides the proposed amino acid sequence of (a) ZmMYB32 1, (b) ZmMYB32-2, (c) HvMYB32 and (d) TaMYB32. Figure 8 illustrates a species tree generated from a list of 20 potential BnMYBl03-1 orthologues derived from an orthologue search of the NCBI database. Each species is represented by one or more MYB genes (see Figure 9). Figure 9 illustrates a gene tree of plant MYB genes related to 25 BnMYB103-1, AtMYB103 and BnMYB103-1 are shown in bold. Figure 10 illustrates AtMYBl03 probes used to screen 3. napus cDNA (600 bp) and genomic DNA (450bp and 500bp) libraries. 30 Figure 11 shows the AtMYBl03 and p8.1.1 CLUSTAL X (1.8) sequence alignment. Both sequences contain conserved tryptophan and phenylalanine residues typical of the R2R3 repeat regions of plant MYB genes (indicated by underscore).

WO 2005/122751 PCT/AU2005/000851 10 Figure 12 shows the AtMYB103 and BnMYBl03-1 CLUSTAL X (1.8) sequence alignment. Arrow heads indicate conserved intron sites. 5 Figure 13 shows the BnMYBl03-l amino acid sequence indicating conserved amino acids by underscore. Figure 14 shows the AtMYB103 and BnMYBl03-1 CLUSTAL X (1.8) sequence alignment. 10 Figure 15 shows the 700bp RT PCR product. The arrow indicates 700bp RT PCR product. The large product represents amplification from genomic DNA. Lane 1 XEcORI/HindIII marker 15 Lane 2 700bp RT PCR product (1.5 mM MgCl 2 ) Lane 3 700bp RT PCR product (2.0 mM MgCl 2 ) Lane 4 products from B.napus genomic DNA Lane 5 no DNA 20 Figure 16 shows the AtMYBI03 and p700-1 CLUSTAL X (1.8) sequence alignment. Figure 17 shows the 800bp RT PCR product. The arrow indicates 800bp RT PCR product. The large product represents amplification 25 from genomic DNA. Lane 1 XEcORT/HindIII marker Lane 2 800bp RT PCR product (1.0 mM MgCl 2 ) Lane 3 800bp RT PCR product (1.5 mM MgCl 2 ) Lane 4 800bp RT PCR product (2.0 mM MgCl 2 ) 30 Figure 18 shows the AtMZYB103 and p800-19 CLUSTAL X (1.8) sequence alignment.

WO 2005/122751 PCT/AU2005/000851 11 Figure 19 shows the 950bp RT PCR product. The arrow indicates 950bp RT PCR product. Lane 1 XEcORI/HindIII marker Lane 2 950bp RT PCR product (2.0 mM MgC1 2 ) 5 Lane 3 950bp RT PCR product (2.5 mM MgCl 2 ) Lane 4 no RNA Figure 20 shows the AtMY13l03 and p900-10 CLUSTAL X (1.8) sequence alignment. 10 Figure 21 shows the AtMYB103, BnMYBl03-1, p700-1 (SEQ ID NO. 24), p800-19 (SEQ ID NO. 25) and p900-10 (SEQ ID NO. 26) CLUSTAL X (1.8) multiple sequence alignment. All sequences contain conserved tryptophan and phenylalanine residues typical of the R2R3 repeat 15 regions in plant MYB genes (indicated by underscore). Differences in the amino acid sequences are highlighted. Figure 22 shows A - AtMYB103 promoter - gusA in A. thaliana (Bar = 15mm). B- BnMYBl03-1 promoter - gusA in A. thaliana (Bar = 25mm). 20 Figure 23 shows sections of GUS stained anthers from A. thaliana. A shows anthers from a plant carrying an AtMYBI03 promoter-gus A construct (Bar = 20pm). B shows anthers from a plant carrying a BnMYB103-l promoter-gus A construct (Bar = 30gm). GUS staining is 25 confined to the tapetum (t) and microspores (m). Figure 24 illustrates the BnMYB103-1 fragment used to probe B. napus Northern blot. 30 Figure 25 shows a Northern blot analysis of B.napus tissues. Panel A shows the results of probing with the 484bp BnMYBl03-1 gene specific probe. Panel B shows the results of probing with the ubiquitin probe.

WO 2005/122751 PCT/AU2005/000851 12 Lane 1 dehiscing anthers Lane 2 anthers from flower buds 2-6 mm Lane 3 flower buds 0.5-2 mm Lane 4 flower buds 2.5-5 mm 5 Lane 5 flower buds 6-8 mm Lane 6 shoots Lane 7 roots Figure 26 RT PCR analysis of B.napus tissues. The top panel shows 10 RT PCR products generated from B. napus tissues using primers specific to BnMYB1L03-l, a product of 979bp is expected). The bottom panel B shows RT PCR generated using primers specific to the 3-8 tubulin gene (a product of 545bp is expected). Lane 1 molecular weight markers 15 Lane 2 dehiscing anthers Lane 3 anthers from flower buds 2-6 mm Lane 4 flower buds 0.5-2 mm Lane 5 flower buds 2.5-5 mm Lane 6 flower buds 6-8 mm 20 Lane 7 shoots Lane 8 roots Figure 27 shows RT PCR analysis of pre and post anthesis anthers for BnMYB103-1 mRNA. 25 Lane 1 molecular weight markers Lane 2 post anthesis anthers, primers A5956A10 & A2485F04 Lane 3 pre anthesis anthers, primers A5956A10 & A2485F04 These primers have amplified a 784bp fragment from B.napus cDNA. Lane 4 post anthesis anthers, primers A5962HO2 & Z1346E10 30 Lane 5 pre anthesis anthers, primers A5962H02 & Z1346E10 These primers have amplified a 979bp fragment from B.napus cDNA.

WO 2005/122751 PCT/AU2005/000851 13 Figure 28 shows TA39 promoter GUS expression. A shows TA39::gusA expression in A. thaliana (Bar = 25 mm). B shows TA39::gusA expression in B. napus (Bar = 2 mm). 5 Figure 29 shows restriction enzyme digests of BnMYB103-l sense and antisense constructs. A & B show sense and antisense BnMYB103-1 constructs cloned into pBluescipt with the TA39 promoter. C shows restriction enzyme digests (PvuII & SnaBI) of two clones, 10 indicating the sense orientation in lane 2 (expect fragments of 2513 bp - vector, 2512 bp & 475 bp) and antisense orientation in lane 3 (expect fragmets of 2513 bp -vector, 2171 bp & 816 bp). Figure 30 shows BnMYB103-1 sense and antisense constructs. NOS pro 15 indicates nopaline synthase promoter, NPTIT indicates neomycin phosphotransferase II and NOS-ter indictes nopaline synthase terminator/ Figure 31 shows PCR anaylsis of A. thaliana plants transformed with 20 sense and antisense BnMYB1103-1 constructs. Lane 1 molecular weight markers Lanes 2-6 PCR products generated from plants transformed with antisense constructs using primers P30 & P32 (903 bp) Lanes 7-11 PCR products generated from plants transformed with 25 sense constructs using primers P30 & P34 (636 bp). Figure 32a shows scanning electron micrographs of pollen from A.thaliana wild type and BnMYBl03-l transgenic plants. 1 shows wild type pollen grains 30 2 gives detail showing flattened pollen grains from AS22 3 shows representative field of pollen from AS15 showing abnormal grain shape and positioning of germinal apertures.

WO 2005/122751 PCT/AU2005/000851 14 Figure 32b shows scanning electron micrographs of pollen from A.thaliana BnMYB103-1 transgenic plants. These panels show pollen from plants transgenic for the sense construct. 1 representative field of pollen from S11 5 2 representative field of pollen from S20. Figure 33 shows production of a construct according to a preferred embodiment of the present invention. 10 Figure 34 shows production of a further construct according to a preferred embodiment of the present invention. Figure 35 A-H illustrate changes in phenotype in A. thaliana as a result of carrying out a method according to a preferred embodiment 15 of the present invention. Figure 36 shows the nucleotide sequence of TaMYB32 and the alignment of the deduced amino acid sequences of AtMYB32 and TaMYB32. The MYB domain sequences are marked in bold letters 20 and the identical amino acids are underlined. The gaps introduced for optimal alignment are represented by broken lines. Figure 37 shows the putative rice AtMYB103 orthologue, OsMYB103. 25 Sequence was obtained from Genbank. Introns are shown in lower case and amino acid sequence is shown below the nucleotide sequence. In the alignment shaded boxes show the position of primers. Conserved MYB repeats R2 and R3 are underlined. Conserved residues important for protein folding are shown in 30 bold. Note - sections of low homology (between 168 and 306 in the AtMYB103 protein) are not shown.

WO 2005/122751 PCT/AU2005/000851 15 Figure 38 shows an alignment of the deduced amino acid sequences of AtMYB103 and TaMYB103 fragments and a partial nucleotide sequence of the wheat TaMYB103 gene. Identical amino acids in the alignment are underlined. 5 Figure 39 shows the structure of repressor (top) and restorer (bottom) constructs. The promoters are represented by the lines and the AtMYB103 gene by light shading, and the repression motif by dark shading. 10 DETAILED DESCRIPTION OF THE INVENTION The present invention relates to the identification by the inventors of certain genes which are expressed in the anthers of plants during pollen formation and their work showing that such 15 genes are vital for pollen formation and thus may be blocked to produce male sterile plants. The inventors' earlier research papers, Li et al., Plant Cell Physiol. 40(3); 343-347 (1999) and Higginson et al., The Plant 20 Journal (2003) 35, 177-192 describe a MYB-like gene from Arabidopsis, AtMYBl03. This gene was shown to be expressed in the anther during the early stages of anther development and was suggested to be involved in pollen formation. Unpublished work by the inventors showed that blocking of this gene alone in 25 Arabidopsis was insufficient to produce 100% male sterility. Without 100% male sterility pollen is still able to be produced and accordingly heterosis may not be achieved. Subsequent work described in this specification shows the WO 2005/122751 PCT/AU2005/000851 16 discovery of an additional Arabidopsis gene expressed in anther' tissue and involved in pollen formation. Provided herein is evidence that blocking the expression of both the Arabidopsis genes allows production of 100% male sterile Arabidopsis plants. 5 Additionally the inventors have sought to find if there are functional ortholgues of the Arabidopsis genes in other plant species and here describe several such orthologues and methods of inducing male sterility in crop species such as Brassica by 10 blocking the expression of the endogenous gene corresponding to the orthologues. "Nucleic acid" as used herein refers to an oligonucleotide, polynucleotide, nucleotide and fragments or portions thereof, as 15 well as to peptide nucleic acids (PNA), fragments, portions or antisense molecules thereof, and to DNA or RNA of genomic or synthetic origin which can be single- or double-stranded, and represent the sense or antisense strand. Where "nucleic acid" is used to refer to a specific nucleic acid sequence "nucleic acid" is 20 meant to encompass polynucleotides that encode a polypeptide that is functionally equivalent to the recited polypeptide, e.g., polynucleotides that are degenerate variants, or polynucleotides that encode biologically active variants or fragments of the polypeptide, including polynucleotides having substantial sequence 25 similarity or sequence identity relative to the sequences provided herein. Similarly, "polypeptide as used herein refers to an oligopeptide, peptide, or protein. Where "polypeptide" is recited herein to refer 30 to an amino acid sequence of a naturally-occurring protein molecule, "polypeptide" and like terms are not meant to limit the amino acid sequence to the complete, native amino acid sequence associated with the recited protein molecule, but instead is meant WO 2005/122751 PCT/AU2005/000851 17 to also encompass biologically active variants or fragments, including polypeptides having substantial sequence similarity or sequence identify relative to the amino acid sequences provided herein. 5 By "isolated" we mean free from material present in nature in the plant from which the nucleic acid molecule is derived, that is in an environment different from that in which the compound naturally occurs. "Isolated" is meant to include compounds that are within 10 samples that are substantially enriched for the compound of interest and/or in which the compound of interest is partially or substantially purified. The nucleic acid molecules of the present invention are expressed 15 in anther tissue during pollen formation. Whilst the nucleic acid molecules are not exclusively expressed in the anther, nor only during pollen development, these molecules are more specific than "house-keeping" genes which are active in all plant cells. The invention does not encompass genes that are not expressed in the 20 anther during pollen development. For the avoidance of any doubt the present invention does not encompass the ZmMYBP2 gene, the NtMYBAS1 gene or the NtMYNAS2 gene. None of the prior art in relation to these genes describes that the gene is expressed in the anther during pollen development. The inventors' studies would 25 tend to suggest that the ZmMYBP2 gene, the NtMYBAS1 gene, and the NtMYNAS2 gene are not expressed in the anther during pollen development. The prior art does not specifically disclose that the AtMYB103 gene is expressed in the anther during pollen development. The inventors have found that it is and have also found that 30 blocking expression of this gene alone is not capable of producing 100% male sterility. Accordingly the present invention does not encompass the AtMYB103 gene, although it does encompass the use of the gene in inducing male sterility.

WO 2005/122751 PCT/AU2005/000851 18 preferably, the majority of the expression of the nucleic acid molecule during pollen development is in the anther. Such molecules may be considered "anther specific". 5 Anther tissue describes the tissue of the male reproductive organs in a plant, be it fully developed or partially developed. The definition of anther tissue used herein is intended to include all structures making up the anther, that is, the epidermis, 10 endothecium, middle layer and tapetum. The nucleic acid molecules according to the second and third aspects of the invention encode proteins belonging to the MYB class of DNA binding transcription factors. MYB (myeloblastosis) genes 15 represent the largest family of transcription factors found in plants with around 180 MYB genes known in the genome of A. thaliana. The first MYB genes were isolated from avian leukaemia viruses and they have now been described in mice, Drosophila melanogaster, Dictyostelium discoideum, yeast and fungi. It has 20 been estimated that over 5% of the Arabidopsis genome codes for transcription factors. This accounts for approximately 1700 genes of which 180 (10.6%) belong to the MYB family. Maize expresses over 80 MYB genes, it is estimated that petunia has at least 40, rice more than 20 and others have been reported in Antirrhinum, 25 barley, pea, sorghum and the moss Physcomitrella patens. The majority of MYB genes found in plants are of the R2R3 type, although some exceptions have been reported. Genes with a single repeat unit have been found in potato and Arabidopsis. 30 The vertebrate family of MYB transcription factors consists of A-, B- and c-MYB and all three are involved in the regulation of cellular proliferation, differentiation and apoptosis. This is in contrast to the many MYB proteins found in plants, none of which WO 2005/122751 PCT/AU2005/000851 19 have been found to share a similar function. A study aimed at the functional characterisation of R2R3 MYB genes in Arabidopsis has shown that they are expressed in different organs (seedlings, young and mature leaves, cauline leaves, stem, flower buds, siliques and 5 roots) and in response to a variety of environmental conditions (that is different sucrose/nitrogen regimes, hormone treatments, pathogen attack, different light regimes and cold and drought stress).' Plant MYB genes are involved in many functions including the regulation of secondary metabolism, cellular development and 10 signal transduction. A number of MYB genes are expressed in response to plant hormones, biotic and abiotic stress. Most of the plant MYB genes with known function are involved in regulating phenylpropanoid metabolism. 15 A number of MYB genes have been suggested to be involved in the regulation of developmental processes such as shoot morphogenesis (AtIYB13), growth and dorsoventrality of lateral organs (PHANTASTICA) embryogenesis (AtMYBR1 & AtMYBR2) and anther 20 development (AtMYBI03 [Li et al., 1999 supra). Pollen is the haploid male gametophyte in flowering plants and carries the sperm cells required for fertilisation of the ovules. These tiny grains develop within the anther and are released as the 25 anther matures by a process referred to as dehiscence. Pollen formation requires the expression of temporally and spatially specific genes in the gametophyte and sporophyte. It has been estimated that 20 000 mRNA transcripts are present in the 30 mature pollen of Tradescantia paludosa and maize. It is likely that many more transcripts are produced in the early stages of pollen formation, which reflects the complex nature of this process and the need for regulatory genes.

WO 2005/122751 PCT/AU2005/000851 20 The study of pollen formation over many years has shown significant similarities in a number of plant species. In recent years attention has focussed on Arabidopsis thaliana, a model plant for 5 research due to its compact size, short life cycle and small genome which has now been entirely sequenced. The male haploid phase of development in angiosperms results in the production of pollen, the male gametophyte. Pollen develops within 10 the anther a structure composed of four layers, the epidermis, endothecium, middle layer and tapetum. Within the anther locules, adjacent to the tapetum, lie sporogenous cells from which pollen will develop. The expression of temporally and spatially specific genes in the gametophyte (meiocytes) and sporophyte (tapetum) is 15 required for complete pollen development. Mature pollen grains consist of a vegetative and generative cell (bicellular) or a vegetative cell and two sperm cells (tricellular), surrounded by the intine and exine. A number of common developmental stages have been identified: 20 Sporogenesis Sporogenous cells form meiocytes and these undergo two meictic divisions (meiosis I & II) resulting in a tetrad of haploid microspores surrounded by a P-1,3-glucan (callose) wall. The 25 pollen wall begins to develop as sporopollenin is deposited, forming the exine. Tapetal cells increase in size, numerous ribosomes are present, the endoplasmic reticulum and Golgi proliferate and the number of secretory vesicles increases. As the callose wall degrades, the tapetum reaches its highest secretory 30 phase and the stage of microgametogenesis development ensues.

WO 2005/122751 PCT/AU2005/000851 21 Microgametogenesis During this stage the free microspores grow rapidly with further deposition of the exine layer. This coincides with sporopollenin synthesis and secretion by the tapetum. The microspores become 5 vacuolated, at first with many small vacuoles, which later fuse to form a single large vacuole. At this stage exine formation is complete, sporopollenin synthesis has ceased, deposition of the intine begins and internal degradation of the tapetum has commenced. 10 Microspore maturation This stage is characterised by an asymmetric mitotic division (mitosis I) that gives rise to a large vegetative cell and a smaller generative cell. The vegetative cell is involved in 15 further pollen development and later in the formation of the pollen tube. The generative cell will undergo another mitotic division (mitosis 1I) to produce two sperm cells. The tapetum ruptures releasing cytoplasmic lipid bodies and elaioplasts that associate with the pollen to form a coat. This coat is involved in 20 interactions at the stigmatic surface enabling hydration and successful germination of the pollen grain. Both A. thaliana and B. napus belong to the 30% of angiosperms that produce mature pollen in a tricellular state, i.e. the generative 25 cell divides before pollen dehiscence. The generative cell in the 70% of angiosperms producing bicellular pollen undergoes a mitotic division following germination on the stigmatic surface. The cytological processes and time frame of pollen development are relatively consistent within angiosperms despite obvious 30 differences in floral morphology, pollen size and pollen shape that distinguish various genera and families. The conserved nature of pollen development suggests that the processes and genes involved are also likely to be conserved. Examples of conserved pollen- WO 2005/122751 PCT/AU2005/000851 22 specific genes already exist. The gene P2 in Oenothera organensis hybridises to similar transcripts in families representing both monocots and dicots. The pollen specific genes Bcp1 and Bnml appear to be conserved within the family Brassicacea. 5 "Derivatives" of nucleic acid molecules as defined herein encompass those nucleic acid molecules comprising non-naturally occurring nucleotides or those naturally occurring nuclectides that have been modified by chemical or other means. 10 "Derivatives" of proteins or peptides as defined herein encompass those proteins or peptides comprising non-naturally occurring amino acids or those naturally occurring amino acids that have been modified by chemical or other means. 15 "Derivatives" as used hereon in relation to nucleic acid molecules, proteins and peptides are also intended to encompass single or multiple nucleotide or amino acid substitutions, deletions and/or additions as well as parts, fragments, portions, homologues and 20 analogues of the nucleic acid molecule or protein or peptide. Functional derivatives as defined herein encompass nucleic acid molecules, proteins and peptides that are derivatives of the wild type MYB nucleic acid molecules, proteins or peptides but that 25 still retain the wild type function of being expressed in the anther during pollen formation. Wild type as defined herein refers to the MYB gene, protein or peptide as found in nature. 30 "Under normal conditions" as used herein is intended to refer to conditions where the inhibiting nucleic acid molecule is not WO 2005/122751 PCT/AU2005/000851 23 present, i.e. in a plant cell not transformed with the nucleic acid molecule according to the second or third aspects of the invention. Complementary as used herein in relation to nucleic acid molecule 5 "complementary" to the B. napus nucleic acid sequence or a region thereof, or a functional derivative thereof, is intended to encompass those sequences that are capable of hybridising under high stringency conditions at 42 degrees centigrade to the nucleic acid molecules defined according to the second or third aspects of 10 the invention. "Stringency" of hybridization reactions is readily determinable by one of ordinary skill in the art, and generally is an empirical calculation dependent upon probe length, washing temperature, and 15 salt concentration. In general, longer probes require higher temperatures for proper annealing, while shorter probes need lower temperatures. Hybridization generally depends on the ability of denatured DNA to re-anneal when complementary strands are present in an environment below their melting temperature. The higher the 20 degree of desired homology between the probe and hybridizable sequence, the higher the relative temperature which can be used. As a result, it follows that higher relative temperatures would tend to make the reaction conditions more stringent, while lower temperatures less so. For additional details and explanation of 25 stringency of hybridization reactions, see Ausubel et al., Current Protocols in Molecular Biology, Wiley Interscience Publishers, (1994) and Sambrook et al., Molecular Cloning, A Laboratory Manual, 2nd Edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY). 30 Reference herein to a "region" of the B. napus nucleic acid molecule encompasses a fragment of the B. napus nucleic acid molecule which is expressed in anther tissue during pollen WO 2005/122751 PCT/AU2005/000851 24 formation and encodes a functional region of an MYB transcription factor. The invention provides specific nucleic acid molecules BnMYB32, 5 BnMYl-03-l and BnMYB103-2 which encode MYB transcription factors and are expressed in anthers during pollen formation. The inventors provide evidence herein to show that BnMYB103-1 is expressed during early pollen development. 10 Experiments performed to ascertain the expression pattern of BnMYBl03-1 show that it is expressed most strongly in flower buds 0.5-2.0 mm in length. These bud lengths correlate with specific stages of early pollen development. 15 BnMYB32 is expressed in developing anthers, its expression commencing in immature anthers and peaking in mature anthers, i.e. from the onset of pollen meiosis to a peak at the conclusion of meiosis. BnMYB32 is also weakly expressed in developing lateral 20 roots and in emerging root tips. In very young flowers anther primordia are composed of undifferentiated meristematic cells. These cells form a sporogenous cell mass, which will become distinguishable from the 25 tapetum, middle layer, endothecium and epidermis. Archesporial cells align themselves in the anther and divide mitotically giving rise to the tapetum and meiocytes. 0.5mm bud length 30 At a bud length of 0.5mm the meiocytes undergo meiosis, an event that occurs synchronously in all locules within the bud. A second mitotic division produces a tetrad of microspores that secrete a P(1,3)-glucan (callose) wall. Mutants in which there is premature WO 2005/122751 PCT/AU2005/000851 25 dissolution of this wall are usually male sterile, indicating the importance of callose deposition to the production of viable pollen. The tapetal cells undergo mitosis without cytokinesis producing tetraploid binucleate cells. 5 1.0mm bud length At 1mm, meiosis is complete and the new microspores form a tetrad enclosed by a callose wall. The tapetal cells increase in size and volume associated with the proliferation of endoplasmic reticulum 10 and Golgi. Vesicles are present, small vacuoles develop and sporadic lipid bodies are observed. The inner tapetal wall is in contact with the tetrad wall and the tapetum loses its tangential walls. 15 1.5mm bud length In 1.5mm buds a cellulosic primexine develops outside the plasmalemma of young microspores. This provides a matrix for the deposition of sporopollenin, the precursors of which are formed within the microspore cytoplasm and secreted outside the cell 20 between the plasma membrane and callose wall. This process begins the formation of the exine, a structure that gives the mature pollen grain its characteristic sculptured appearance. The callose wall is digested by (1,3)-glucanase (callase), which is secreted into the locule from the tapetum and releases the microspores. 25 Vacuolation of the microspores is initiated with numerous small vacuoles present. The tapetum secretes fibrillar and small osmiophillic vesicles. 2.0mm bud length 30 At a bud length of 2mm the unicellular microspores are characterised by the formation of a single large vacuole in the cytoplasm. Sporopollenin deposition continues until exine formation is complete. The radial cell walls of the tapetum are WO 2005/122751 PCT/AU2005/000851 26 lost and it has started its final degeneration. Starch granules appear transiently in the tapetal plastids. 2.5mm bud length 5 In 2.5mm buds an important asymmetric mitotic division occurs producing bicellular microspores containing a large vegetative cell and a smaller generative cell. The tapetal cells continue to secrete nutrients into the locule but the cellular organelles exhibit characteristics of degradation. Many of the important 10 developmental processes are complete by this stage as evidenced by the fact that meiotically dividing meiocytes can be removed from the anther and will continue development in vitro . The fact that BnMYB103-1, BnMYB103-2 and BnMYB32 expression is 15 strongest in flower buds 0.5-2.0mm in length has implications for its role in regulating genes expressed during these early stages of pollen development. Various other anther expressed genes have been identified in A. thaliana and Brassica sp. Function has not yet been assigned to most of these genes and many other genes involved 20 in this process are likely to be identified in the future. However, genes involved in meiotic division, the synthesis of a variety of compounds found in the pollen wall, functions of the tapetum and associated processes are all required during the period in which BnMYB103-1, BnMYB103-2 and BnMYB32 are expressed. The 25 regulated expression of these genes will be necessary and transcriptional regulators such as BnMYB103-1 , BniMYB103-2 and BnMYB32 will play an important role. As well as providing isolated nulciec acid molecules involved in 30 pollen formation in B. napus, the invention also provides othologues in other plant species and encompasses homologues.

WO 2005/122751 PCT/AU2005/000851 27 A "homologue" is defined as a nucleic acid molecule sharing the same function as another nucleic acid molecule. Homologues are generally determined by sequence similarity as defined by alignment using algorithms such as that in the Advanced BLAST2 service 5 provides by EMBL. Homologous sequences are generally those with a percentage sequence identity of at least 50% at nucleotide or amino acid level according to BLAST analysis. Similarities of at least 60%, 70%, 10 80% and 90% that are functionally active are said to be homologous sequences. "Percent (%) sequence identity" with respect to the nucleic acid sequences identified herein is defined as the percentage of 15 nucleotides in a candidate sequence that are identical with the nucleotides in the specific MYB nucleotide sequence, after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity, and not considering any conservative substitutions as part of the sequence identity. 20 Alignment for purposes of determining percent nucleotide sequence identity can be achieved in various ways that are within the skill in the art, for instance, using publicly available computer software such as BLAST, BLAST-2, ALIGN or Megalign (DNASTAR) software. Those skilled in the art can determine appropriate 25 parameters for measuring alignment, including any algorithms needed to achieve maximal alignment over the full length of the sequences being compared. For purposes herein, however, % nucleotide sequence identity values are generated using the WU-BLAST-2 computer program (Altschul et al., Methods in Enzymology 266:460 30 480 (1996)). Most of the WU-BLAST-2 search parameters are set to the default values. Those not set to default values, i.e., the adjustable parameters, are set with the following values: overlap span = 1, overlap fraction = 0.125, word threshold (T) = 11, and WO 2005/122751 PCT/AU2005/000851 28 scoring matrix = BLOSUM62. For purposes herein, a % nucleotide sequence identity value is determined by dividing (a) the number of matching identical nucleic acid residues between the nucleotide sequence of the MYB gene of interest having a sequence derived from 5 the MYB gene and the comparison nucleotide sequence of interest (i.e., the sequence against which the MYB sequence of interest is being compared which may be a MYB variant) as determined by WU BLAST-2 by (b) the total number of nucleotides of the MYB gene of interest. 10 Percent nucleic acid sequence identity may also be determined using the sequence comparison program NCBI-BLAST2 (Altschul et al., Nucleic Acids Res. 25:3389-3402 (1997)). The NCBI-BLAST2 sequence comparison program may be downloaded from 15 http://www.ncbi.nlm.nih.gov. NCBI-BLAST2 uses several search parameters, wherein all of those search parameters are set to default values including, for example, unmask = yes, strand = all, expected occurrences = 10, minimum low complexity length = 15/5, multi-pass e-value = 0.01, constant for multi-pass = 25, drop-off 20 for final gapped alignment = 25 and scoring matrix = BLOSUM62. "MYB homologue" or "MYB variant nucleic acid sequence" means a nucleic acid molecule which encodes an active MYB polypeptide as defined below and which has at least about 50% nucleic acid 25 sequence identity with a nucleotide acid sequence encoding a full length native sequence MYB polypeptide sequence as disclosed herein, or any fragment of a full-length MYB polypeptide sequence as disclosed herein. Ordinarily, a MYB variant polynucleotide will have at least about 50% nucleic acid sequence identity, more 30 preferably at least about 51% nucleic acid sequence identity, more preferably at least about 52% nucleic acid sequence identity, more preferably at least about 53% nucleic acid sequence identity, more preferably at least about 54% nucleic acid sequence identity, more WO 2005/122751 PCT/AU2005/000851 29 preferably at least about 55% nucleic acid sequence identity, more preferably at least about 56% nucleic acid sequence identity, more preferably at least about 57% nucleic acid sequence identity, more preferably at least about 58% nucleic acid sequence identity, more 5 preferably at least about 59% nucleic acid sequence identity, more preferably at least about 60% nucleic acid sequence identity, more preferably at least about 61% nucleic acid sequence identity, more preferably at least about 62% nucleic acid sequence identity, more preferably at least about 63% nucleic acid sequence identity, more 10 preferably at least about 64% nucleic acid sequence identity, more preferably at least about 65% nucleic acid sequence identity, more preferably at least about 66% nucleic acid sequence identity, more preferably at least about 67% nucleic acid sequence identity, more preferably at least about 68% nucleic acid sequence identity, more 15 preferably at least about 69% nucleic acid sequence identity, more preferably at least about 70% nucleic acid sequence identity, more preferably at least about 71% nucleic acid sequence identity, more preferably at least about 72% nucleic acid sequence identity, more preferably at least about 73% nucleic acid sequence identity, more 20 preferably at least about 74% nucleic acid sequence identity, more preferably at least about 75% nucleic acid sequence identity, more preferably at least about 76% nucleic acid sequence identity, more preferably at least about 77% nucleic acid sequence identity, more preferably at least about 78% nucleic acid sequence identity, more 25 preferably at least about 79% nucleic acid sequence identity, more preferably at least about 80% nucleic acid sequence identity, more preferably at least about 81% nucleic acid sequence identity, more preferably at least about 82% nucleic acid sequence identity, more preferably at least about 83% nucleic acid sequence identity, more 30 preferably at least about 84% nucleic acid sequence identity, more preferably at least about 85% nucleic acid sequence identity, more preferably at least about 86% nucleic acid sequence identity, more preferably at least about 87% nucleic acid sequence identity, more WO 2005/122751 PCT/AU2005/000851 30 preferably at least about 88% nucleic acid sequence identity, more preferably at least about 89% nucleic acid sequence identity, more preferably at least about 90% nucleic acid sequence identity, more preferably at least about 91% nucleic acid sequence identity, more 5 preferably at least about 92% nucleic acid sequence identity, more preferably at least about 93% nucleic acid sequence identity, more preferably at least about 94% nucleic acid sequence identity, more preferably at least about 95% nucleic acid sequence identity, more preferably at least about 96% nucleic acid sequence identity, more 10 preferably at least about 97% nucleic acid sequence identity, more preferably at least about 98% nucleic acid sequence identity and yet more preferably at least about 99% nucleic acid sequence identity with the nucleic acid sequence encoding a full-length native sequence MYB polypeptide sequence as disclosed herein, or 15 other fragments of a full-length MYB polypeptide sequence as disclosed herein. Variants do not encompass the native nucleotide sequence. Ordinarily, MYB variant polynucleotides are at least about 30 20 nucleotides in length, often at least about 60 nucleotides in length, more often at least about 90 nucleotides in length, more often at least about 120 nucleotides in length, more often at least about 150 nucleotides in length, more often at least about 180 nucleotides in length, more often at least about 210 nucleotides in 25 length, more often at least about 240 nucleotides in length, more often at least about 270 nucleotides in length, more often at least about 300 nucleotides in length, more often at least about 450 nucleotides in length, more often at least about 600 nucleotides in length, more often at least about 900 nucleotides in length, or 30 more.

WO 2005/122751 PCT/AU2005/000851 31 An "orthologue" may be defined as genes or gene products which are derived from a common ancestor and share a common function. Orthologues are generally homologues. 5 A study of collinearity between regions of the A. thaliana and B. napus genomes has shown that genes present as a single copy in Arabidopsis are present in between two to eight copies in B. napus. Southern blot data reveals four hybridising bands in the B. napus genome when probed with AtMY103. Southern blotting of the B. 10 oleracea and B. rapa genomes with the same probe shows two hybridising bands in each species. This result is consistent with the AACC genome of B. napus and the collinearity data available. It is likely that the four clones obtained in this study represent two genes from each of the parental species that have come together 15 in the B. napus genome. Studies of the A. thaliana and B. napus genomes provide important insights into the relationship between these species. Genes that are orthologous between these species share an average of 87% DNA 20 sequence identity and extensive conservation of marker order has been found. Work done to date suggests that Brassica genomes contain three representations of a basic genome with each representation being extensively collinear with the genome of A. thaliana. 25 A practical example of gene conservation between these species is the senescence specific gene SAG12. This gene which codes for a cysteine protease in A. thaliana was used to probe a B. napus genomic DNA library and resulted in the isolation of two 30 homologues. These genes show between 83-86% nucleotide sequence identity with SAG12. The similarities in gene structure and expression patterns of these genes indicate they are orthologues.

WO 2005/122751 PCT/AU2005/000851 32 The ultimate test of functional homology between the two MYB genes would be the rescue of an AtMYB103 mutant by transformation with BnMYB103-1. It is highly likely that AtMYBl03 and BnMY1303-1 share a common function in the regulation of early pollen development and 5 as such represent orthologues. In this context "heterologous" DNA means that the DNA introduced into the cells is a DNA not naturally occurring in the cells in this form. On the one hand, it may be DNA which does naturally not 10 at all occur in these transformed cells or DNA which, even if it does occur in these cells, is integrated at other genetic positions as exogenous DNA and is therefore situated within another genetic environment. 15 Comparison of the AtMYBI3 and BnMYBl03-1 sequences show a number of similarities that indicate these two genes are likely to be orthologues. Both genes consist of three exons interrupted by two introns at conserved sites and both have an open reading frame of 960bp coding for a protein of 320aa. There is 90% identity between 20 the two sequences at the nucleotide level and 93% at the amino acid level. Within the R2R3 repeat region there is 99% identity at the amino acid level. In the first 400bp of the non-coding region upstream of the ATG start codon there is 74% identity. In Arabidopsis it has been shown that this 400bp promoter region is 25 responsible for the correct spatial and temporal expression of AtMYB1 03. Other data support the likelihood that these two genes are orthologues, including promoter gusA experiments where 30 AtMYB103::gusA and BnMYB103-1::gusA constructs transformed into A. thaliana both show identical staining patterns. GUS activity is evident in the anthers of flower buds from 0.5-3.0mm long. Cross sections through stained anthers show that GUS activity is confined WO 2005/122751 PCT/AU2005/000851 33 to the tapetum and developing microspores. We show herein that antisense and sense BnMYB1O3-1 constructs under the control of a tapetum-specific promoter (TA39 from tobacco), produce an abnormal pollen phenotype in A. thaliana. The abnormalities observed are 5 similar to those seen in B. napus plants transformed with the same constructs, indicating that functional homology exists between these two genes. A search of the literature indicates that this is not an isolated 10 case. Examples exist where antisense and sense technology has been used to down regulate the expression of an endogenous gene in a temporally and spatially specific manner. This has been achieved using an homologous gene from another species under the control of a non-native promoter. The tapetum and microspore expressed 15 Brassica gene Bcpl shows 73% identity to its Arabidopsis homologue. A 0.5kb fragment of the Brassica Bcpl cDNA was used in an antisense construct under the control of a promoter from a pollen-expressed gene of tomato, LAT52. Transformation of this construct into A. thaliana was found to inhibit expression of the endogenous gene. 20 The results showed that the Brassica Bcpl antisense cDNA was able to inhibit function of the Arabidopsis Bcpl homologue in haploid microspores, leading to pollen abortion. During phenylpropanoid biosynthesis the first step is catalysed by 25 phenylalanine ammonia-lyase (PAL), an enzyme active in the tapetum during pollen development. Antisense and sense constructs of PAL cDNA from sweet potato under the control of a tapetum-specific promoter from rice, Osg6B, have been transformed into tobacco. Analysis of anthers from transgenic plants showed a reduction in 30 PAL activity, which was accompanied by a partial male sterility phenotype. Another example involves the use of a genomic DNA fragment for the alternative oxidase gene from A. thaliana. Expression of this gene has been demonstrated in the tapetum and WO 2005/122751 PCT/AU2005/000851 34 pollen during anther development. An antisense construct of this fragment under the control of the tapetum-specific rice promoter Osg6B was transformed into tobacco. Expression of the alternative oxidase gene in transgenic plants was much lower than in wild type 5 and resulted in partial male sterility. The ultimate test of functional homology between the two MYB genes would be the rescue of an AtMYB103 mutant by transformation with BnMYB1l03-1. It is highly likely that AtMYBl03 and BnMYB103-1 share 10 a common function in the regulation of early pollen development and as such represent orthologues. The Advanced BLAST2 and Orthologue Search Service at EMBL was used to find sequences likely to represent orthologues of BnMYBl03-1. 15 Using the BnMY13l03-1 amino acid sequence as the query sequence the highest score of 1449 3.6e 48 was obtained for AtMYB103. The other genes shown in this analysis are all recognised MYB genes or described as MYB-like, putative MYBs or putative transcription factors. Further analysis with this service produced a species 20 tree (Figure 8) using the NCBI database and a gene tree (Figure 9) based on the neighbour-joining tree computed By CLUSTALW. The species tree is a phylogenetic tree representing evolutionary relationships between the species under consideration. Genes from several species were represented in this analysis and most grouped 25 along family lines, Lycopersicon esculentum and Petunia hybrida from the Solanaceae, Arabidopsis thaliana and Brassica napus from the Brassicaceae and Oryza sativa, Zea mays, and Hordeum vulgare from the Poaceae. However, the grouping of Antirrhinum majus, Gossypium hirsutum and Pimpinella brachycarpa does not reflect a 30 familial relationship. Unsurprisingly, the moss Physcomitrella patens does not group with any of these flowering plants. The sorting of genes according to this taxonomic tree groups BnMYBl03-1 with thirty-one A. thaliana sequences. Fifteen of these genes are WO 2005/122751 PCT/AU2005/000851 35 MYBs, including AtMY13l03, the others are described as MYB-related or transcription factor-like proteins. The inventors have isolated orthologues of the B. napus MYB genes 5 from an A. thaliana cDNA library and these genes have been characterised: AtMYB32 This gene is expressed in developing anthers from the onset of pollen meiosis and expression reaches its peak at the 10 conclusion of meiosis. AtMYB32 is also expressed in developing lateral roots and in emerging root tips. Antisense and cosuppression AtMYB32 transgenic plants exhibit an abnormal pollen phenotype. 15 AtMYBl03 Expression studies using a promoter::GUS construct suggest this gene is specifically expressed in developing flowers at a stage where the pollen mother cells separate and undergo meiosis to form a tetrad of microspores. GUS activity is highest during the tetrad stage and was expressed in the tapetum and middle 20 layer of the anther and in microspores. The GUS activity results have been confirmed with in situ hybridisation experiments [Li et al., 1999 supra]. As AtMYB32 and AtMYBl03-l are expressed at the time of important 25 processes in the anther and pollen indicates that these genes are likely to be involved in regulating other genes required for pollen development. Thus, the isolation of the BNMYB32 and the BnMYB103-l and BnMYBl03 30 2 genes and their homologues and orthologues of the present invention provides a strategy for pollen-control. The gene BnMYBI03-1 is exclusively active in developing anthers, whereas the gene BnMYB32 is active during lateral root initiation, as well as WO 2005/122751 PCT/AU2005/000851 36 being expressed (turned on) in a number of tissues of the developing anther, including the tapetum (a tissue providing nutrients for pollen development as well as pollen-wall components and enzymes for microspore release from the enclosing callose wall) 5 and the microspores themselves. Expression of the BnMYBl03-l and BnMYB103-2 genes and the BnMYB32 gene occurs only during the early stages of pollen development, the genes subsequently becoming "silent". 10 Although the precise functions of the genes are not known, they code for proteins that belong to a family of proteins (MYB-like) that bind to specific DNA sequences. Such MYB-like proteins regulate the activity of other genes. Two MYB-like proteins in maize, for example, control anthocyanin synthesis. The inventors' 15 results indicate that the MYB103 gene(s) and the MYB32 gene regulate genes important in pollen development. Experiments in which antisense MYBl03 gene(s) and MYB32 gene constructs were placed under control of the Cauliflower Mosaic 20 Virus 35S promoter and introduced into A. thaliana plants, resulted in plants with defective pollen. The inventors have found that their MYB-like genes, express in the anther during pollen development are conserved between species. 25 Consequently, the antisense approach using the A. thaliana or B. napus genes can be expected to function in crops both closely related and distinct (e.g. wheat). Moreover, the introduced (antisense) MYBl03 gene(s) and MYB32 gene 30 can be genetically engineered so that the genes are expressed at high levels in anthers and the expression can be induced by simple chemical treatment of plants.

WO 2005/122751 PCT/AU2005/000851 37 Thus, anti-sense constructs of the specified gene(s) of the invention can be utilized to put the invention into practice by being integrated into the genome of a number of relevant transformable plant species (e.g. canola, tomato, grain legumes, 5 cereals) with anthers and pollen developed therefrom. The inventors propose that the MYB-like genes of the present invention may be used in conjunction with other genes that are involved in male-specific development, such as those encoding 10 putative transcription factors NOZZLE, MS1 and MYC class transcription factor (see for example Sorensen, A-M et al., The Plant Journal (2003) 33, 413-423). The use of such genes is likely to improve the percentage efficiency of the induction of male sterility. 15 "Antisense nucleic acid molecules" as described herein defines sequences that are complementary to the nucleic acid molecules of the first aspect of the invention or part thereof. Such antisense nucleic acid molecules, may bind to the endogenous gene and block 20 prevent expression of the functional gene in a plant cell. Antisense techniques generally use short 10 to 20 oligonucleotide fragments which hybridise to essential parts of the gene thereby blocking its expression. Such essential regions of the gene may include regions within the 5' regulatory region such as enhancer 25 and promoter regions and may also include the transcription start site. By "promoter" is meant a minimal sequence sufficient to direct transcription. "Promoter" is also meant to encompass those promoter 30 elements sufficient for promoter-dependent gene expression controllable for cell-type specific, tissue-specific or inducible by external signals or agents; such elements may be located in the 5' or 3' regions of the native gene.

WO 2005/122751 PCT/AU2005/000851 38 A promoter may be constitutive but most preferably is inducible. The construct may contain one or more promoters. 5 Plasmids, phage and vectors are all terms to describe carriers for introducing nucleic acid into cells. There are a number of methods available for introducing foreign nucleic acid molecules into plant cells and it is not intended that the invention is restricted to any one of these carriers. The most generally used transformation 10 methods include particle bombardment, polyethylene-mediated transformation and microinjection. The particular method chosen to obtain transformed plants containing nuclear genomes with the inserted nucleic acid will depend on the plant species. 15 In practice, any nucleic acid used to transform plant cells will be in the form of a nucleic acid construct. In practice, a construct used to transfect the plant nucleus will generally additionally comprise various control elements. Such 20 control elements may include promoters and optionally a ribosome binding site (RBS), positioned at an appropriate distance upstream of a translation initiation codon to ensure efficient translation initiation. 25 Constructs envisaged according to the present invention include those constructs comprising an anther specific promoter and at least one gene coding for an MYB transcription factor, which gene is expressed in the anther during pollen formation. 30 A person skilled in the art will be readily able to determine suitable constructs. For example we provide a construct comprising an AtMYB103 gene under the control of the TA39 promoter, the AtMYB103 gene being provided in either the sense or antisense WO 2005/122751 PCT/AU2005/000851 39 direction, i.e. 5'to 3' or 3' to 5'. We also provide a construct comprising the BnMYBIO3-1 and BnMYBl03-2 genes lined together, and an AtMYB-32 gene, each gene in either the sense or antisense direction and all genes all under the control of an anther specific 5 promoter. Preferably most or all of the constituents of the construct are operably linked. 10 Due to the homology between the genes in different species the inventors propose that MYB genes from one species may be used in constructs used to transform other species and that species specific genes be used in constructs with orthologues from other species, for use in transforming any plant species. 15 For example a construct comprising BnMYB.lO3-1 and BnMYBl03-2 genes may also include an AtMYB32 gene and be used to transform Brassica species under the control of an anther specific promoter such as TA39 from tobacco. 20 A preferred construct further comprises GVG, a chimeric transcription factor which binds to and activates the 6UAS promoter in the presence of dexamethasone (see Aoyama, T. and Chua, N.H., The Plant Journal (1997) 11(3), 605-612). The construct may also 25 comprise a selectable marker, for example HPT, encoding hygromycine phosphotransferase, to allow for selection of transgenic plants comprising the construct. A further preferred construct further comprises a NOZZLE, MS1 or 30 MYC gene, or other nucleic acid encoding proteins involved in male specific development (see for example Sorensen, A-M, et al., supra) WO 2005/122751 PCT/AU2005/000851 40 The construct may incorporate into the plant nuclear genome through recombination events. An anther specific promoter defines any promoter which targets the 5 construct to the anther in preference to other positions. A number of regulatory sequences from the promoter region of anther or pollen-specific genes have been identified using promoter deletion analysis. Minimal promoter regions required for tissue 10 specific gene expression have been described (Twell et al., (1991) Genes & Development 5: 496-507). The chiA promoter from petunia contains the CCACAAAAA motif (van Tunen et al., (1990) Plant Cell 2: 393-401) while the Bp19 gene from Canola contains the motif TGACG that is also present in the nos and 35S promoters (Albani et 15 al., (1991) Plant Molecular Biology 16: 501-513. The A9 gene from Arabidopsis contains a TATATATA motif in the promoter region thought to direct tapetum-specific expression (Paul et al., (1992) Plant Molecular Biology 19: 611-622). The 52/56 box TGTGGTTATATA and the 56/59 box TGAAATTGTGA are conserved sequence motifs in the 20 tomato LAT genes (Eyal et al., (1995) Plant Cell 7: 373-384). The pollen-specific a-tubulin gene from Arabidopsis also contains sequences similar to the 56/59 box (Carpenter et al., (1992) Plant Cell 4: 557-571). The AAATGA sequence directs pollen specificity in alfalfa promoters (Wu et al., (1998) Sexual Plant Reproduction 11: 25 181-182). Sequence similarities among tissue specific promoters are restricted to short sequence motifs. Promoters may share a similar sequence but are also influenced by upstream regulatory elements that influence expression levels (Twell et al., 1991 supra). 30 The tobacco TA39 promoter regulates tissue-specific gene expression in the anther tissue (Goldberg et al., (1993) Plant Cell 5: 1217 1229. This promoter is expressed in the microspheres, connective tissue and most highly in the tapetum during early pollen WO 2005/122751 PCT/AU2005/000851 41 development. The TA39 promoter is widely used in gene constructs to control the expression of transgenes directed to the tapetum tissue in the anther and is a preferred promoter used in the constructs of the present invention. Another preferred anther 5 specific promoter is the AT anther specific promoter described in our co-pending application entitled "Promoter and Uses Thereof". The anther specific promoter is preferably a constitutive but tissue specific promoter. 10 The construct may also comprise an inducible promoter so that expression of the construct may be achieved at a desired time point. Aoyama, T. and Chua, N.H. The Plant Journal (1997) 11(3), 605-612 describe a chemical induction system for transcription in 15 plants. This system utilises a chimeric transcription factor, GVG, consisting of the DNA-binding domain or the yeast transcriptional factor GAL4, the transactivating domain -of the herpes viral protein VP16 and the receptor domain of the rat glucocorticoid receptor (GR). The promoter is induced in response to dexamethsone (DEX), a 20 strong synthetic glucocorticoid. Once activated by dexamethasone, GVG can induce the transcription from the chimeric promoter by about 100 fold. This system, and other inducible promoter systems, may be used in 25 the constructs of the present invention. The construct may alternatively or in addition comprise a repressor/restored system (as described below) to allow control of expression of the genes in the construct. 30 The present invention provides transformed cells comprising a nucleic acid molecule or fragment thereof according to a first aspect of the invention or a construct according to the second aspect of the invention. By means of methods known to the skilled WO 2005/122751 PCT/AU2005/000851 42 person the transgenic plant cells can be regenerated to whole plants. Thus, the plants obtained by regenerating the transgenic plant cells of the invention are also the subject-matter of the present invention. A further subject-matter of the invention are 5 plants which contain the above-described transgenic plant cells. The transgenic plants may in principle be plants of any desired species as previously defined. The invention also relates to propagation material of the plants of 10 the invention, e.g. fruits, seeds, tubers, root-stocks, seedlings, cuttings etc. The plants cells may be transformed with a first construct comprising one or more MYB genes and then be subsequently 15 transformed with a further construct comprising one or more same or other MYB genes or fragments thereof or another gene involved in male-specific development, it is not necessary that all the genes to be introduced on the cell be provided on the same construct. In other words, rather than transforming the genome with a single 20 construct comprising at least two MYB genes and optionally another gene, the genes may be introduced on separate carriers. When two nucleic acid sequences are used they are preferably introduced together by co-transformation. 25 The invention in a fourth aspect provides method for controlling pollen development in a plant by introducing into the plant means of inhibiting the expression of the endogenous MYB gene expressed in the anther during pollen development. 30 Based on the nucleic acid sequences provides persons skilled in the art would be able to'design suitable means to substantially block expression of the endogenous MYB gene.

WO 2005/122751 PCT/AU2005/000851 43 An "endogenous gene" is a gene that is naturally present in a cell. The endogenous MYB gene as used herein describes the MYB gene that would be naturally expressed in a plant cell under normal 5 conditions. The production of plant cells which lack the expression of one or more endogenous MYB genes may for example be achieved by the expression of a corresponding antisense-RNA, of a sense-RNA for 10 achieving a co-suppression effect or the expression of a correspondingly constructed ribozyme, which specifically cleaves transcripts encoding one of the proteins of the invention, using the nucleic acid molecules of the invention or the AtMYB103 gene. 15 In order to block expression of endogenous gene(s) according to the first aspect of the invention, antisense-RNA is preferably expressed in plant cells and the technique known as RNA interference (mIAi) is used. (Waterhouse et al., (1998) Proc. Natl. Acad. Sci. USA 95, 13959-13964). 20 In brief canola plants expressing the sense sequence of BnMYBl03-l are crossed with plants expressing the same BnMYB103-1 sequence in antisense orientation. F1 plants exhibited pollen abnormality and some male sterility. The F1 phenotypes were more severe than 25 either parent. It is likely that the sense and antisense transcripts of the EnMYBl03-l transgenes formed double-stranded RNA molecules leading to the degradation of endogenous BnMYB103-1 transcript. The RNAi system using double stranded RNAs to block gene expression is widely used. 30 In order to express an antisense-RNA, on the one hand DNA molecules can be used which comprise the MYB gene(s) complete sequence, including possibly existing flanking sequences as well as DNA WO 2005/122751 PCT/AU2005/000851 44 molecules, which only comprise parts of the coding region. These parts have to be long enough in order to prompt an antisense-effect within the cells. Basically, sequences with a minimum length of 15 bp, preferably with a length of 100-500 bp and for an efficient 5 antisense-inhibition, in particular sequences with a length of more than 500 bp may be used. Generally DNA-molecules are used which are shorter than 5000 bp, preferably sequences with a length of less than 2500 bp. 10 Use may also be made of nucleic acid sequences which are highly homologous, but not completely identical to the sequences of the nucleic acid molecules of the invention. The minimal homology should be more than about 65%. Preferably, use should be made of sequences with homologies between 95 and 100%. 15 In an embodiment the cells of the invention differ from naturally occurring cells in that they contain a heterologous recombinant DNA molecule encoding an antisense RNA, a ribozyme or a cosuppression RNA. Due to the expression of this heterologous recombinant DNA 20 molecule the expression of the MYB gene(s) is prevented, resulting in lack of production of the MYB transcription factors in the anther. As the MYB genes are implicated in the formation and production of pollen, preventing expression of the MYB nucleic acid molecules according to the present invention provides transgenic 25 plants which do not produce pollen. Preferably the method comprises the use of a system that controls the expression of the MYB gene(s) and allows expression of the gene(s), and hence the production of pollen, to be turned on and 30 off as desired. In one embodiment the system is an inducible system. The inducible system may involve induction of the GVG promoter in a transgenic WO 2005/122751 PCT/AU2005/000851 45 plant with dexamethasone or some other inducible systems may be used. It is preferred that the system be inducible so that production of male sterile plants may be reversible. The method of choice would involve the use of a cheap, non-toxic inducer molecule 5 that can be sprayed on plants to activate antisense gene(s) and block pollen development. It is preferred that in the absence of inducer, the transgenic plants will produce normal pollen. Another system for controlling expression of a gene of interest, 10 which may be used in the present invention but may also find utility outside the constraints of the present invention, is a repressor/restorer system. In this system the functioning of an essential gene is repressed using a transcription repressor. 15 Transcription repressors use various repressive motifs to block the transcription of their target genes. One repressor motif suitable for use in plants has the amino acid sequence LDLDLELRLGFA (SEQ ID NO. 27) (Hiratsu, K. et al., (2003) Plant J. 34, 733. Another suitable repressor for use in plants has the amino acid sequence 20 LDLNLELRISPP (SEQ ID NO. 28) (Jin, H. et al., (2000) EMBO J 15, 6150). AtMYB103 was chosen to test the inducible repressor system in plants. This gene was a suitable candidate as it is strongly 25 expressed in Arabidopsis and encodes a transcription factor, an essential factor in the development of viable pollen. In the system it was proposed that functioning of the AtMYB103 gene could be repressed using a repressor motif, but turned back on (restored) by crossing. 30 The two peptide sequences exhibit nine identical or homologous residues. The repressive motif (LDLDLELRLGFA) (SEQ ID NO. 29) from Arabidopsis was fused to four different transcription WO 2005/122751 PCT/AU2005/000851 46 factors with resultant loss-of-function phenotypes demonstrating that the motif converts transcription factors into strong dominant repressors. Inspection of AtMYB7 (Li, S.F., Parish, R.W. (1995) Plant J. 8, 963) and AtMYB32 C-terminal sequences 5 identified the repressive sequence LDLNL.ELRISPP (SEQ ID NO. 30). The strategy is to test the repressor/restorer system proposed by studying if you could convert AtMYB103 into a strong repressor using the motif of twelve amino acids to obtain male 10 sterility. The heterozygous male sterile inbred A containing the chimeric AtMYB103 repressor (AtMYB103Rep) is crossed with inbred B containing a "restorer" to produce male fertile F1 hybrid seeds. The "restorer" will be AtMYB103 gene under the control of the strong anther-specific At39 promoter or multiple 15 copy of the AtMYB103 promoter or their equivalents. Higher levels of the AtMYB103 protein (activator) in the hybrid plants will alleviate the repression by AtMYB103Rep protein leading to viable pollen production and seed setting. 20 The repressor/restorer system produces 50% male fertile inbred A seedlings. A selectable marker linked to the repressor will provide a selection to cull the male fertile inbred A seedlings before crossing with inbred B. Alternatively, a simple chemical which can overcome the repression by the AtMYB103Rep may be used 25 to induce seed setting of the male sterile plants to obtain the homozygous inbred A seed. The homozygous and male sterile inbred A plants will then be crossed with inbred B plants to obtain Fl hybrid seed. 30 The repressor/restorer reversible male sterility system is being successfully tested in Arabidopsis and is suitable for hybrid seed production in any crop plants including canola and wheat.

WO 2005/122751 PCT/AU2005/000851 47 The repressor/restorer system is the preferred system for use controlling gene expression in the method of the first aspect of the invention because it potentially avoids the need to spray crops to achieve control of expression. 5 By disrupting pollen formation according to the first aspect of the invention there is provided a method for providing male sterile plants, that is transgenic plants which do not produce pollen. The ability to bring about male sterility by the directed expression of 10 genes able to inhibit pollen development or the expression of antisense and co-suppression constructs designed to inhibit genes required for pollen formation in a controlled manner is particularly important. 15 To allow male sterility to be achieved it is important that pollen formation is completely disrupted rather than being merely produced. Blocking one or more of the MYB nucleic acid sequences according to the present invention may provide complete disruption of pollen formation and hence produces 100% male sterile plants. 20 The percentage efficiency of blocking may be improved if, in addition to MYB32 and MYB103, other genes involved in male-specific development are also blocked. With regard to the specific embodiments of the present invention, 25 in Brassica napus blocking expression of the BnMYB103-1 gene alone does not give 100% male sterile plants. It is also necessary to block 1BnMYB32 and probably also BnMY2103-2. It may also be necessary to block MYC and/or other genes involved in male-specific development (see Sorensen, A-M, et al., supra). 30 In Arabidopsis blocking expression of AtMYB32 and AtMYB103 gives 100% male sterile plants.

WO 2005/122751 PCT/AU2005/000851 48 It is proposed that because of the extensive homology of the MYB genes of the present invention between species that the sequence used to block expression of an MYB gene in one species could be derived.from an MYB gene from anther species. For example a 5 construct comprising the BnMYB103-1 and BnMYBl03-2 genes and the AtMYB32 gene was shown to produce male sterile plants in B. napus. It is proposed that the Brassica and Arabidopsis MYB genes according to the present invention could be used to produce male sterility in other plant species, including wheat, barley etc. 10 Although such methods may have utility it is especially preferred that the MYB genes used are species specific, particularly to appease any problems perceived in relation to genetically modified organisms. 15 For the purposes of this specification it will be clearly understood that the word "comprising" means "including but not limited to", and that the word "comprises" has a corresponding meaning. 20 It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents forms part of the common general knowledge in the art, in Australia or in any 25 other country. It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit 30 or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

WO 2005/122751 PCT/AU2005/000851 49 Embodiments of the present invention will now be described in the following non-limited examples. EXAMPLES 5 EXAMPLE 1 - Genes from Arabidopsis expressed in the anther during pollen development a) AtMYB32 10 Isolation of Genomic Clone An Arabidopsis thaliana (Landsberg erecta) genomic library in the vector EMBL3 was screened with a 38-mer oligonucleotide directed against the conserved region of the third repeat of the mammalian MYB DNA-binding domain. 15 Screening procedures were carried out according to Ausubel et al. supra. Hybridisation was performed at 40 0 C. with the 32 P-labelled 38-mer probe 38mer probe: CCTGGTCGTACTGA(C/T)AA(C/T)GA(A/G)ATTAA(A/G)AA(C/T)TA(C/T) TGGAA (SEQ ID NO. 31), followed by washes of 6x SSC at 50 0 C. Positive clones 20 were isolated and subjected to Southern analysis by digesting with restriction enzymes BamHI, EcoRI or Sa/I and probing with the 32 P-labelled 38-mer. A 3.5kb EcoRI-BamHI fragment and a 5.Okb BamHI-Sa/I fragment were isolated and cloned into the vector pTZ18U (Bio-Rad). These clones were further subcloned and sequenced. 25 Plant Transformation The following procedures were principally performed as outlined by Ausubel et al. supra unless otherwise specified. A Bbsl restriction fragment of approximately 1000 bp from the 3.5kb EcoRI-BamHI clone 30 comprising the 5' promoter region of AtMYB32 was blunt ended with T4 DNA Polymerase (Promega) and cloned into the Smal site of pBluescript (Strategene). The 1000 bp 5' promoter fragment in pBluescript was excised with BamHI and EcoRV and cloned into the WO 2005/122751 PCT/AU2005/000851 50 BamHI and Smal sites of pBI101.3 (Clontech). This construct was transformed into Agrobacterium tumefaciens strain AGL1 using the method outlined by Ditta et al., (1980) PNAS 77: 7347-7351. Agrobacterium harbouring the construct were used to transform 5 Arabidopsis root explants using the methods described by Valvekens et al., (1988) PNAS 85: 5536-5540. Transformed cells were selected on a shoot inducing media containing kanamycin (Sigma) 100 gg/ml, and subsequently transferred to a growth medium containing kanamycin at 50 gg/ml. 10 Growth of Plants Transgenic seeds were sown onto plates containing germination media and grown under constant illumination at 22 0 C. for approximately four weeks. Plants were then transferred to hydroponic pots and 15 grown in a nutrient solution under the above conditions. After approximately two weeks, auxin (2,4-Dichlorophenoxy-acetic acid) was added to the pots to a concentration of 25 gM. Plants were harvested and screened for GUS activity as outlined below. 20 Polymerase Chain Reaction (PCR) The PCR was carried out according to Klimyuk et al., (1993) Plant J. 3: 493-494, in which whole plant tissue was subjected to an alkali treatment and used directly in the PCR. A primer pair was generated against regions outside the multi-cloning site of pBl101 25 (Clontech). Primers used were 5'TGTGGAATTGTGAGCGGATA (SEQ ID NO. 32) (725-744 of pB1101) and 5'ATTCCACAGTTTTCGCGATC (SEQ ID NO. 33) (929-910 of pBl101). Cycling conditions were 94 0 C. for 30 sec; 58 0 C. for 20 sec; 72 0 C. for 1 min; 36 cycles, followed by a 10 min extension at 72 0 C. PCR products were visualised on a 2.5% agarose 30 gel.

WO 2005/122751 PCT/AU2005/000851 51 GUS Assays and Histochemistry Tissues from transformed plants were assayed for GUS activity by incubating overnight at 37 0 C. in a 0.5 mg/ml X-gluc(5-bromo-4-chloro-3-indoyl-B-D-glucuronide cyclohexylammonium 5 salt) substrate solution. Histochemical analysis was carried out according to Caissard et al., (1992) Protoplasma 170: 68-76 with samples embedded in LR White resin (London Resin Company). Sections of 10 gm were stained with 1% Saffranin 0 and GUS activity observed under a light microscope. 10 Results AtMYB32 clones were isolated and subjected to Southern analysis. Sequencing indicated that an Arabidopsis myb-like gene had been isolated and this was designated AtMYB32 (Fig. 4). The genomic 15 organisation and amino acid sequence of AtMYB32 were determined through sequence comparison with previously reported plant myb-like genes. The AtMYB32 gene possesses one intron which occurs in the R 3 repeat of the DNA- binding domain. The putative AtMYB32 protein product is comprised of 274 residues with an estimated molecular 20 weight of 31.5 kDa. The N-terminal domain is comprised of the two MYB repeats, (R 2 and R 3 ) and displays high sequence similarity to previously reported plant MYB-related proteins. An Arabidopsis thaliana genomic library was screened with a 25 degenerate 38-mer probe 38mer probe: CCTGGTCGTACTGA(C/T)AA(C/T)GA(A/G)ATTAA(A/G)AA(C/T)TA(C/T)TGGAA (SEQ ID NO. 34), directed against a conserved region Three highly conserved tryptophan residues separated by 18-19 amino acids in each of the three c-MYB repeats constitute a hydrophobic core 30 around which three a-helical bundles are positioned in each repeat. Three tryptophan are residues present in the R 2 and two in the R 3 repeat of AtMYB32. As has been found in many plant MYB-related WO 2005/122751 PCT/AU2005/000851 52 proteins, the first tryptophan in R3 is substituted by phenylalanine. The binding capacity of c-MYB is not impaired when this tryptophan is exchanged for an aromatic/hydrophobic residue. The carboxyl region of AtMYB32 shows little similarity with other 5 MYB-like proteins except for the sequences GIDPATH, CLDLNLELRISPP and GLNNTRVLDFSTLEMK. A serine-rich region (10 of 14 amino acids are serine) occurs within the carboxyl terminus. Expression of the AtMYB32 Promoter-Gus Construct 10 To elucidate the expression pattern of AtMYB32, the 5' promoter region was fused to the GUS reporter gene and used to transform Arabidopsis root explants. Fourteen transformant lines were obtained. To confirm the integrity of the constructs in the transgenic plants, PCR was carried out on plant tissue using 15 primers directed against regions outside the cloning site of the binary vector pB1101 (data not shown). GUS activity was initially identified in the anthers during early stages of floral development. Initial GUS expression coincided with stage 9 of Arabidopsis floral development, while maximal GUS levels were, 20 observed at floral stage 10. Hence, initial GUS activity coincided with the onset of pollen mother cell meiosis and peaked at the conclusion of the meiotic stage when the microspores were present as tetrads. GUS activity subsequently declined during later stages. This developmental sequence takes approximately 72 hours 25 (based on the estimates of floral development outlined in Bowman J (ed) Arabidopsis an atlas of morphology and development. Springer Verlag NY (1994)). GUS activity was also observed in emerging lateral roots. GUS 30 activity was associated with the primordia involved in the initiation of lateral roots. As the roots extended, some activity remained in this region and also occurred in the root tip. This is probably because some cells expressing GUS in the primordia are WO 2005/122751 PCT/AU2005/000851 53 destined to form the tip of the emerging root. Subsequently, GUS activity disappeared. Exogenously applied auxin promotes the formation of lateral roots and recent genetic evidence supports such a role for auxin in vivo. Transgenic lines were grown in 5 hydroponic solution and treated with auxin. After treatment times of between 36 and 48 hours, a substantial increase in GUS activity was observed throughout the roots, reflecting both the hormone stimulation of lateral root initiation and new staining within the roots (e.g. primary root apices, vascular tissue). No stimulation 10 of GUS activity was detected in the anthers of auxin-treated plants (data not shown). The levels of AtMYB32 expression in both developing anthers and lateral roots appear to be relatively low. Incubation periods of 15 16-18 hours in X-gluc were needed to obtain adequate staining. However, an 8-10 hour incubation was sufficient to detect activity in auxin-treated roots. Untransformed plants or control transgenic plants transformed with a promoterless GUS construct did not stain (results not shown). 20 Clearly, the presence of AtMYB32 in such divergent tissues as pollen and emerging roots, need not imply that it regulates identical genes in those tissues. Tissue specific MYB partners may be required to form functional transcriptional activators or 25 repressors. b) AtMYB103 AtMYBI03 was isolated from a genomic library of Arabidopsis as 30 described in Li et al., supra which is incorporated herein in its entirety by reference. The nucleotide sequence of AtMYB103 has been deposited in GenBank under accession number AF048839. The gene was shown to be expressed during the early stages of anther WO 2005/122751 PCT/AU2005/000851 54 development and it was proposed, amongst other things, that the gene may regulate pollen development. EXAMPLE 2 - Genes from Brassica expressed in the anther during 5. pollen development Materials and methods carried out in relation to Brassica napus are essentially the same as those carried out for Arabidopsis thaliana as described in detail above. 10 a) BnMYB32 Total RNA from young canola flower buds was subjected to RT-PCR amplification using nested primers, Primer sequences: ATGGGAAGGTCTCCTTGCTGTG (SEQ ID NO. 35), TCATTTCATTTCCAAAGTGCTA (SEQ 15 ID NO. 36), designed using the nucleotide sequence of AtMYB32 and a fragment was sequenced. The protein sequence, shown in Figure 1, codes for a peptide with 80% sequence identity with AtMYB32 (97% identity in the MYB domain) and is designated BnMYB32. 20 The expression pattern of BnMYB32 in canola was determined using Northern blot hybridisation as described in relation to AtMYB32, and was found to be similar to that of AtMYB32 in Arabidopsis. Poly A RNA samples were extracted from canola shoots, roots flower 25 buds and anthers of various developmental stages. A 420 bp BnMYB32 fragment coding for the C-terminal was used as a probe in the Northern blot. The BnMYB32 expressed weakly in roots and young shoots. Its expression commenced in immature anthers and peaks in mature anthers. The nucleotide sequence of BnMYB32 is provided in 30 Figure 1. Another clone identified encodes a protein identical to BnMYB32 but with the deletion of the R2 repeat in the MYB domain. The WO 2005/122751 PCT/AU2005/000851 55 inventors propose that BnMYB32-R3 may be derived from BnMYB32 via alternative splicing of the BnMYB32 mRNA. 5 BnMYB32R3 peptide sequence: MGRSPCCEKDHTNKGAWTKEEDDLIIKLHSLLGNKWSLIATRLPGRTDNEIKNYWNTHVKRKLLRGG IDPTTHRPINEAKAPRDSSETRETEDSLVKFLSFSRQLEKKESFGEERNDQKGLICKKERVEYSIVE EKCLDLNLELRISPPWQDQQHHDETKLWFGKEKYMCTACRFGLGNGKKCSCDNVKCQVEYSSSSSSH SSSDISSSVIGYDFLG (SEQ ID NO. 37) 10 BnMYB32R3 nucleotide sequence: TGATAAGCTTATGGGAAGGTCTCCTTGCTGTGAGAAGGACCACACGAACAAAGGAGCTTGGACTAAA GAAGAAGACGATCTCATCATCAAACTCCATAGCCTCCTTGGAAACAAATGGTCTCTTATCGCGACGA GATTACCGGGGAGAACAGATAACGAGATCAAGAACTACTGGAATACACACGTAAAGAGGAAGCTTTT 15 GAGAGGAGGGATTGATCCCACGACTCATCGGCCGATCAACGAAGCCAAAGCTCCTCGTGATTCGTCT GAGACTAGAGAGACAGAGGACTCGCTTGTGAAGTTTCTATCTTTCAGTCGTCAACTGGAGAAAAAGG AAAGTTTTGGGGAAGAGAGAAATGATCAGAAAGGACTGATTTGCAAAAAAGAGAGAGTTGAGTATTC GATTGTTGAAGAAAAGTGCTTAGATTTGAATCTTGAGCTTAGAATCAGCCCGCCATGGCAAGACCAA CAGCACCATGATGAGACCAAACTTTGGTTTGGGAAAGAGAAGTACATGTGCACTGCATGCCGTTTTG 20 GGTTGGGAAACGGCAAGAAGTGTAGCTGCGATAATGTTAAATGTCAAGTCGAGTACAGTAGTAGCAG CAGCAGCCATTCTTCAAGCGATATTAGTAGTAGCGTTATTGGTTATGACTTCTTGGGTA (SEQ ID NO. 38) b) BnMYB103-1 25 Wild type Arabidopsis thaliana (Landsberg Erecta ecotype), Agrobacterium tumefaciens (AGLI strain) and the plasmid pRK2013 were supplied by Dr. David Smyth (Department of Genetics and Developmental Biology, Monash University, Melbourne, Australia). 30 Plasmids pGEM@-3Zf (+) cloning and transcription vector. pBluescript II SK (ampr) phagemid vector.

WO 2005/122751 PCT/AU2005/000851 56 pBI101.2 (kanr) plasmid containing a promoter-less GUS cassette in the binary vector pBIN19, used for plant transformation. [Jefferson, 1987 EMBO J. 6: 3901-3907]. 5 pRK2013 (kanr) a helper plasmid for mobilisation of non-self transmissible plasmids [[Ditta, 1980 supra]. Bacterial strains E. coli strain DH5c (F~, end Al, hsd R17 [rk~, mk_] . supE44, thi -1, 10 ~, rec Al, gyr A96, rel Al, A [lac/ZYA-arg F] U169 deoR, [#80d lac ZAM15]). E. coli strain MV1190 (A[lac-proAB], thi, supE, A[srt-recA] 306::TnlO [tetR] [F': tra D36, proAB, lacI" 1acZAM15]) 15 Agrobacterium tumefaciens strain AGLI (AGLO recA::bla pTiBo542AT Mop+ CbR) [Lazo, 1991 Biotechnology 9: 963-967] Arabidopsis thaliana line 20 Landsherg erecta ecotype (isolated Landsberg, Germany) containing a homozygous recessive mutation, erecta, which confers a more compact phenotype. Oligonucleotides 25 Universal primer - 5' GTAAAACGACGGCCAGT 3' (SEQ ID NO. 39) 17mer used for sequencing in pGEM and pBluescript vectors. Reverse primer - 5' AACAGCTATGACCATG 3' (SEQ ID NO. 40) 30 17mer used for sequencing in pGEM and pBluescript vectors. pBI101 forward primer - 5' TGTGGAATTGTGAGCGGATA 3' (SEQ ID NO. 41) 20mer used for sequencing and amplification.

WO 2005/122751 PCT/AU2005/000851 57 pBI101 reverse primer - 5' ATTCCACAGTTTTCGCGATC 3' (SEQ ID NO. 42) 20mer 80 base pairs 3' to the multicloning site, used for sequencing and amplification. 5 A5962H02 - 5' GAAAGAAGAAATGGGTCGGA 3' (SEQ ID NO. 43) 20mer situated at the 5' end of AtMYB103 including the ATG start A6004A01 - 5' CAACCACTTCTCCCTTCAGC 3' (SEQ ID NO. 44) 20mer situated 150bp upstream of the AtMYB103 TGA stop codon. 10 A5956A10 - 5' GGGGGGGTCGACATGGGTCGGATTCCATGTTG 3' (SEQ ID NO. 45) 32mer beginning at the ATG start codon of AtMYB103, including a Sal I restriction site. 15 A2485F04 - 5' GGGGAAGCTTTCCTCCTCCTCGTGCGCGGT 3' (SEQ ID NO. 46) 30mer situated 180bp upstream of the AtMYB103 TGA stop codon. Z1346E10 - 5' TAAAAATCAAACCATATGAT 3' (SEQ ID NO. 47) 20mer situated at 3' end of AtMYB103 including the TGA stop codon. 20 SRK1 - 5' TTAAAGAGAGACGATCGAGAG 3' (SEQ ID NO. 48) 21mer situated 20bp upstream of the BnMYB103-1 ATG start codon. SRK2 - 5' CTATTTGGCGTCCTGGACCT 3' (SEQ ID NO. 49) 25 20mer situated 176bp downstream of the BnMYB103-1 ATG start codon. SRK7 - 5' CCAATGGGATCCAAAATGAATCA 3' (SEQ ID NO. 50) 23mer situated at the 3' end of BnMYB103-1 including the TGA stop codon. 30 Isolation of plasmid DNA: Alkaline-lysis extraction WO 2005/122751 PCT/AU2005/000851 58 A rapid alkaline extraction method for the isolation of plasmid [Birnboim, 1983 Enzymology 100: 243-255]. Alkaline-lysis/PEG preciptation procedure for sequencing 5 Plasmid DNA for sequencing was prepared using the modified "mini" alkaline-lysis/PEG preciptation procedure outlined in User Bulletin Number 18 from Perkin Elmer.' Appendix A of the PRISM"'Ready Reaction DyeDeoxy' M Terminator Cycle Sequencing kit protocol Rev.A 10 Restriction enzyme digests 0.5 - 1.0pig DNA was digested in a 20gl reaction using 2 k 1 of the required 1OX buffer and 2 - 4 units of the appropriate restriction enzyme. The reaction mix volume was made up to 20p1l with deionised water and incubated for 1 - 2 hours at the appropriate temperature. 15 Resolution of DNA fragments Agarose gel electrophoresis was used to resolve DNA fragments. The agarose concentrations ranged from 0.5 - 1.0 % (w/v) agarose, 20 [Maniatis, 1982 Molecular Cloning: A Laboratory Manuel: Cold Spring Harbour, New York]. Isolation and purification of DNA fragments The BRESAclean"' DNA purification kit was used to purify DNA 25 fragments from agarose gels. Subcloning of DNA fragments Vector DNA was digested with appropriate restriction enzymes as described in section 2.2.3. The restriction enzymes were heat 30 inactivated at 70 0 C for 15 minutes before treating the lineralised vector with calf intestinal phosphatase (CIP) to remove the 5'phosphate residues. This was achieved by adding 2p1l of 10 x CIP buffer plus 1.4 units of CIP to the reaction mix and incubating for 20 minutes at 37cC [Maniatis, 1982 supra] . To remove contaminating WO 2005/122751 PCT/AU2005/000851 59 proteins and salts 40011 of WizardTM purification resin was added to the reaction. It was mixed by inversion for 1 minutes and passed through a Wizardm minicolumn as per manufactures instructions. 150 - 300ng fragment DNA and 50 - 100ng of vector DNA was incubated in 5 the presence of 3 units of T4 DNA ligase and 1lpl of 10 x T4 DNA ligase buffer in a total volume of 10pl at room temperature for 2 4 hours (sticky end ligation) or at 15 0 C overnight (blunt end ligation) [Ausubel, 1994 supra]. 10 Preparation of electrocompetent E. coli cells A single colony of the E. coli strain to be prepared was inoculated into a 5ml overnight culture of 2YT broth (10g yeast extract, 16g trptone, 5g NaCl, per litre, pH 7.0) containing 10mM MgSO4 and grown at 37cC. This was diluted into 500ml of 2YT containing 10mM 15 MgSO4 and grown in two one litre flasks with shaking at 37 0 C. When the optical density at 550nm reached between 0.8 - 1.0 the cultures were cooled on ice for lhr. The cells were then pelleted by centrifugation at 4 0 C for 5min. at 5000xG. The supernatant was discarded and the cells resuspended in 10ml of ice cold 1mM HEPES, 20 pH 7.0. 200ml of the HEPES was then added and the cells mixed by gentle inversion. This step was repeated and the supernatant discarded. The cells were resuspended in 200ml ice cold 15% (v/v) glycerol and pelleted. Following centrifugation the supernatant was discarded and the cells resuspended in the glycerol remaining 25 in the tube. Aliquots of the cells were frozen in microfuge tubes using liquid nitrogen and stored at -70 0 C. Electroporation of E. coli cells 1 11 of a ligation reaction or 25ng of plasmid DNA was added to 40pl 30 of thawed eletrocompetent cells and incubated on ice for 1 minute. The cells were then transformed using the Invitrogen Eletroporator II. Cuvettes with a 0.1cm gap were used and the apparatus set to 1.5kV, SOF and 150 ohms. Electroporation was performed according WO 2005/122751 PCT/AU2005/000851 60 to the manufacturers instructions. 200pIl of SOC broth was added to the transformed cells and incubated at 37 0 C for 1 hour. The cells were then plated onto appropriate selective media and incubated overnight at 37'C. 5 Screening of plasmid size The colonies to be screened were patched onto appropriate media and grown overnight at 37 0 C. A sterile toothpick was used to remove and resuspend cells in 40 gl of 10mM Tris-HC1, pH8.0, 100mM NaCl, 10mM 10 EDTA in a microfuge tube. An equal volume of phenol-chloroform isoamyl alcohol (25:24:1) and 54 of stop mix was then added. The tubes were vortexed then centrifuged at full speed for two minutes. 2001 of the aqueous layer was loaded onto an agarose gel for electrophoresis and analysis using a vector-only sample as a 15 marker. Constructs found to be larger that the vector were then further assessed via restriction enzyme analysis. DNA Sequencing Dye terminator cycle sequencing was performed using the ABI Prism" 20 BigDye' m Terminator Cycle Sequencing Ready Reaction kit from Perkin Elmer and template prepared according to the manafacturers instructons. An MJ Research MiniCycler' m (PTC-150) was used to amplify the extension products with the following parameters, 96 0 C for 30 seconds, 50 0 C for 15 seconds and 60 0 C for 4 minutes. This 25 cycle was repeated twenty five times. The extension products were precipitated, dried under vacuum and sent to the Microbial Biotechnology and Diagnostic Unit at Monash University (Melbourne, Australia) for electrophoresis. 30 The polymerase chain reaction DNA amplification was performed using taq DNA polymerase and the MJ Research MiniCycler' with Hot Bonnet'M. The reaction mixture contained the following components, taq DNA polymerase reaction WO 2005/122751 PCT/AU2005/000851 61 buffer (lX), MgCl (ranging from 1.0 - 2.0mM), dNTP's (200yM each dNTP), primers (10yM each primer), template DNA (5 - 10ng), taq DNA polymerase (2.5 units) and double distilled water to make the final volume of 20yl. Cycling parameters varied according the size of 5 the fragment to be amplified and Tm of the primers. Extraction of total RNA The method used is a modification of the procedure by [Logemann, 1987 Analytical Biochemistry 163: 16-201 for RNA extraction from 10 plant tissue. One gram of plant tissue was frozen in liquid nitrogen and ground to a fine powder using a mortar and pestle. The powder was transferred to a 15ml tube and 2ml of extraction buffer (8M guanidine hydrochloride, 20mM MES, pH7.0, 20mM EDTA, 50mM mercaptoethanol) added. Three millilitres of 15 phenol/chloroform(phenol/chloroform/iso-amylalcohol, 25:24:1) was added and the tube shaken to extract the nucleic acids. The tube was centrifuged for 10min at 3 0OOxG, the upper phase transferred to a new tube containing 3ml phenol/chloroform and the extraction and centrifugation steps repeated. The upper phase was transferred 20 to another 15ml tube containing 0.2 volumes 1M acetic acid and 0.7 volumes freezer-cold ethanol. The RNA was allowed to precipitate at -20 0 C overnight. The tube was centrifuged for 25min at 9500xG, the supernatant removed and the pellet air dried for 2min. The pellet was then resuspended in 1ml of DEPC-saturated water, 25 transferred to a 1.5ml microfuge tube and LiC1 added to a final concentration of 2M. The tube was mixed and incubated overnight at 4 0 C. The tube was centrifuged for 15min at full speed at 4'C. The supernatant was discarded and the pellet resuspended in 500y1 DEPC saturated water. To this was added 50l 3M Na-acetate, pH5.2, and 30 1000l freezer-cold ethanol. This was mixed and incubated overnight at -20 0 C. The mixture was centrifuged at full speed for 15min at 4 0 C, the pellet resuspended in DEPC-saturated water at a final concentration of 5pg/ml and stored at -70 0

C.

WO 2005/122751 PCT/AU2005/000851 62 Quantification of total RNA An aliquot of total RNA was diluted 1/1000 in 1X TNE buffer (loX 100mM Tris base, 10mM EDTA, 2.OM NaCl, pH 7.4) and absorption of 5 the sample measured at different wave lengths to determine concentration and purity. Absorbance at 260nm was used to determine concentration. The ratio of absorbance at 260 and 280nm was used to indicate purity and absorbance at 325nm to indicate contaminating particulate matter. 10 Isolation of polyadenylated RNA Dynabeadso Oligo (dT) 2 5 were used to isolate polyadenylated RNA from total RNA according to the manufacturers instructions. 15 Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) SuperScript" II Rnase H reverse transcriptase (Gibco BRL) was used to synthesize first strand cDNA according to the manufacturers instructions using gene specific primers. This first strand cDNA was used in a PCR to amplify the fragment of interest according to 20 the protocol suggested by the manufacturer. Extraction of genomic DNA from Arabidopsis 2 - 5g of ground frozen tissue was transferred to a 50ml tube containing 25ml of CTAB extraction buffer. The sample was shaken 25 vigorously and incubated at 650C for 20 minutes with occasional shaking. 10ml of chloroform was added to the sample and placed on an inverter for 20 minutes. To resolve the phases, the sample was centrifuged for 5 minutes at 3,000 rpm in a bench top centrifuge. The aqueous phase was transferred into a fresh 50ml tube containing 30 17ml of isopropanol, mixed, and incubated on ice for 10 minutes. The sample was centrifuged at 3,000 rpm for 5 minutes and the supernatant was discarded. The precipitate was resuspended in 4ml WO 2005/122751 PCT/AU2005/000851 63 of TE buffer by gently passing the solution up and down a pipette tip. 4ml of 4 M LiAc was added and the sample was incubated on ice for 20 minutes. The sample was centrifuged for 5 minutes at 3,000 rpm. The supernatant was discarded and the pellet resuspended in 5 900 1l of TE buffer. To remove contaminating RNA, 25pgq/ml of RNase was added and incubated at 37 0 C for 5 minutes. 100 11 of 3M NaOAc was added, mixed, and divided equally into two microfuge tubes. The samples were extracted with 500p1l of buffered phenol (pH 8.0), buffered phenol:chloroform (1:1) and chloroform, the aqueous phase 10 was transferred to a fresh microfuge tube after each extraction. Two volumes of 100% ice cold ethanol was added and the samples incubated on ice for 5 minutes. The DNA precipitate was collected by spinning the samples for 5 minutes at 13,200 rpm at 4 0 C. The supernatant was discarded, the pellet washed twice with ice cold 15 70% (v/v) ethanol, dried under vacuum and resuspended in 200p1l of sterile deionised water. DNA Quantification DNA was diluted 1/1000 in TEN buffer and absorbance measured at 20 260nm. The ratio of absorbance at 260 and 280nm was used to indicate purity. Construction of Brassica napus cDNA library A B. napus flower-bud cDNA library was constructed using the ZAP 25 ExpressD cDNA synthesis kit and ZAP Expresso cDNA Gigapack III Gold cloning kit from Stratagene. Total RNA was extracted from B. napus flower-buds using the protocol outlined above and poly(A)* RNA isolated using Dynabeads. The synthesis of cDNA was carried out using 5pg of poly(A)* RNA and subsequent steps required for 30 construction of the cDNA library were performed according to the instruction manual accompanying the kit (Revision #080012 and WO 2005/122751 PCT/AU2005/000851 64 4200403-12). Plaque lifts, library screening and in vivo excision were all performed according to the instruction manual. Northern blotting 5 Approximately 3yg of Poly(A)* RNA from each sample was electrophoresed on a 1% agarose gel (50% formamide) with an RNA molecular marker (Promega). The RNA was blotted onto a positively charged nylon membrane (Hybond N, Amersham) using the upward capillary method of transfer [Ausubel, 1994 supra]. The filter was 10 prehybridised (5x SSC, 5x Denhardt's, 1% SDS, 50% formamide with 100pg/ml salmon sperm DNA) for two hours at 42 0 C. The probe was labeled with 32P dCTP (250ACi, Geneworks) using the Megaprime DNA labeling kit (Amersham). Unincorporated nucleotides were removed using a Nickspin column (Amersham). The denatured probe was added 15 to the prehybidisation solution in a glass tube and allowed to hybridise to the filter overnight at 42 0 C with constant rotation in a Hybaid oven. Post hybridisation stringency washes included 2xSSC, 0.1%SDS at RT for 2x5min; 0.2x SSC, 0.1%SDS, at RT for 2xl5min and 0.2xSSC, 0.1%SDS, at 42 0 C for 2x15min. The filter was 20 then wrapped in plastic and exposed to x-ray film overnight at -70 0 C in a film cassette. Methods: Plant transformation The plant transformation protocol described below is based on 25 methods devised by Lazo, 1991 supra and modified by C. Johnson, Department of Genetics and Developmental Biology, Monash University, (Melbourne, Australia). Tri-parental mating 30 Fresh liquid cultures of Agrobacterium tumefaciens (AGLI strain), E. coli containing the required binary vector with insert and E. coli containing the plasmid pRK2013, were prepared. A 100pl aliquot of each culture was mixed in a tube then plated onto 2YT WO 2005/122751 PCT/AU2005/000851 65 plates and incubated overnight at 28 0 C. Two streaks of cells were resuspended in 100tl of 2YT broth. The resuspended cells were streaked for single colonies on 2YT plates containing 75pg/ml kanamycin and 100[tg/ml ampicillin. Several single colonies were 5 inoculated in 2YT broth containing the appropriate antibiotics and grown for 2 days at 28 0 C with vigorous shaking. Plasmid DNA extracted from the culture was used in a PCR with insert specific primers. Presence of the amplified insert indicates that successful tri-parental mating has occurred. 10 Growth and transformation of Arabidopsis thaliana roots Arabidopsis thaliana seeds were surface sterilised in a microfuge tube by washing the seeds with 0.5ml of 70% (v/v) ethanol. The ethanol was removed, replaced with seed sterilising solution (4% 15 chlorine, water, 5% SDS at a ratio of 8:15:1 respectively) and incubated for 10 minutes. The seed sterilising solution was removed and the seeds washed several times in sterile water. Sterilised seeds were sown into 12.5mm petri dishes containing 50ml of GM media and sealed with micropore tape. The seedlings were 20 grown at 22 0 C for 2 weeks under constant illumination (80 LEm- 2 S-') Roots harvested from Arabidopsis plants were placed parallel onto CIM plates and sealed with micropore tape. The plates were wrapped in foil and incubated at 22 0 C for 2 days. The roots were cut into 25 0.5cm pieces, added to liquid cultures of AGLI, shaken and strained through a sterilised steel tea strainer. Excess AGLI was removed by lightly blotting onto sterile 3MM paper and the roots spread over a CIM plate. The plates were sealed with micropore tape, wrapped in foil and incubated for a further 2 days at 25 0 C. Roots 30 removed from CIM plates were washed to remove bacteria by vigorous shaking with sterile water in 50ml tubes. A sterile steel tea strainer was used to strain the roots, and they were blotted dry on WO 2005/122751 PCT/AU2005/000851 66 sterile 3 MM paper. The roots were carefully spread over SIM media containing 100pg/ml kanamycin and 200gg/ml timetin, sealed with micropore tape and incubated at 22 0 C under constant illumination (80 Em 2 s') 5 Selection of transformants Shoots produced from root cultures were transferred into 150mm pots containing GM media with 50pjg/ml kanamycin and 200pg/ml timentin. A small leaf from each shoot was used to check for successful 10 transformation. The leaf was placed into a microfuge tube with 40pl of 0.25M NaOH, boiled for 30 seconds and neutralised by adding 40ptl 0.25MHCl and 20pl 0.SM Tris-HCl (pH 8.0) containing 0.25% (v/v) Nonident P-40 before boiling for a further 2 minutes. The leaf sample was immediately used for PCR as described above. An 15 amplified PCR product of the desired size indicated that successful transformation had occurred. Transformed shoots containing the insert of interest were grown to maturity Histochemical assay of transformed Arabidopsis plants 20 Seeds generated from plants transformed with pBI101 containing the promoter::gusA construct were surfaced sterilised and sown onto GM plates containing 50 pg/pLl kanamycin and grown under constant illumination at 22'C for 2 weeks. Germinated plants that retained their chlorophyll content were transferred to soil. Floral 25 clusters were removed from the plant and placed into a microfuge tube containing X-gluc solution and incubated at 37'C for 12 - 15 hours. The chlorophyll was leached from the plant tissue by placing the tissue into 100 % ethanol. GUS staining was examined under a dissecting microscope. 30 WO 2005/122751 PCT/AU2005/000851 67 Staining, embedding and sectioning of transformed plants Floral heads were pre-fixed with 1% (v/v) glutaraldehyde made up in 50mM sodium phosphate buffer (pH 7.4) for 5 minutes at room temperature. Tissue was washed 4 times with 50mM sodium phosphate 5 buffer (pH 7.4) for 5 minutes each time. The tissue was covered with X-gluc and incubated overnight at 37'C. The tissue was post fixed in a solution of 6% (v/v) glutaraldehyde and 4 % (v/v) paraformaldehyde made up in 50mM sodium phosphate (pH 7.4) and incubated for 2 hours at 4 0 C after 1 minute vacuum infiltration. 10 The tissue was washed three times with 50mM sodium phosphate buffer (pH 7.4) for 5 minutes each time. The floral heads were slowly dehydrated by incubating them for 15 minutes each time with increasing concentrations of ethanol (15%, 50%, 70%, 90% and 100% (v/v)). The tissue was transferred into a fresh tube containing 15 30% LR white : 70% ethanol (v/v) and incubated at room temperature for one hour. This solution was replaced with 50% LR white : 50% ethanol (v/v) and incubated for a further hour. The tissue was placed in 100% LR white and incubated overnight at 4 0 C. The next day, the tissue samples were placed into plastic boats; covered 20 with 100% LR white; covered to exclude oxygen and incubated overnight at 650C. Embedded plant tissue was mounted on a microtome, cut into 10pLm sections and mounted on a clean slide. The sections were stained with 1% (w/v) safranine dye for 2 minutes with gentle heating. Excess dye was washed off with water and then 25 with 50% (v/v) ethanol. Finally the sections were rinsed with 100% ethanol, arranged on the slide and dried with gentle heating. A small drop of clearene and 4 small drops of mounting medium were placed on the sections. The sections were covered with a coverslip and placed on a slide heater. The slide was incubated overnight at 30 50 0 C with a small weight of 200 grams placed on top. The slides were viewed under the light microscope.

WO 2005/122751 PCT/AU2005/000851 68 Electron microscopy Samples analysed using the scanning electron microscope were critical point dried in liquid CO 2 . The samples were coated with 5 palladium-gold and examined using a JEOL JSM 6340F field emission scanning electron microscope. Flower buds at various stages of development were embedded in LR White Hard according to Chaudhury et al. (1994) supra. Sections of 10 70-90nm were mounted on 50-100nm FORMVAR coated copper grids and double stained in uranyl acetate for 6.5 minutes followed by staining in lead citrate for 1.5 minutes. The stained sections were then examined at 80kV in a JEOL 2010 transmission electron microscope. 15 Results Isolation of clone p8.1.1 from a B. napus cDNA library The isolation of an AtMYBl03 homologue from B.napus began with the construction of a flower bud cDNA library (see above). Flower buds 20 representing a number of developmental stages, from 0.5mm to the point before flower opening were used for the extraction of RNA. Flower buds of these sizes contain microspores at a variety of developmental stages. These include meiocytes, tetrads and newly released microspores in the smaller buds and binucleate or 25 trinuleate microspores prior to anthesis in the largest buds. The cDNA library was screened using a 600bp probe derived from AtMYBlO3 including the R3 repeat and part of the C-terminal region, (figure 10). An initial screen of 1x10 6 plaques led to the 30 isolation of four clones following tertiary screening at high stringency (0.2X SSC, 0.1% SDS, 65 0 C). These four clones were sequenced and found to be identical.

WO 2005/122751 PCT/AU2005/000851 69 The clone p8.1.1 consists of an ORF 969bp long with a putative amino acid sequence of 323aa. It contains the imperfect R2R3 repeats characteristic of plant MYB genes, including conserved tryptophan residues. However, compared with AtMYB103 it shows an 5 overall identity of 43% at the nucleotide level and 31% identity at the amino acid level (Figure 11). A BLASTN search of the NCBI data base gave the following results. At the nucleotide level the highest identity is with an A. thaliana putative transcription factor, MYB3S (AF062877) with a score of 361 4e 97 . This partial 10 sequence contains only 27bp of the R2 repeat but shows significant homology within the repeat region, namely 92%. Portions of significant homology occur at the 3' end of the sequence with an overall identity of 62%. Romero et al. (1998) The Plant J. 14: 273-284 identified MYB35 in a systematic search for R2R3-MYB 15 regulatory genes in the genome of A. thaliana but no functional characterisation has been reported. A phylogenetic tree based on the amino acid sequence outside the R2R3 domain classifies the MYB proteins into 22 subgroups [Romero, 1998 supra]. However, MYB35 does not fall within any of these subgroups. A BLASTP search 20 showed highest identity with a transcription factor-like protein found on chromosome three of A. thaliana. This protein has a 95.6% identity in the R2R3 region and an overall identity of 75% with the clone p8.1.1. 25 Both the BLASTN and BLASTP searches showed identity to AtMYB103 with scores of 62 Se -07 and 213 le54 respectively. This may be considered significant, however the greatest identity between these sequences is in the conserved R2R3 repeats showing 69% identity at the amino acid level. The overall amino acid identity is only 31%, 30 indicating that clone p8.1.1 is unlikely to represent an AtMYBl03 homologue.

WO 2005/122751 PCT/AU2005/000851 70 Isolation of an AtMYB103 homologue from a B. napus genomic library A B. napus genomic DNA library (Clonetech FL112d) was screened in this laboratory by Roger Kalla (personal communication) with AtMYB103 3' and 5' probes mixed (Figure 11). The 3' probe was gene 5 specific and excluded the R2R3 repeats. The 5' probe includes 130bp of the R2 repeat and 320bp of the sequence upstream from the ATG start codon. This screening identified two recombinant phage particles containing homologous sequences. Sequence data from these clones show a gene of similar structure to AtMYB103 10 consisting of three exons interrupted by two introns at conserved sites (Figure 12). An ORF of 960bp coding for a protein of 320aa similar to that of AtMYBl03 was identified (Figure 13). This clone, BnMY1B103-l, shows 93% identity to AtMYB103 at the amino acid level, with 99% identity in the R2R3 region and 89% identity in the 15 putative activation domain (Figure 14). A BLASTP search showed the highest score of 561 e 159 with AtMYBl03. Isolation of clones p700-1, p 8 00

-

19 and p900-10 using the reverse transcriptase polymerase chain reaction 20 RT PCR was used in an attempt to isolate an AtMYBl03 homologue. First strand cDNA derived from the polyA* RNA of immature B.napus flowers was used as template in a reaction with AtMYBl03 specific primers. A 700bp product was obtained as a result of two rounds of amplification using nested primers. The first reaction used 25 primers A5962H02 and A6004A01 situated at the 5' start site and putative 3' stop site of the AtMYBlo3 sequence respectively. (Subsequent analysis of the 3' region of AtMYBl03 has placed the stop codon further downstream). A set of nested primers, A5956A10 and A2485F04 were then used to ensure specificity of the amplified 30 product (Figure 15). The 700bp product was cloned as a blunt ended fragment into the Sma I site of pBluescript. Sequence comparison of the 700bp RT PCR product, p700-1 (SEQ ID NO. 24), with AtMYB103 shows 93% identity at the amino acid level, 99% identity in the WO 2005/122751 PCT/AU2005/000851 71 R2R3 region and 89% in the 3' region, although this sequence is incomplete (Figure 16). Further RT PCR was performed using primers designed from the gDNA 5 sequence and first strand cDNA obtained from immature B. napus flowers as template. The first round of PCR was performed using primers SRK1 and SRK7 situated at the 5' end upstream of the ATG start and at the 3' end including the TGA stop respectively. No products of the expected size were observed following 10 electrophoresis on an agarose gel. A second round of PCR was performed using the first round PCR products as template with SRK2, a primer 180bp downstream of SRK1, and SRK7. This time a product of approximately 800bp was amplified which corresponds to the expected size of 805bp (Figure 17). Cloning and subsequent 15 sequencing of this product (p800-19 - SEQ ID NO. 25)) shows 99% identity to the genomic BnMYB023-1 sequence, with only three nucleotide differences. These differences may reflect the different cultivars used to produce the gDNA and cDNA libraries. Compared with AtMYB103 there is an overall amino acid identity of 20 93%, with 99% identity in the R2R3 region and 89% identity in the putative activation domain (Figure 18). Screening of the B. napus flower cDNA library with a 350bp 3' probe (gene specific) derived from this RT PCR product failed to give any positive signals. 25 Using new primers designed from the AtMYBl03 sequence another round of RT PCR was performed. First strand cDNA was generated from polyA+ RNA extracted from immature B. napus flower buds using the primer Z1346E10. This primer is situated at the 3' end of the gene and includes the TGA stop codon. This was used in a PCR with 30 primers A5962H02, situated at the 5' end including the ATG start and Z1346E10. A 950bp product was generated, the approximate size predicted from the AtMYBI03 sequence (Figure 19). This product was cloned (p900-10 - SEQ ID NO. 26)) and sequenced and shares 97% WO 2005/122751 PCT/AU2005/000851 72 identity with AtMYBl03 at the amino acid level (Figure 20). It also shares homology with p800-19 and BnMYBl03-l, although the greatest homology is with p700-1. 5 A BLASTN and BLASTP search of the Genbank data base with p700-1, p800-1 9 and p900-10 all produce the most significant alignment with AtMYB103. Scores of 394 e' 0 ', 428 e~"' and 407 e"' respectively, were obtained at the nucleotide level. Scores of 520 e-4, 434 e- 12 1 and 450 e 126 respectively, were obtained at the amino acid level. A 10 Southern blot of B. napus genomic DNA probed with the 3' end of p700-1 (R. Kalla, personal communication) indicates that at least four different homologues of this gene are present in B. napus. Brassica napus which originates from a cross between B. rapa AA (n=10) and B. oleracea CC (n=9) contains an AACC genome. Southern 15 blotting of genomic DNA from these parental species and probing with the same p700-1 probe provides evidence for at least two homologues in each organism (results not shown). The fact that these four clones, BnMYBl03-l, p700-1, p800-19 and 20 p900-10, were isolated from B. napus flower buds and share significant identity with each other is in keeping with the amphidiploid nature of B. napus. Their homology to AtMYBl03 is in keeping with other Brassica genes that have been isolated in a similar manner. Figure 21 gives an overview of homology between 25 these four clones and AtMYBl03. It is likely that these four clones are homologues of AtMYB103. BnMYB103-l, a likely orthologue of AtMYB103 The Advanced BLAST2 and Orthologue Search Service at EMBL was used 30 to find sequences likely to represent orthologues of BnMYB103-l. Using the BnMYBl03-l amino acid sequence as the query sequence the highest score of 1449 3.6e was obtained for AtMY1l03. The other genes shown in this analysis are all MYB genes or described as MYB- WO 2005/122751 PCT/AU2005/000851 73 like. Further analysis using this service produced a species tree using the NCBI database and a gene tree based on the neighbour joining tree computed by CLUSTALW (Figures 8 & 9) 5 The taxonomic tree shows the closest evolutionary relationship between the B. napus query sequence and the A. thaliana sequences. The sorting of genes according to this taxonomic tree groups nMYBl03-1 with thirty-one A. thaliana sequences. Fifteen of these are MYB genes the others are described as MYB-related or 10 transcription factor-like proteins. Expression pattern of BnMYB103-1 The expression pattern of a gene may be determined using a variety of methods. Reporter genes such as gusA fused to a specific 15 promoter are useful for determining temporal and spatial expression patterns. More sensitive and specific methods include Northern blotting, RT PCR and in situ hybridisation. These techniques involve detection of mRNA that has been transcribed from the gene of interest. Where the gene under investigation is expressed only 20 in certain tissues and at specific stages of development it is important to have reporter gene information regarding expression of the gene in the whole plant as guide to the selection of material. BnMYB103-1 promoter-gusA studies were followed by a more detailed analysis of BnMYBl03-1 expression using Northern blotting and RT 25 PCR. BnMYB103-1 promoter-gusA expression patterns A 725bp region upstream of the BnMYBl03-1 start site was cloned into the Sal I/Xba I sites of pBI1O.2 immediately upstream of gusA 30 reporter gene. When transformed into A. thaliana, the X-Gluc stained inflorescences showed GUS staining only in the anthers of immature flowers. This corresponds to the staining pattern observed using the AtMYBl03 promoter-gusA construct (Figure 22).

WO 2005/122751 PCT/AU2005/000851 74 Sections through these stained anthers show that GUS expression is found in the tapetum and microspores (Figure 23) Northern blot analysis of B. napus tissues 5 The method used follows the protocol outlined by Ausubel et al. (1994) supra. Brassica napus tissues from different stages of development were selected for analysis. These included the roots and shoots from ten day old seedlings, anthers harvested from flower buds 2-6mm in length, mature dehiscing anthers and flower 10 buds of different sizes. The flower buds were divided into three groups representing different developmental stages, 0.5-2mm, 2.5 5mm and 6-8mm in length. Flower bud length has been correlated with the stages of anther and pollen development (data not shown). 15 Total RNA was extracted using Trizol reagent (Gibco). The integrity of RNA was assessed via agarose gel electrophoresis. Absorbance readings on a spectrophotometer at 260 and 280nm were taken to assess the purity of RNA extractions using the A 2 6o to A 2 80 ratio. The A 2 6 0 reading was also used to calculate the concentration 20 of RNA in each sample. PolyA* RNA was extracted from total RNA using oligo dT dynabeads (Dynal) according to the manufacturer's instructions. For each sample, 75 Lg of total RNA was used in the extraction to obtain approximately 3pig of PolyA" RNA. 25 A 474bp BnMYEBlO3-1 gene-specific probe was prepared from the 3' end of the sequence using PCR (96 0 C 3min x 1, 96 0 C 30sec, 50 0 C 30sec, 72 0 C 30sec x 30, 72'C 5min x 1) using primers P16 and Z2346F02 (Figure 24). To check for equal loading of polyA* RNA the filter was stripped and reprobed with a ubiquitin probe. The ubiquitin 30 probe, a 1.2kb fragment from the UBQ4 gene [Burke, (1988) Molecular and General Genetics 213: 435-4431 was gel purified (Bresaclean kit, Geneworks) and labeled as described above. Figure 25 shows the results of probing with both the gene-specific and ubiquitin WO 2005/122751 PCT/AU2005/000851 75 probes. The strongest signal with the gene specific probe is found in flower buds 0.5-2mm in length. A weaker signal is seen in pre anthesis anthers. Probing with ubiquitin indicates that the loading of polyA RNA for each sample was even. 5 Reverse transcriptase polymerase chain reaction for detection of BnMYB103-1 mRNA RT PCR is another sensitive and specific method for the detection of mRNA. The total RNA samples used to prepare polyA* RNA for 10 northern blotting were also used for RT PCR. The Onestep RT PCR kit (Gibco) was used to generate cDNA via reverse transcriptase and PCR was performed in the same tube. Tubulin primers were initially used to assess whether RT PCR 15 products could be obtained from the RNA samples (50 0 C 30min x 1; 940C 2min x1; 94 0 C 30sec, 45 0 C 30 sec, 72 0 C 60sec x 25; 72 0 C 10min x 1). The primers were designed using the published sequence of A. thaliana P-8 tubulin [Snustad, 1992 The Plant Cell4: 549-556]. A product of 545bp is expected from cDNA using the tubulin primers. 20 A product from contaminating genomic DNA would be 760bp as the primers span two introns. All of the samples tested with tubulin primers produced a strongly amplified product of the correct size (Figure 26). Two pairs of BnMYBl03-l specific primers were tested for their ability to produce a RT PCR product using RNA from pre 25 and post-anthesis anthers. The primer pairs employed were A5956A10 & A2485F04 and A5962H02 & Z1346E10 with fragments of 784bp and 979bp expected from cDNA respectively. RNA from pre-anthesis anthers gave a strongly amplified product of -784bp with primers A5956A10 & A2485F04 while primers A5962H02 & Z1346E10 generated a 30 product of -979bp. No product was seen when RNA from post-anthesis anthers was used as template (Figure 27).

WO 2005/122751 PCT/AU2005/000851 76 The remaining RNA samples were then used as templates in RT PCR with primers A5962H02 & Z1346E10. Products corresponding to the size expected for BnMYBl03-1 were only obtained from RNA isolated from pre-anthesis anthers and flower buds 0.5-2mm in length (Figure 5 26). These RT PCR results agree with the Northern blotting data and BnMYB1303-1 promoter-gusA experiments. Expression of BnMYB103-1 is restricted to pre-anthesis anthers and is most strongly expressed 10 in flower buds 0.5-2mm in length. At this stage of development the pollen exists as meiocytes, tetrads and newly released, pre-mitotic microspores and the tapetum shows its highest secretory activity. b) BnMYB1 03-2 15 The second orthologue of AtMYBl03 is named BnMYBI03-2 and its nucleic acid sequence and putative amino acid sequence are provided in Figure 2. Example 3 - Orthologues of the MYB genes 20 BLAST searching of the GenBank database has identified putative orthologues of BnMYB32, namely GhMYB9 from cotton (Figure 5b) TMH27 from tomato (Figure 5a) and AtMYB32 from wheat (Triticum aestivum) (Figure 36). 25 BLAST searching of the GenBank database has identified putative orthologues of E3nMYB103-1, namely OsMYBl03 from rice (Figure 6 and Figure 37). BLAST searching of the wheat Grain gene database (http://wheat.pw.usda.gov) identified a putative orthologue of 30 AtMYB103 from wheat (Triticum aestivum) (Figure 38). The wheat and rice putative sequences were used to clone the orthologue genes.

WO 2005/122751 PCT/AU2005/000851 77 The wheat orthologue TaMYB32 was cloned using RT-PCR with nested primers. The deduced amino acid sequence exhibits 63 % identity to the AtMYB32 sequence (Figure36). 5 A homologous fragment of AtMYB103 from rice (Oryza sativa) anther cDNA was cloned using primers Tr-F1 and Tr-R2 (see Figure 37). The fragment includes part of the R3 sequence and sequence immediately downstream from R3. The rice orthologue of AtMYB103, 10 termed OsMYB13, exhibits a high level of sequence identity to AtMYB103 at the amino acid level. Figure 38 shows an alignment of the deduced amino acid sequences of AtMYB103 and a fragment of its wheat orthologue TaMYB103. 15 Underlined amino acids are identical. The deduced amino acid sequence is about 90% identical to that of AtMYB103 sequence (Figure38). Example 4 - Functional Analysis of the MYB genes 20 Functional analysis of BnMYB103-1 A number of strategies are available to obtain information about the function of a specific gene. The expression of BnMYB103-1 in immature anthers suggests a role in the development of the male reproductive structures. Homology with genes of known function may 25 also provide some clue to function. BnMYBl03-1 is highly homologous to the A. thaliana gene AtMYBl03. As previously stated, AtMYBI03 is expressed in the tapetum and developing microspores of A. thaliana at the stage where meiocytes are dividing meiotically to produce microspore tetrads [Li, 1999 supral. The spatial and 30 temporal expression patterns of AtMYB103 imply a function in the regulation of pollen development. An estimated 20 000 mRNA WO 2005/122751 PCT/AU2005/000851 78 transcripts are present in the mature pollen of Tradescantia paludosa and maize, reflecting the complex nature of pollen development and presumably the requirement for regulatory genes. 5 Mutants also provide important information about gene function. Transposon mutagenesis can be used to disrupt a specific gene and function implied from any consequent phenotypic changes. Moreover, the gene can then be cloned and function verified using gene rescue. If the gene of interest is responsible for the mutant 10 phenotype, the wild type phenotype will be regained. Mutagenesis has been used to generate a number of A. thaliana mutants that are defective in pollen development. Antisense and sense techniques have been used in plants to disrupt 15 the function of known genes. Many examples exist where a changed phenotype has been produced using this approach. Both techniques appear to silence expression of the endogenous gene being targeted. Post-transcriptional gene silencing (PTGS) has received significant attention in recent years and some of the mechanisms involved are 20 now beginning to be understood. The TA39 promoter Successful targeting of a specific endogenous gene using antisense and sense constructs requires a strong tissue specific promoter. 25 The promoter from a pollen/anther specific gene of tobacco, TA39, was selected for this work. This promoter is expressed in the microspores, connective tissue and most highly in the tapetum during early pollen development [Goldberg, 1993 supral. 30 A TA39 promoter-gusA fusion (TA39::gusA) transformed into A thaliana or B. napus drove strong GUS expression in the young anthers (Figure 28). This expression pattern closely resembles that obtained with an AtMYB103 promoter fused to gusA [Li, 1999 WO 2005/122751 PCT/AU2005/000851 79 supra]. AtMYB103 antisense and sense constructs driven by the AtMYB203 promoter had previously failed to produce a changed phenotype when transformed into A. thaliana. Hence, the TA39 promoter was selected for use with the BnMYBl03-1 antisense and 5 sense constructs. Antisense and sense BnMYB103-1 constructs The TA39 promoter was cloned as a 1.8kb BamHI/SalI fragment from p4AI-LIT into pBluescript. An 800bp fragment of BnMYB103-1, from 10 the R3 repeat to the stop codon, was cloned as a BamHI fragment in front of the TA39 promoter. Restriction enzyme analysis (PvuII/SnaBI) identified constructs with the BnMYB103-1 fragment in both the sense and antisense orientation (Figure 29). The 2.6kb TA39::BnMYBl03-1 SacI/SalI fragment from each of these constructs 15 was cloned into the binary vector pBI101.2 for transfer into A. thaliana via root transformation (Figure 30). Characterisation of A thaliana transgenics transformed with antisense and sense BnMYB103-1 constructs 20 It has been predicted that the expression of antisense constructs to Brassica anther-specific mRNAs in Arabidopsis may produce informative mutants...and even suggest a function for the gene product. Transformation of A. thaliana and the generation of transformed plants for analysis can be accomplished in about two 25 months. The same process in B. napus takes at least nine months. Taking these time frames into consideration, both A. thaliana and B. napus were transformed with the same constructs. Considering the high degree of homology between AtMYBl03 and BnMYB103-1 it was felt likely that BnMYBI03-l antisense and sense constructs would 30 interfere with the expression of AtMYBl03 in A. thaliana plants. Following root transformation and selection for transformed tissue, individual plants were grown for phenotypic analysis. The To plants WO 2005/122751 PCT/AU2005/000851 80 were analysed for presence of the transgene using PCR with primers specific to the antisense or sense construct, respectively (Figure 31). Plants containing the transgene were then selected for further analysis. Pollen was taken from dehiscing flowers for 5 analysis using the scanning electron microscope (SEM). Table 1 summarizes the results of this analysis. The percentage of affected pollen ranges from <10% to 100% abnormal pollen. The abnormal pollen phenotype is variable with flattened, small, 10 rounded or misshapen pollen resulting (Figures 32a & b). There appears to be no significant difference between the antisense and sense constructs with regard to the percentages of abnormal pollen or the phenotypes observed. 15 Table 1 - Scanning electron microscope analysis of transgenic A. thaliana lines

T

0 plants PCR positive for To plants PCR positive for Antisense construct sense construct LINE % ABNORMAL POLLEN LINE % ABNORMAL POLLEN ASI 90 S1 20 AS2 80 S2 20 AS3 90 S3 90 AS4 90 S4 80 AS5 90 S5 10 AS6 30 S6 80 AS7 90 S7 50 AS8 20 S8 50 AS9 30 S9 40 AS10 100 S10 30 AS11 90 811 90 AS12 <10 S12 40 AS13 10 813 50 AS14 10 814 10 WO 2005/122751 PCT/AU2005/000851 81 AS1S 90 S15 60 AS16 90 S16 10 AS17 10 S17 50 AS18 20 S18 30 AS19 100 S19 90 AS20 30 S20 100 AS21 40 S21 40 AS22 80 S22 90 AS23 30 AS24 90 AS25 60 AS26 100 AS27 50 AS28 70 AS29 90 All lines, no matter how severely the pollen phenotype was affected were able to produce viable seed. T, plants from selected lines were raised to determine whether the pollen phenotype observed in To 5 plants would be carried over into the next generation. Five of the most severely affected plants from each of the construct types were chosen. In each case the abnormal pollen phenotypes were maintained in the T, generation. PCR analysis of these plants confirmed the presence of the relevant transgene. 10 Characterisation of B. napus transgenics transformed with antisense and sense BnMYB103-1 constructs Transformation of B. napus with the antisense and sense constructs was carried out and viability of transgenic pollen ascertained. 15 All of the PCR positive transgenic plants contained viable pollen as determined by iodine staining. Scanning electron microscope analysis of pollen from these plants showed no significant abnormalities (results not shown). All of the transgenic lines produced viable seed.

WO 2005/122751 PCT/AU2005/000851 82 Crossing of antisense and sense lines Waterhouse et al. (1998) PNAS 95:13959-13964 showed that transgenic tobacco transformed with both sense and antisense forms of the 5 protease (Pro) gene from potato virus Y displayed greater levels of immunity to the virus than plants transformed with one or other of the constructs. In an attempt to obtain a male sterile phenotype, antisense and 10 sense lines from B. napus or A. thaliana and were crossed to obtain transgenic plants of each species carrying both transgenes. Plants shown to be PCR positive for either the sense or antisense transgene were selected for use as both male donors and female recipients. The pre-dehiscent anthers from newly opened flowers 15 were removed and the stigma inspected for the presence of pollen. A dehiscing anther from the male donor was then used to deposit pollen directly onto the stigmatic surface. Treated flowers were tagged with the relevant information and siliques allowed to develop. Seed from these siliques was germinated on selective 20 media, GM kanamycin (50pg/ml) for A. thaliana, GM hygromycin (100gg/ml) for B. napus. Seedlings able to grow on these media were transferred to soil and grown to maturity. The presence of both transgenes was confirmed with PCR using primers specific to each construct. Lines raised from each of these crosses were 25 analysed for abnormal pollen phenotypes using the SEM. All of the A. thaliana lines showed abnormal pollen similar to that observed in the antisense or sense parents. All of the plants produced viable seed, indicating that complete male sterility had 30 not been achieved. The B. napus lines showed a range of phenotypes with up to 90% abnormal pollen, a feature not observed in either of the antisense or sense parents. Some of the plants generated from these crosses were PCR positive for only one of the transgenes, WO 2005/122751 PCT/AU2005/000851 83 indicating a degree of self-fertilsation. As these plants represent a Ta population and may be homozygous for the transgene it was of interest to analyse their pollen. Some of these T 1 plants did show abnormal pollen phenotypes similar to those seen in A. 5 thaliana. SEM analysis of these transgenic plants showed that the plants were able to produce viable seed. Pollen phenotypes Wild type A. thaliana and B. napus pollen both appear very similar, 10 apart from an obvious size difference. An A thaliana WT pollen grain is approximately 30x15pm and a B. napus pollen grain approximately 40x2Opm. Both are tricolpate with germinal apertures that run the length of the grain and are positioned equidistantly. The exine has a characteristic network of ridges that form a 15 regular pattern. Using the SEM a range of abnormal pollen phenotypes was observed in A thaliana transformed with the sense, antisense or both constructs. Similar phenotypes are observed in T 1 sense and antisense B. napus plants or when both constructs are present. The abnormalities can be divided into a number of 20 categories: 1. Pollen exhibiting normal exine patterning but displaying abnormalities in shape, size and germinal aperture positioning; 2. Aborted pollen that appears small and deflated but with normal exine patterning; 3. Abnormally shaped grains that appear partially or fully deflated with normal exine patterning but 25 coated in parts as if the contents have "oozed" out. In some cases, pollen displaying this phenotype occurred in large clumps, apparently held together by the secreted contents; and 4. Pollen that is small and rounded. 30 In many cases the abnormal pollen from BnMYBl03-l sense and antisense transgenic A. thaliana and B. napus plants are larger than wild type. Abnormally large pollen from A. thaliana showed an average increase of 15% in length and 30% in diameter. The largest WO 2005/122751 PCT/AU2005/000851 84 increases observed were 26% in length and 46% in diameter. In B. napus the average increases were 5% in length and 13% in diameter. The largest increases observed were 10% in length and 30% in diameter. 5 Transmission electron microscope analysis of abnormal pollen Abnormal pollen grains from a number of A. thaliana lines transgenic for either the sense or antisense BnMYB103-1 transgene were selected for analysis using the transmission electron 10 microscope (TEM). Thin sections through these pollen grains were compared with sections of wild type pollen at the same developmental stage. The exine, intine, middle layer, tapetum and microspores of pollen from the transgenics all appear normal. No obvious differences can be detected in these sections. 15 Example 5 - Production of Male Sterile Plants From Example 4 it is shown, that at least in B. napus, blocking expression of BnMYB103 alone does not produce 100% sterile plants. 20 The inventors propose that at least two and preferably three genes need to be blocked in B. napus to produce 100% male sterile plants. In an attempt to provide 100% male sterile plants in B. napus the following construct containing a glucocorticoid (dexamethasone) 25 mediated transcriptional induction system was made. The strong TA39 promoter was incorporated in the system to ensure an adequate level of GVG is produced in developing anthers. Once activated by dexamethasone (DEX) GVG can induce the transcription from the chimeric promoter. A GUS reporter gene was placed under the .30 control of the chimeric promoter to examine the effectiveness of the system.

WO 2005/122751 PCT/AU2005/000851 85 Exact details of the construct are as follows, with reference to Figures 33 a & b and Figure 34: The BamHl-digested BnMYB03-l fragment was blunt-ended and inserted 5 into the EcoRV site of the plasmid containing BnMYB103-2 sequence. The BamH1 AtMYB32 fragment was then ligated to the BamHI site of the resultant plasmid pBnMYB103-12 creating pMYB103-32 (all sequences in the same orientation) and pMYB103-23 (AtMYB332 in opposite orientation). The Xhol-Xbal fragments from the two 10 plasmids were cloned into the Xhol-Xbal sites of pLITMUS28 (New England Biolab) producing pLTMYB103-32 and pLTMYB103-23, respectively. The MYB fragments were then released from pLTMYB103 32 and pLTMYB103-23 using Spel restriction enzyme and inserted into the Spel site of pTADEX1 generating pTADEX31 and pTADEX32, pTADEX71 15 and pTADEX72, respectively. Plasmid pTADEX1 was constructed by replacing the Sbfl-Pmel CaMV35S fragment in pTA7002 (Nam-Hai Chua, Rockefeller University USA) with a Pst1-EcoRV TA39 promoter fragment. The numbers marking the ends of the MYB fragments indicate the number of base pairs downstream from the translational 20 start site (ATG). AtMYB32 sequence includes a translational start site at position 1 and a stop codon at position 822. Boxes represent promoters. GVG encodes a chimeric transcription factor which binds to and 25 activates the 6UAS promoter in the presence of dexamethazone. HPT gene encodes hygromycine phosphotransferase used for the selection of transgenic plants. The system was introduced into Arabidopsis plants and the GUS 30 expression of four lines was examined. The GUS activity was detected only in the anthers sprayed with DEX (Figure 35B). Two allelic sequences of BnMYBI03, BnMYB103-1 and BnMYB103-2 and one copy of AtMYB32 sequence were linked together in a sense or WO 2005/122751 PCT/AU2005/000851 86 antisense orientation. These fragments containing multiple MYB sequences were placed under the control of the chimeric promoter.The constructs were transformed into Arabidopsis plants and PCR analysis was used to ascertain the presence of the 5 transgenes in the plants. The seed setting of 25 transgenic lines was examined in repsonse to the DEX spray and eight of these lines exhibited abnormality in seed setting. The majority of the siliques (50-95%) in these lines 10 failed to elongate (aborted (Figure 35H) and the elongated siliques contained fewer seeds than the untreated siliques. These transgenic lines produced abnormal pollen in response to DEX induction (Figure 35F). Most of these abnormal pollen are devoid of contents as demonstrated by Alexander staining and the presence 15 of aggregates inside the pollen sacs (Fig 35D). However, the gynoecia of the male sterile flowers remained fertile as cross pollination with untreated pollen restored the seed setting. Example 6 - A repressor/restorer male sterility system 20 Transcription repressors use various repressive motifs to block the transcription of their target genes. One of the plant repressive motifs consists of about twelve amino acids (LDLDLELRLGFA, or LDLNLELRISPP). The two peptide sequences 25 exhibit nine identical or homologous residues. The repressive motif (LDLDLELRLGFA) from Arabidopsis was fused to four different transcription factors with resultant loss-of-function phenotypes demonstrating that the motif converts transcription factors into strong dominant repressors. Inspection of AtMYB7 30 (Li, S.F., Parish, R.W. (1995) Plant J. 8, 963) and AtMYB32 C terminal sequences identified the repressive sequence

(LDLNLELRISPP).

WO 2005/122751 PCT/AU2005/000851 87 The strategy is to convert AtMYB103 into a strong repressor using the motif of twelve amino acids to obtain male sterility. The heterozygous male sterile inbred A containing the chimeric AtMYB103 repressor (AtMYB103Rep) is crossed with inbred B 5 containing a "restorer" to produce male fertile F1 hybrid seeds (Figure44). The "restorer" will be AtMYB103 gene under the control of the strong anther-specific At39 promoter or multiple copy of the AtMYB103 promoter or their equivalents. Higher levels of the AtMYB103 protein (activator) in the hybrid plants 10 will alleviate the repression by AtMYB103Rep protein leading to viable pollen production and seed setting. The system produces 50% male fertile inbred A seedlings. A selectable marker linked to the repressor will provide a 15 selection to cull the male fertile inbred A seedlings before crossing with inbred B. Alternatively, a simple chemical which can overcome the repression by the AtMYB103Rep may be used to induce seed setting of the male sterile plants to obtain the homozygous inbred A seed. The homozygous and male sterile inbred 20 A plants will then be crossed with inbred B plants to obtain F1 hybrid seed. The repressor/restorer reversible male sterility system is being successfully tested in Arabidopsis and is suitable for hybrid seed production in any crop plants including canola and wheat. 25 A sequence coding for the repressor motif of twelve amino acids, LDLDLELRLGFA, was fused in frame to the 3' end of the AtMYB103 coding sequence producing a chimeric fusion peptide. The chimeric gene was placed under the control of the AtMYB103 promoter 30 (Figure 39). The repressor construct was transformed into Arabidopsis plants and more than 50% of the sixty transformants exhibited male sterility. A restorer (activator) containing the WO 2005/122751 PCT/AU2005/000851 88 AtMYBl03 gene under the control of the At39 promoter was constructed (Figure 39). Eight Arabidopsis lines transgenic for the restorer were obtained and exhibited normal male fertility. Pollen from these restorer plants was used to pollinate some male 5 sterile plants containing the repressor construct and the F1 hybrid seeds were collected for further analysis.

Claims (18)

1. A method for disrupting pollen development in a plant, the 5 method comprising inhibiting the expression of an endogenous nucleic acid molecule which is, under normal conditions, detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding transcription factors. 10

2. A method according to claim 1, in which the nucleic acid molecule whose expression is blocked encodes MYB 32 or MYB 103.

3. A method according to claim 2, in which MYB32 has the amino 15 acid sequence provided in Figure 1 or is a homologue, orthologue or derivative thereof having at least 50% similarity to the amino acid sequence of Figure 1 and which capable of being detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding 20 transcription factors.

4. A method according to claim 3, in which the plant is Brassica napus and MYB32 has the amino acid sequence provided in Figure 1. 25 5. A method according to claim 3, in which the plant is Arabidopsis thaliana and MYB32 has the amino acid sequence provided in Figure 4.

6. A method according to claim 3, in which the plant is tomato 30 and MYB32 has the amino acid sequence provided in Figure 5a.

7. A method according to claim 3, in which the plant is cotton and MYB32 has the amino acid sequence provided in Figure 5b. WO 2005/122751 PCT/AU2005/000851 90

8. A method according to claim 3, in which the plant is Zea mays and MYB32 has the amino acid sequence provided in Figure 7a. 5 9. A method according to claim 3, in which the plant is Zea mays and MYB32 has the amino acid sequence provided in Figure 7b.

10. A method according to claim 3, in which the plant is Hordeum vulgare and MYB32 has the amino acid sequence provided in Figure 10 7c.

11. A method according to claim 3, in which the plant is Nicotiana tabacum and MYB32 has the amino acid sequence provided in Figure 7d. 15

12. A method according to claim 3, in which the plant is Triticum aestivum and MYB32 has the amino acid sequence provided in Figure

36. 20 13. A method according to claim 2, in which MYB 103 has the amino acid sequence provided in Figure 2, Figure 3 or as Atmybl03 in Figure 11, or is a homologue, orthologue or derivative thereof having at least 50% similarity to the amino acid sequence of Figure 2, Figure 3 or as Atmyb1O3 in Figure 11 and which is capable of 25 being detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein belonging to the MYB class of DNA binding transcription factors. 14. A method according to claim 13, in which the plant is Brassica 30 napus and MYB 103 has the amino acid sequence provided as Figure 2, Figure 3 or as Atmyb103 in Figure 11. WO 2005/122751 PCT/AU2005/000851 91 15. A method according to claim 13, in which the plant is rice and MYB103 has the amino acid sequence provided as Figure 6 or Figure 37 5 16. A method according to claim 13, in which the plant is wheat and MYB103 and has the amino acid sequence provided as Figure 38. 17. A method according to any preceding claim, in which expression of nucleic acid molecules encoding MYB 32 and MYB 103 is blocked. 10 18. A method according to any preceding claim, in which expression of the endogenous nucleic acid molecule is effected by sense or co suppression technology. 15 19. A method according to any one of claims 1 to 17, in which expression of the endogenous nucleic acid molecule is effected by antisense RNA interference technology. 20. A method of inducing male sterility in plants, said method 20 comprising disrupting pollen formation according to any one of claims 1 to 19. 21. A method according to claim 20, in which the male sterility is reversible. 25 22. An isolated nucleic acid molecule capable of blocking expression of a gene encoding MYB32 having the amino acid sequence provided in Figure 1 or an orthologue thereof having at least 50% sequence homology to the amino acid sequence provided in Figure 1, 30 which MYB32 gene or orthologue are detectably expressed in anther tissue during pollen formation. WO 2005/122751 PCT/AU2005/000851 92 23. A nucleic acid molecule according to claim 22, in which the nucleic acid molecule comprises the nucleotide sequence provided in Figure 1, 4, or Figure 36 or a sequence complementary or antisense thereto, or a fragment of the nucleotide sequence or its 5 complement, which nucleic molecule i's capable of blocking expression of the gene encoding MYB32. 24. An isolated nucleic acid molecule capable of blocking expression of a gene encoding MYB103 having the amino acid sequence 10 provided in Figure 2 or Figure 3 or an orthologue thereof having at least 50% sequence homology to the amino acid sequence provided in Figure 2 or Figure 3, which MYB103 gene or orthologue are detectably expressed in anther tissue during pollen formation. 15 25. A nucleic acid molecule according to claim 22, in which the nucleic acid molecule comprises the nucleotide sequence provided in Figure 2, 3, 37 or Figure 38 or a sequence complementary or antisense thereto, or a fragment of the nucleotide sequence or its complement, which nucleic molecule is capable of blocking 20 expression of the gene encoding MYB103. 26. A nucleic acid construct comprising a nucleic acid molecule according to any one of claims 22 or 23. 25 27. A nucleic acid construct comprising a nucleic acid molecule according to any one of claims 24 or 25. 28. A nucleic acid construct according to claim 26 or 27 further comprising a promoter capable of targeting expression of 30 the nucleic acid molecule in the construct to the anther. WO 2005/122751 PCT/AU2005/000851 93 29. A nucleic acid construct according to claim 28 in which the promoter is capable of targeting expression of the nucleic acid molecule in the construct to the anther during pollen formation. 5 30. A nucleic acid construct according to any one of claims 26 or 28 to 29, in which the construct also comprises a nucleic acid molecule according to claim 22 or 23. 31. A nucleic acid construct according to any one of claims 26 to 10 30 further comprising one or more additional genes involved in male specific development. 32. A nucleic acid construct according to any one of claims 26 to 31 comprising an inducible promoter. 15 33. A plant cell transformed with the isolated nucleic acid molecule according to any one of claims 22 to 26 or the construct according to any one of claims 27 to 32. 20 34. Transgenic plants generated from plant cells according to claim 33. 35. Progeny of plants generated from the transgenic plants according to claim 34. 25 36. Seed, tissues, roots, shoots, tubers or propagating material of transgenic plants according to claim 34 or progeny according to claim 35. 30 37. Use or the nucleic acid molecule according to any one of claims 22 to 26 or nucleic acid construct according to any one of claims 27 to 32 for producing male sterile plants. WO 2005/122751 PCT/AU2005/000851 94

38. A male sterile plant in which expression of an endogenous nucleic acid molecule is blocked, which nucleic acid molecule is, under normal conditions, detectably expressed in anther tissue of a plant during pollen formation, and which codes for a protein 5 belonging to the MYB class of DNA binding transcription factors.

39. A male sterile plant according to claim 38, in which expression is blocked using a nucleic acid molecule according to any one of claims 22 to 25 or a nucleic acid construct according to 10 any one of claims 26 to 32.

40. A male sterile plant according to claim 38 or claim 39, in which the plant is monocotyledonous or dicotyledonous. 15 41. A male sterile plant according to claim 38 or claim 39, in which the plant is a legume, crop, cereal, native grass, fruiting plant or flowering plant.

42. A male sterile plant according to claim 38 or claim 39, in 20 which the plant is a Brassicaceae or other Solanaceae species, including potato and cole vegetables, cabbage, kale, collards, turnips, rutabaga, kohlrabi, Brussels sprouts, broccoli and cauliflower, mustards and oilseeds, crucifers, broccoli, canola, tomato, grain legumes, wheat, barley, maize, tobacco and rice. 25

43. A male sterile Arabidopsis plant, in which expression is blocked using the nucleic acid molecule according to claim 22 and the nucleic acid molecule according to claim 24 or the nucleic acid construct according to claim 30. 30

44. A method according to claim 1 in which the nucleic acid molecule whose expression is blocked is not ZmMYBP2, NtMYBAS1, NtMYNAS2, or AtMYB103. WO 2005/122751 PCT/AU2005/000851 95

45. A nucleic acid molecule according to any one of claims 22 to 25 which is not ZmMYBP2, NtMYBAS1, NtMYNAS2, or AtMYB103. 5