Methods and constructs for producing male sterile plants.

Pí'T WORLDINTELLECTUAL PROPERTY ORGANIZATION 5HQÍ7 · * <> * International Burean

INTERNATIONAL APPLICATION PUBUSHED UNDER THE PATENT COOPERATED TREATY (PCT)

<td>(SI) International Patent Cbudficitioo 6: C12N 15 / 82,15 / 52, A01H 1Λ2,5Λ »</td><td>To the</td><td colspan="2">(11) International Publication Number; WO 96/40950 (43) International Publication Date: 19 December 1996 (19.12.96)</td>

<td colspan="3">(21) International Appücation Number: PCT / US96 / 08692 (22) International Filfas Date: 7 June 1996 (07.06.96) (30) Priority Data: 08/476364 June 7, 1995 (07X16.95) US (71) Applfcant: PIONEER HI-BRED INTERNATIONAL, INC. [US / US]; 700 Capitol Squarc, 400 Locust Street, Des Moines, IA 50309 (US). (72) Inventor: FABUANSKL Steven, F .; 6068 Forestglen Cres- oent, Gloucester, Ontario K1C 594 (CA). ALBANI, Diego; John Monday Center, Colney Lañe, Narfolk NR4 7VJ (GB). LAURIAN, Robert, Sd Mant Researeh Center, Central Experimental Patín, Agricultura and Agrifood, KW Neatby, 960 Carting Avenue, Ottawa K1A 0C6 (CA). ARNISON, Paúl, G ,; Farr Biotochnology Group, 1612 Marcoux Drive, Orieans, Ontario K1E 2K6 (CA). (74) Agent: BENT, Stephen, A. et si .; Foley &amp; Lardner, Suite 500, 3000 K Street, NW, Washington, DC 20007-5109 (US).</td><td>(81) Deaignated Statea: AL, AM, AT, AU, AZ, BB, BO, BR, BY, CA CH, CN, CZ, DE, DK, EE, ES, FI, OB. GE, HU, IL, IS. JP, KE, KG, KP, KR, KZ, IX LR, LS, LT, LU, LV, MD, MG, MK, MN, MW, MX, NO, NZ, PL, PT, RO, RU, SD, SE , SG, SI, SX TJ, TM, TR, TT, UA UG, UZ, VN, ARIPO patent (KE, LS, MW, SD, SZ, UG), Euratian patent (AM, AZ, BY, KG, KZ, MD, RU, TJ, TM), European patent (AT, BE, CH, DE, DK, ES, FI, FR, GB, GR, IE, IT, LU, MC, NL, PT, SE), OAPI patent (BF, BJ, CF, CG, d, CM, GA, GN, ML. MR, NE, SN, TD, TG). PuMished With intematíonai tearch repon. Befare the expirado »of the time limit for amending the claime and to be republúhed in the event of the receipt of amendments.</td>

<td colspan="4">(54) Tltle: METHODS AND CONSTRUCTS FOR PRODUCING MALE STERDLE PLANTS (57) Abstract A procesa is described for producing fettüe hybrid seed or hybrid seed comprising fertile and stetile seed using male-sterile plants cteated by employing molecular techniques lo manipulate genes that ate capable of controlling the production of fertile pollen in plants. Hybrid seed production is sitnpUfled and improved by this spproech, which can be extended to plant crop species for which commercially acceptable hybrid seed production methods have not been avaflable.</td>

MOLECULAR METHODS OF HYBRID SEED PRODUCTION *

TECHNICAL FIELD

The present invention relates to amethod for producing plants with male sterility and hybrid seed, to the genetic material used to impart male sterility and to new products produced through this method, mainly, genetically transformed plants bearing the trait of male sterility, Male sterile plants and hybrid seeds are produced by pollinating such plants with pollen from male fertile plants. The present invention also relates to a method of producing a plant bearing the male sterility trait, or its seeds, produced through pollination of such a plant with pollen, from genetically transformed male fertility plants which, by virtue of of the genetic material used, they are resistant to herbicides, and in this desirable way, they are restored with respect to the fertility of the hybrid seed.

BACKGROUND

Hybrid seed production for commercial sale is a huge industry. Hybrid plants developed from hybrid seed benefit from the heterotic effects of crossing two genetically distinct breeding lines. The agronomic performance of this result is superior both in origins, typically in vigor, production and uniformity. The best performance of hybrid seeds varies compared to open pollinated varieties which makes hybrid seeds more attractive to plant farmers and thus demands a premium price on the market.

In order to produce uncontaminated hybrid seeds with equal seeds, pollination control methods must be implemented to ensure cross-pollination and not self-pollination. Pollination control mechanisms can be mechanical, chemical, or genetic.

A simple mechanical method for the production of hybrid seeds can be used if the plant species in question have spatially separate male and female flowers or separate male and female plants. The corn plant, for example, has male pollen-producing flowers influencing the apex of the plant and female flowers in the leaf axils along the stem. Crossing is ensured through mechanical de-donation of female plants to avoid uniformity.

However, most of the major crop plants of interest have both male and female functional organs within the same flower, so emasculation is not a simple procedure. Pollen-forming organs can be removed by hand prior to removal of the pollen. But this form of hybrid seed production is extremely laborious, intense, and expensive. The seed is produced in this way if the value and quantity of seed recovered guarantees the effort.

A secondary general method of producing hybrid seeds is to use chemicals to kill or block the formation of viable pollen. These chemicals, called gametocides, are used to impart transient male sterility. The commercial production of hybrid seeds through the use of gametocides is limited by the cost and availability of the chemical products and the reliability and duration of action of the applications. These chemicals are not effective for crops with an extended flowering period, since new flowers will be produced that will not be affected. Repeated application of chemicals is impractical due to cost.

Many current commercial hybrid seed production systems for field crops are based on a genetic method of pollination control. Plants that are used as females either fail to make pollen, fail to separate pollen, or produce pollen that is biochemically unsuitable for self-fertilization. Plants that are not able (through any combination of different means) to self-pollinate biochemically are called self

<img file="MX9709745A_D0001.tif" />

incompatible. The difficulties associated with the use of self-incompatibilities are: availability and spread of the self-incompatible female line and the stability of the self-incompatibility. In some cases, self-incompatibility can be overcome chemically or immature shoots can be pollinated by hand before the biochemical mechanism that blocks pollen is activated. Self-incompatible systems that can be inactive are generally very vulnerable to stressful weather conditions that break or reduce the effectiveness of the biochemical block to self-pollen.

A widespread interest for the production of commercial seeds are pollen control systems based on genetic mechanisms that cause male sterility. These systems are of two general types: (a) male gene sterility, which is the failure of pollen formation due to one or more nuclear genes, or (b) genetic cytoplasmic male sterility (commonly called cytoplasmic male sterility or CMS) where pollen formation is blocked or aborted due to a defect in the cytoplasmic (mitochondrial) organelle (for general discussions of gene sterility, CMS, and plant hybrid formation see Frankel, R., et al., Pollination Mechanisms, Reproduction and Plant Breedinq; Springer V., et al., Monoqraphs on Theoretical and Applied Genetics, NY, 1977; Edwardson, JP, Bot. Rev. 36: 341-420, 1970).

Nuclear (gene) sterility can be either dominant or recessive. Dominant sterility can only be used for the production of hybrid seeds if the fertility of hybrid plants is not critical, and if the propagation of the female line 5 is reliable, for example, through clonal propagation or through the use of a marker will select closely linked to the sterility gene.

Many successful hybridization schemes involve the use of CMS. In these systems, a specific mutation in the cytoplasmically located mitochondrium can, when in the appropriate nuclear background, lead to failure of mature pollen formation. In some other cases, the nuclear background can compensate for the cytoplasmic mutation and normal pollen formation occurs. The nuclear trait that enables pollen formation in plants with 15 CMS mitochondria is called restoration and is the property of specific restorative genes. Generally, the use of CMS for commercial seed production involves the use of three breeding lines, the male sterility line (female origin), a maintenance line, which is isogenic to the male sterility line, but contains fully functional mitochondria and the line of male origin.

The male line may carry the specific restorer genes (usually designated as a restorer line), which then impart fertility to the hybrid seeds. For crops such as vegetables, for which seed recovery from the hybrid is not important, a CMS system without restoration can be used. For crops for which the hybrid fruit or seed is the commercial product, then the fertility of the hybrid seed must be restored through the specific restorer genes in the male origin or the male sterility hybrid must be pollinated. Pollination of unrestored hybrids can be accomplished by including with hybrids a small percentage of male fertility plants to effect pollination. In most species, the CMS trait is maternally inherited (since all cytoplasmic organelles are inherited only from the egg cell), which may restrict the use of the system.

In a crop of particular interest herein, the oilseed crop of the species Brassica napus or Brassica campestris, termed as canola, no commercial hybrid system has been perfected so far. Mechanical flower emasculation is not practical for hybrid seed production on any scale. The use of currently available gametocides is not practical due to the undetermined nature of flower production. Repeated application of chemicals is expensive and the method is prone to contamination with the pure seed.

Genes that result in self-incompatibility are widely spread in Brassica species and self-incompatible hybrid systems have been used for the production of hybrid seeds in plants. The main difficulties are associated with the spread of feminine lines and the interruption of self-incompatibilities under stressful conditions. Adaptation of these systems to Brassica oilseeds is restricted by the expense of increasing female lines and the availability of appropriate self-incompatible genes in the dominant species of canola, Brassica napus.

A variety of sources of male sterility are available in Brassica species. Both recessive and dominant gene systems have been reported, however, their use is restricted since large-scale in vitro propagation or variation of female lines is in most cases impractical for large-scale seed production.

In addition, a number of CMS systems have been reported in the Brassica species. Four of these systems have been explored as possible vehicles for the production of hybrid seeds: pol, nap, anand and ogu. The Polima (pol) system has been extensively studied and is probably the closest to commercial use. Good restoration and maintenance of pol CMS have been achieved, however the system suffers from potential CMS instability with high temperature, a reduction in the heterotic effect of different crossing lines (due to faulty mitochondria) and reduction in the oil content of the hybrid seed. The use of other CMS systems is also restricted by thermal sensitivity (nap), difficulty in restoring fertility (ogu, anand), difficulty in maintaining sterility (nap), and low temperature chlorosis associated with sterile cytoplasm (ogu ). Improving this system is the object of considerable search, however, all systems have some inherent weakness that limits their usefulness.

For a general discussion of male sterility in Brassica see Shiga, T., Male Sterilitv and Cvtoplasmic Differentiation. ln Brassica Crops and Wild Alies, Bioloqy and Bredinq, Japan Scientific Societies Press, Tokyo pp. 205-221; Thompson, KF, Heredity 29: 253-257).

It is recognized that a desirable system for hybrid seed production in any crop could be a form of male gene sterility that can be regulated or overcome to allow male fertility for the spread or increase of female lines or to allow fertility in hybrids. . This recognition has stimulated the search for the use of molecular systems to effect male gene sterility that can be used for the formation of hybrid seeds. Furthermore, the advent and wide application of recombinant DNA techniques may provide a mechanism for introducing novel DNA sequences to a wide variety of different crop species that is not possible through the limited sexual methods of genetic exchange between different species. . A molecular approach has the advantage that the hybridization system can be imposed on all breeding lines or crops of any given crop without the need for extensive crossing and interruption of established production lines leading to rapid production of production lines. male sterility with a characterized and superior agronomic performance.

BRIEF DESCRIPTION OF THE INVENTION

In accordance with one aspect of the present invention, a recombinant DNA molecule has been provided for use in preparing a male sterile plant of a pollen-bearing species that is capable of transformation. The recombinant DNA molecule comprises:

(A) a first DNA sequence that codes for a first gene product that converts a first substance, which is present in the plant to a second substance, where the first and second substances do not inhibit pollen formation or function, (B) a second DNA sequence encoding a second gene product that converts the second substance to a third product that inhibits pollen formation or function, (C) a first promoter that controls the expression of the first DNA sequence in plants,

-1010 (D) a second promoter that controls the expression of the second DNA sequence in plants, and (E a ​​third DNA sequence that confers resistance to a selective agent.

In a preferred embodiment, the aforementioned resistance is to a herbicide, such as a herbicide selected from the group consisting of glyphosate, glyphosate, and TBL.

In another preferred embodiment, at least one of the first and second promoters is specific to cells that are critical to pollen formation or function. In particular, either or both of the first and second promoters can be specific to other cells or, alternatively, to microspores. In yet another preferred embodiment, the second promoter is a promoter capable of induction.

In a recombinant DNA molecule as described above, the first DNA sequence can be lamS and the second lamH DNA sequence. Alternatively, the second substance and the second expression product can be carried in pairs as follows: 2-amino-4-methoxy-butanocium acid and methoxynine dehydrogenase or rhizobitoxin synthase; a non-toxic analog of glucuronic acid and β-glucoronidase; naphthalenacetamide or indolacetamide and indolacetamide hydrolase.

In accordance with another aspect of the present invention, there is provided a method of producing a male sterility plant, comprising the steps of

-1111 (A) introducing a recombinant molecule, as described above, into the genome of a pollen-producing plant, which is capable of being genetically transformed; and (B) developing the pollen-producing plant under conditions such that male sterility is achieved as a result of the expression of the first DNA sequence and the second DNA sequence, providing the male sterile plant. In a preferred embodiment, the method further comprises, after step (B), the steps of (C) crossing the male sterile plant with an isogenic, fertile plant to obtain the seed, then (D) developing the seed to produce progeny plants, and (E) selecting male sterility plants from the progeny through exposure to the selective agent.

In accordance with a further aspect of the present invention, there is provided a method of producing hybrid seeds from a male sterility plant, comprising the steps of:

(A) introducing a first recombinant molecule as described above, to the genome of a pollen-producing plant which is capable of being genetically transformed;

(B) developing the pollen-producing plant under conditions such that male sterility is achieved as a result of the expression of the first DNA sequence and the second DNA sequence, producing a male sterile plant; and

-1212 (C) crossing the male sterility plant with pollen from a male fertility line, pollen has integrated into its genome a second recombinant DNA molecule that comprises a selective marker gene and a selective marker gene expression regulatory promoter in plants.

In accordance with a preferred embodiment of this method, the second recombinant DNA molecule further comprises, (i) a third DNA sequence and (i) a third promoter that causes transcription of the third DNA sequence in plants, wherein the third DNA sequence is transcribed as an antisense RNA or encodes a gene product that negates the effect of the first DNA sequence or the second DNA sequence, so that the hybrid seed with restored fertility is produced through the male sterility plant, and the third promoter causes transcription, so that the expression of the third DNA sequence overlaps in time with the expression of the first sequence of DNA or the second DNA sequence. In an alternative, the third DNA sequence may have an anti-sense RNA transcript, which may be a repressor gene, or may encode a gene product that is a ribozyme, a dominant negative transactivator, or a transdominator.

In accordance with another preferred embodiment, selective marker genes impart herbicidal resistance.

Various products are also within the scope of the present invention. These include, for example,

-1313 hybrid seeds produced through a method as described above, a plant cell comprising elements (A) to (E) of the aforementioned recombinant DNA molecule, and a differentiated plant comprising such a cell.

The following advantages over other hybridization systems can be obtained with the present invention:

(a) Hybrid seed production is not as laborious and intensive and can be accomplished on a large scale with commercially acceptable costs.

(b) Male sterility is simply inherited and stable in response to environmental stresses that limit the effectiveness of self-incompatibility and CMS-based schemes.

(c) The seed that is produced will be relatively uncontaminated by the pure seed.

(d) The system avoids the use of defective cytoplasmic organelles that can decrease the performance of the hybrid seed.

(e) The system will greatly accelerate development and increase the number of lines that can be proven as origins in a hybrid crossing as it can be imposed on any plant or production line capable of being transformed and regenerated to plants without the inclusion of DNA additional genomic. Additionally, plant lines can be tested by combining skill prior to the inclusion of the hybridization system which can modify the breeding strategy.

-1414

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a representation of the construction of an antisense gene vector that was used for inhibition of anti-sense RNA of B-glucuronidase gene activity in transgenic plants.

Figure 2a is a schematic representation of a restriction map and coding region of clone number L 4, and a specific microspore clone isolated from a genomic collection of Brassica napus. The clone contains three similar copies of an individual gene. These genes are identified as Bp4A, Bp4B, and Bp4C. The first (Bp4A) and third (Bp4C) genes are functional, the second gene has modifications that most likely make it non-functional. The restriction map is done diagrammatically since the untranscribed regions are shown as a single line, while the transcribed regions are shown as an area in a box. The second gene (Bp4B) is identified based on sequence homology and is shown as an area in a box with a dotted line. The 220 annotation refers to a deletion / rearrangement of approximately 220 base pairs that presumably has inactivated the second gene (Bp4B) in this clone. The start of transcription is located on the leftmost side of each area in a box (except for the Bp4B gene) and the exon and intron positions are noted.

-1515 for exons that are filled in black and intron positions that are not filled. A small arrowhead is shown above the 5 'untranscribed region of each gene, this arrowhead serves to indicate the promoter region of each gene. Restriction sites are identified such that the number of the first nucleotide of the restriction enzyme recognition site is shown. No restriction sites are shown, only those relevant to the detailed constructions. Genes are represented with the 5 'region on the left side and the 3' region on the right side. The DNA sequence numbering is from left to right, 5 'to 3' in all cases.

Figure 2b is a schematic representation of a restriction map and coding region of clone number L 10, a specific microspore clone isolated from a genomic collection of Brassica napus. the clone contains an individual gene. Transcription initiation, exon, intron, and promoter positions are as seen in Figure 2a. Restriction sites are identified such that the number of the first nucleotide of the restriction enzyme recognition site is shown. Genes are represented with the 5 'region on the left side and the 3' region on the right side. The DNA sequence numbering is from left to right, 5 'to 3' in all cases.

Figure 2c is a schematic representation of the restriction map and coding region of clone number L 16, a specific microspore clone isolated from a collection

-1616 Brassica napus genomics. The clone contains a single gene that shows similarity to clone L 10. The positions of intron and exon are as shown in Figure 1. The restriction sites are identified, so that the number of the first nucleotide of the site of restriction enzyme recognition. Genes are presented with the 5 'region on the left side and the 3' region on the right side. The DNA sequence numbering is from left to right, 5 'to 3' in all cases.

Figure 2 is a schematic representation of the restriction map and coding region of clone number L 19, a specific microspore clone isolated from a genomic collection of Brassica napus. The gene contains an individual gene. The start of transcription, and the positions of exon, intron and promoter are as shown in Figure 1. The restriction sites are identified so that the number of the first nucleotide of the restriction enzyme recognition site is shown. Genes are presented with the 5 'region on the left side and the 3' region on the right side. The DNA sequence numbering is from left to right, 5 'to 3' in all cases.

Figure 3a is the complete nucleotide sequence of clone L 4 depicted in Figure 2a. Only the coding chain structure is shown, for clarity.

Figure 3b is the nucleotide sequence of the portion of clone L 10 shown in Figure as underlined in

-1717

Figure 2b. Only the coding chain structure is shown, for clarity.

Figure 3c is the nucleotide sequence of the portion of clone L 16 shown in Figure underlined in Figure 2C. Only the coding chain structure is shown, for clarity.

Figure 3d is the nucleotide sequence of the portion of clone L 19 shown in the Figure as underlined in Figure 2d. Only the coding chain structure is shown, for clarity.

Figure 4 is the nucleotide sequence of 3 cDNA clones isolated from a cDNA collection derived from the Brassica napus microspore. These clones are named cBp401, cBp405, and cBp408. These three cDNA clones are extremely homologous to members of the Brassica napus L 4 microspore specific gene family (Bp4A, Bp4B, Bp4C). The nucleotide sequence of two of the 3 members of the Brassica napus L4 microspore specific gene family is shown in this Figure (Bp4A, Bp4C). The Bp4C gene was chosen as a master sequence for comparison. The deduced nucleotide coding sequence for the Bp4A and Bp4C genes is shown as a sequence from which the two exons of the genes have been split together at the normally split positions in vivo. This gives rise to the coding sequence in the mature mRNA. The cDNA clones are aligned with the Bp4C sequence of

-1818 so that only nucleotide changes are shown. The sequences, therefore, are represented as variants of an individual master sequence of the Bp4C gene which is shown in line 1. The ATG start codon as well as the TGA or TAA stop codons are underlined. These three cDNA clones correspond to related members of the Brassica napus microspore-specific gene family, a portion of which is contained in clone L4.

Figure 5 shows a partial nucleotide sequence of a cDNA clone that is closely homologous to the gene contained in clone L10, the restriction map of which is shown in Figure 2b.

Figure 6 is the nucleotide sequence of the cDNA clone that is from the gene product of clone L19, the restriction map of which is shown in Figure 2d.

Figure 7 (7A, 7B, 7C, 7D) are schematic representations describing the production of vectors containing the promoter and promoter regions of the L4 clone. Specific examples are discussed in more detail below. Figure 7E is a schematic representation of the promoter constructs produced as schematically shown in Figures 7A to 7D.

Figure 8 is a schematic representation describing the production of vectors containing the regions

-1919 promoters of clone L10, the details of which are discussed below.

Figure 9 is a schematic representation describing the production of vectors containing the promoter regions of clone L19, the details of which are described below.

Figure 10 is a schematic representation of a restriction map of a Brassica napus genomic clone containing a gene (termed HP1O1) that is constitutively expressed at high levels in all cells including cells that develop pollen. The portion of the clone that was used to provide promoter regions for anti-sense RNA production is shown. This construct gives rise to an anti-sense RNA that contains a region of the RNA transcribed from this gene.

Figure 11 is a schematic representation depicting the production of a gene encoding a protein of po I i I isin a.

Figure 12 is a schematic representation depicting the production of an anti-sense gene specific to the intro region of clone L19 and a restorer lacking the intron region that is targeted for inhibition of anti-sense RNA.

Figure 13 is a schematic representation depicting the production of clones containing versions of a ricin A chain coding region.

-2020

Figure 14 is a histogram showing GUS activity in plants transformed with sense and anti-sense GUS genes.

Figure 15 illustrates the procedure used for the isolation of T-DNA gene 2 (lamH: indole acetamide hydrolase gene) from Agrobacterium tumefaciens plasmid Ti derived from pPCV 311 (described by Koncz, C. and Schell, J., Molecular and General Genetics, 1986, 204: 383-396) and the construction of a promoter-free version of this gene.

Figure 16 illustrates the procedure used for the isolation of T-DNA gene 1 (lamS: indole acetamide synthase gene) from the Agrobacterium tumefaciens plasmid Ti derived from pPCV 311 and the construction of a promoter-free version of this gene.

Figure 17 illustrates the production of hybrid seeds using the binary cryptocytotoxicity method using the lamH and lamS genes described in Figures 15 and 16.

Figure 18 illustrates the segregation patterns of the laMH e lams genes in the F1 and F2 populations when the genes are in the same segregation unit.

Figure 19 is an alternative representation of the procedure used to isolate the T-DNA gene 2 (the lamH gene) from Agrobacterium tumefaciens and the production of a promoter-free version of this gene.

-2121

Figure 20 illustrates hybrid seed production using a variation of the binary cryptotoxicity method and a herbicide resistance gene. Genes 1 and 2 can be the lamS and lamH genes. The herbicide resistance gene can be any resistance gene against any herbicide. The herbicide resistance gene has a dual role: it results in a hybrid plant and seed protected from herbicide treatment and is essential for selection of origin A when the origin is maintained through a cross with origin A¹. The resulting hybrid seed is essentially 100% herbicide resistant and is restored with respect to male fertility.

Figure 21 illustrates the production of hybrid seeds with origin A being produced and maintained as shown in Figure 20. Origin B is homozygous for herbicide resistance. The hybrid is essentially 100% herbicide resistant.

Figure 22a illustrates the use of the binary cryptotoxicity method for the production of an herbicide resistant origin A of male sterility. The resulting Origin A is essentially 100% male sterility with G1 and G2 on different chromosomes. Origin A is resistant to herbicide.

Figure 22b illustrates the method used to maintain origin A shown in Figure 22a. Observe the use of herbicide spray to maintain the origin A of male sterility.

Figure 22c illustrates the crossing of origin A shown in Figure 22a with origin B. Origin B is the result of a

-2222 transformation, where the restorer gene and the herbicide resistance gene are linked into a recombinant DNA molecule. The result of the crossing of origin A and origin B is essentially a hybrid that is restored to male fertility and is herbicide resistant.

Figure 23 is a map of plasmid pPHP5838 which contains the Bp10 promoter operably linked to the IAMS gene and to the 3 'untranslated region.

Figure 24 is a map of plasmid pPHP5839 which contains the Bp10 promoter operably linked to the gene and IAMH to the 3 'untranslated region.

Figure 25 is a map of plasmid pPHP5840, which is a TA39 (8B3) promoter and the 0 'untranslated reader operably linked to the IAMS gene and the 3' Pl-lll untranslated region.

Figure 26 is a map of plasmid pPHP5841, which contains the TA39 (8B3) promoter and the operably untranslated reader linked to the IAMH gene and the 3 'Pl-ll untranslated region.

Figure 27 is a partial sequence of pPHP5838 where the nucleotide coordinates correspond to those used in the map in Figure 23. Few nucleotides are represented by small letters to indicate insecurity at that point in the sequence.

-2323

Figure 28 is a partial sequence of pPHP5839 where the nucleotide coordinates correspond to those used in the map in Figure 24. Few nucleotides are represented by small letters to indicate insecurity at that point in the sequence.

Figure 29 is a partial sequence of pPHP5840, where the nucleotide coordinates correspond to those used in the map in Figure 25. Some nucleotides are represented in small letters to indicate insecurity at that point in the sequence.

Figure 30 is a partial sequence of pPHP5841, where the nucleotide coordinates correspond to those used in the map in Figure 26. Some nucleotides are represented in small letters to indicate insecurity at that point in the sequence.

DETAILED DESCRIPTION OF THE INVENTION

As mentioned above, the invention relates to recombinant DNA molecules for use in preparing a plant or seed that has male sterility or that carries a male sterility trait. The male sterility plant is created in order to facilitate hybrid crossings. The invention also relates to the use of recombinant DNA molecules introduced into the male fertility origin of

-2424 such crosses, resulting in a restored hybrid seed regarding fertility. Recombinant DNA molecules to be used to produce male sterility contain one or more DNA sequences, which may be the same or different, 5 which encode a gene product. Gene products include a nucleotide sequence (in particular an RNA sequence), a protein, or a polypeptide. When produced in a cell and / or plant tissue, which is essential to pollen formation and / or function, the gene product is directly or indirectly capable of interfering with cell function and / or development. The presence of the recombinant DNA molecule can be selected if the DNA also carries a selective marker gene, preferably for herbicidal resistance. A plant cell and / or tissue that is essential to pollen formation and / or function includes cells and / or tissues that are instrumental in the development or function of pollen, including cells and / or tissues from which pollen (for example, premiotic microspore cells or uninucleates), cells and / or tissues are developed, which are part of the male structure where pollen develops (for example, anther, tapetum or filament) and the same pollen. The recombinant DNA molecule may also contain one or more promoters, which may be the same or different, which are capable of regulating the expression of the DNA sequences in the recombinant DNA molecule. The DNA sequences and promoters in the recombinant DNA molecule are selected so

-2525 that the gene product selectively interferes with the function and / or development of a plant cell that is essential to the formation and / or function of pollen.

The recombinant DNA molecule of the invention can contain one or more DNA sequences which can be the same or different. DNA sequences can be selected from the following DNA sequences:

(a) a DNA sequence, which codes for an RNA when produced in a plant cell, which is essential to the formation and / or function of pollen, substantially interferes with the expression of:

(i) a gene which is essential to the formation and / or function of pollen;

(I) a gene, which is essential to the continued development and / or function of all metabolically competent cells in a plant; or (iii) a gene that confers cells with a plant resistance to a chemical agent or physiological stress;

(b) a DNA sequence that encodes a protein or polypeptide, which when produced in a plant cell that is essential to the formation and / or function of pollen, is substantially cytotoxic to the cell; and (c) a DNA sequence that encodes a protein or polypeptide that when produced in a plant cell that is

-2626 essential to the formation or function of pollen, makes a non-toxic substance substantially cytotoxic to the cell, and (d) a DNA sequence that encodes the gene product, which when produced in a plant cell that it is essential to the formation and / or function of pollen, it makes the cell susceptible to chemical agents or physiological stress.

(e) A DNA sequence that encodes a gene product, which can be converted to a substance that is cytotoxic to cells / tissues essential for the formation and / or function of pollen.

The DNA sequences above (a) to (e) are further described below in relation to the various embodiments of the invention.

According to a first aspect of the present invention, gene male sterility can be produced by transforming plant cells that are capable of regeneration to a fully differentiated plant, with a recombinant DNA molecule, containing an antisense gene, which encodes a gene product, particularly RNA, that is capable of interfering with the expression of a target gene. Preferably, the recombinant RNA molecule encodes RNA, which is complementary to, and capable of hybridizing to, the RNA encoded by the target gene.

An antisense gene is a DNA sequence produced when a gene is inverted relative to its normal display for transcription. An antisense gene can be constructed in a number of different ways, as long as it is able to interfere

-2727 with the expression of a target gene. Preferably, the antisense gene is constructed by inverting the coding region of a gene relative to its normal display for transcription to allow transcription of its complement; therefore, the RNA encoded by the antisense and the target gene is complementary. It is understood that a portion of an antisense gene incorporated into the recombinant RNA molecule of the invention may be sufficient to selectively interfere with the expression of a gene, and thus the term antisense gene used herein encompasses a functional portion of the antisense gene.

The target gene for inactivation is a gene expressed in plant cells and / or tissues that are essential for the formation and / or function of pollen. Preferably, the gene is expressed only at specific stages during the development of pollen, the regulation of which is tightly controlled. The gene may also be a gene that is essential for the continued development or function of all metabolically competent cells / tissues such as, but not limited to genes involved in essential cell structures, essential metabolism, including essential biosynthesis. The gene can also be a gene that confers on the cells / tissues of a plant, resistance to a chemical agent or physiological stress.

It is understood herein that it is not essential to identify and isolate the new target gene. Rather, the gene may have been described in the literature or obtained commercially.

-2828

Isolation of genes that are essential to pollen formation and / or function can be accomplished through a variety of procedures. A detailed description of a procedure to isolate such genes is set forth below.

Since the gene in question can initially be isolated in its native form, it should be understood that the term gene as used herein, may refer to one or more parts of the gene including 5 'untranslated leader sequences, functional sequences in the ribosome binding, coding sequences, promoter sequences, intron sequences and 3 'non-transcribed sequences, or any substantial fragment of these sequences.

The inventors herein have isolated DNA sequences from a plant of the Brassica napus spp oleifera w Westar species, which are expressed only in microspores or whose expression is essential to the function of the microspore and / or its development, and therefore a preferred embodiment of the present invention provides a recombinant DNA molecule containing one or more antisense genes to these DNA sequence isolates. A schematic representation of the restriction maps and coding regions of specific genes in microspores identified as L4, L10, L16, and L19 are shown in Figures 2a, 2b, 2c, and 2d, respectively. The complete nucleotide sequence in clones L4 and relevant sequences of L10, L16 and L19 are shown in Figures 3a, 3b, 3c and 3d, respectively. Nucleotide sequences of cDNA clones

-2929 isolates corresponding to related genes or gene family members within clones L4, L10 and L19 are shown in Figures 4, 5 and 6, respectively. In a preferred embodiment of the invention, the recombinant DNA molecule contains one or more than one antisense gene to the specific microspore genes identified as L10 and L19.

It is believed that microspore-specific DNA sequences homologous to those Brassica napus isolates described herein will be found expressly and exclusively in the pollen of other pollen-bearing plant species, particularly plant species within the Brassica genus and the Cruciferae family (also known as Brasicaceae), and more particularly other crops of Brassica napus.

The occurrence of specific microspore sequences isolated from Brassica napus in other pollen-bearing plant species can be routinely ascertained through known hybridization and DNA polymerase extension techniques. It is believed that the similarity of plant genes from species to species will allow the modalities of the present invention to be performed using the DNA sequence in a number of pollen-bearing plant species that are capable of being genetically transformed. The universality of plant genes has been extensively documented in the literature and plant genes homologous to plant actins have been described (Shah, DM, et al., J.

-3030

Mol. Appl. Genet. 2: 111-126, 1983), phytochrome (Hershey, HP, et al., Proc. Nati. Acad. Sci. USA 81: 2332-2337, 1984) storage proteins (Singh, NK et al., Plant Mol. Biol . 11: 633-639, 1988) enzymes such as glutamine synthases (Lightfoot, DA, et al., Plant. Mol. Biol. 1 1: 191-202, 1988, and references within) and nitrate reductase (Cheng, C. , et al., EMBO Jour. 7: 3309-3314). These and other examples in the literature clearly show that many plant genes are highly conserved. It is also evident that this conservation applies not only to structural proteins, but also to enzymatic proteins important to cellular physiology. Therefore, it is believed that DNA sequences, when found in other plant species, will be essential to microspore development and will be useful in carrying out the present invention in such species.

It has also been shown that antisense RNA sequences derived from one plant species can effectively inhibit the expression of homologous DNA sequences in a different species. Van der Krol, et al., 1988, Nature 333: 866-869. Therefore, it is expected that antisense RNA derived from all or part of the specific DNA sequences in the Brassica microspore will be functional in other plants. Furthermore, identification of the homologous gene in other plants through the above-mentioned hybridization and / or DNA polymerase extension techniques will allow one skilled in the art, if necessary, to design an antisense gene, which will be more

-3131 precisely complement the target gene and will be more effective in the inactivation of such a gene.

As mentioned above, the target gene may also be essential for the development and / or function of all metabolically competent cells / tissues, such as, but not limited to, genes involved in essential cell structures, essential biosynthesis, and essential metabolism. Examples of such sense genes can be found in the literature and include genes encoding actin, tubulin, or ubiquitin; three 10 proteins which are essential for cell growth and development.

Actin gene sequences isolated from plants have been published (eg, Baird WV, and Meagher, RB, EMBO J. 6: 3223-3231, 1987, or Shah, DM, Hightower, RC and 15 Meagher, RB, Proc Nati Acad Sci USA 79: 1022-1026, 1982) and actin is known to play a critical role in normal cellular function, especially during mitosis and meiosis where actin is part of the cellular apparatus for cell division.

The sequence for tubulin from 20 plants has also been described (Raha, D., Sen, K. and Biswas, BB Plant Mol Biol 9: 565571, 1987). Tubulin, like actin, is known to be important in the cell life cycle particularly with respect to the shape, transport, and spindle formation of the cell during mitosis and meiosis.

-3232

The DNA sequence for plant ubiquitin has also been published (Gausing, K. and Barkardottir, R. Eur J. Biochem 158: 57-62, 1986). Ubiquitin is a protein involved in the production of cellular proteins and as such has a critical role in regulating specific levels of cellular protein. Furthermore, ubiquitin is one of the most highly conserved proteins in eukaryotic cells. Interference with ubiquitin expression can cause abnormalities in the production of cellular proteins.

If any of the aforementioned proteins are not present, or are present, but not in a normal functional form in the cell, proper cellular function is interfered with, and the cell fails to develop properly.

A gene found to be essential for the continued development or function of all metabolically competent cells in one plant species is believed to have a similar counterpart in another plant species, as it is generally understood that within the plant kingdom there are genes that are absolutely identical or very homologous involved in the basic procedures that control or are the result of cell development. Furthermore, it is believed that a gene encoding a gene product which interferes with the expression of the gene (i.e. an antisense gene) in one plant species will have the ability to do so in other plant species.

-3333

The tissue-specific and developmentally regulated expression of a wheat endosperum protein synthesized in tobacco plants genetically transformed with this wheat gene has been reported (Flavell, RB, et al., Second Intemational Congress for Plant Molecular Biology, Abstract # 97). In this example, the wheat gene worked in the tobacco plant in a way identical to the way it works in a wheat plant. Other literature clearly shows the regulation of a specific gene, which can be complex in many cases, is maintained in transgenic plants. An example of this is the phytochrome-mediated regulation of a wheat chlorophyll binding protein a / b in transgenic tobacco (Nagy, F. et al., EMBO Jour. 5: 11191124, 1986). In this example, the specific regulation of the light response of the wheat gene was maintained in the foreign genetic environment. Not only do cereal genes work in a conserved way, but genes from other plant species that are more closely related maintain functionality in heterologous genetic systems. Pea seed proteins are appropriately expressed in tobacco plants (Higgins, TJV, et al., Plant Mol. Biol. 1 1: 683-696, 1988), as are soybean proteins, (Barker , SJ, et al., Proc. Nati. Acad. Sci. USA 85: 458-462, 1988) and pea rcbS genes (Nagy, F. et al., EMBO Jour. 4: 3063-3068, 1985). The specific literature has numerous other examples of genes that have been used to genetically transform plants to

-3434 those genes that maintain their ability to function properly in this new genetic environment. Therefore, the conserved nature of these genes, not only in the DNA sequences that control the expression of these genes, but the current structure of the protein encoded by these genes, is similar between plant species.

As mentioned above, the target gene may be a gene that confers on the cells of a plant resistance to a chemical agent or to a naturally or artificially incurred physiological stress. Such a gene can be native to a plant cell or it can be a foreign gene, for example a gene from another plant species. If the gene is a foreign gene, it can be introduced into the genome of a plant cell before, after, or concurrently with the antisense gene as will be described later. Examples of such genes can be identified in the literature and include genes that confer resistance to a herbicide such as acetolactate synthase, which confers resistance to chlorosulfuron; genes that confer resistance to physiological stress such as the gene encoding dismutase for its superoxide, which actively reduces oxygen radicals; genes conferring resistance to UV light and wounds (phenyl lylase and alanine-ammonia or chalcone synthase); genes that confer resistance to disease or pests (eg, 3-glucanases, chitinase, or proteinase); genes that confer resistance to antibiotics and certain toxic drugs such as the gene that

-3535 encodes neomycin phosphotransferase II that confers resistance to kanamycin.

The recombinant DNA molecule containing an antisense gene of the present invention further contains a promoter, which regulates the expression of the antisense gene. If more than one DNA sequence containing an antisense gene (or one or more than one gene encoding a cytotoxic substance or a gene that confers susceptibility to a chemical agent or physiological strain) is present in the recombinant DNA molecule of the invention DNA sequences can be regulated through an identical promoter or each can be regulated through a different promoter.

If the target gene is a gene that is only expressed in plant cells / tissues that are essential to pollen formation and / or function, then a promoter that works in all, many, or a variety of cell types including cells / tissues essential to the formation and / or function of pollen can be used in the construction of an antisense gene. An example of such a constitutive promoter is CaMV 35S or preferably HP 101, which has been isolated from Brassica napus as described below. In this way, the promoter that is used can be constitutively active in all or many of the cell types, but the antisense gene whose expression is regulated through such a promoter, will only interfere with the expression of the target gene in those cells / tissues. that produce the target gene in question, is

-3636 say, cells / tissues essential to the formation and / or function of pollen.

It is also possible to use a promoter capable of being induced to regulate the expression of the antisense gene. A promoter capable of induction is a promoter that is capable of directly or indirectly activating transcription of one or more DNA sequences or genes in response to an inducer. In the absence of an inducer, the DNA or gene sequences will not be transcribed. Typically, the protein factor that specifically binds a promoter capable of being induced to activate transcription is present in an inactive form, which is then directly or indirectly converted to the active form through the inducer. The inducer may be a chemical agent such as a protein, metabolite (sugar, alcohol, etc.), a growth regulator, herbicide, or a phenolic compound, or a physiological stress such as heat, salt, toxic elements, etc., or the action of a pathogen or disease agent such as a virus. A plant cell containing a promoter capable of being induced can be exposed to an inductor by externally applying the inductor to the cell, such as by spraying, water, heating, or other similar methods.

Examples of promoters capable of induction include the 70 KD heat shock promoter capable of induction of d. melanogaster (Freeling, M., Bennet, DC, Maize DNA 1, Ann. Rev. of Genetics 19: 297-323) and the alcohol dehydrogenase promoter which is induced through ethanol (Nagao, RT, et al. , Miflin, BJ,

-3737

Ed. Oxford Surveys of Plant Molecular and Cell Biology, Vol. 3, p. 384-438, Oxford University Press, Oxford, 1986). The promoter capable of being induced can be in an Induced state through pollen formation or at least one during a period corresponding to the transcription of the sense gene. A promoter that is capable of being induced through a simple chemical is particularly useful, since the male sterility plant can easily be maintained through self-pollination when it grows in the absence of such a chemical.

Alternatively, if the target gene is an essential gene for the development and / or function of all metabolically competent cells / tissues or a gene, which confers on the cells of a plant resistance to a chemical agent or physiological stress, a producer that is active only in cells / tissues essential to the formation and / or function of pollen (for example, a promoter essential for the development and / or function of pollen) is preferably used to regulate the expression of the antisense gene.

The invention also relates to a promoter essential for the development and / or function of pollen or functional portions of the same promoter and chimeric promoters using the promoter essential for the development and / or function of pollen, and / or its portions.

An essential promoter for the development and / or function of pollen is a DNA sequence that selectively regulates the expression of a DNA sequence in the cells / tissues of a

-3838 plant essential for the formation and / or function of pollen, and / or limits the expression of such a DNA sequence to the period of pollen formation in the plant.

Preferably, the promoter essential for the development and / or function of pollen is a DNA sequence that corresponds to the promoter sequence in the specific microspore genes identified as L4, L10, L16 and L19, herein, or their functional fragment. ; or a chimeric promoter sequence containing one or more of a promoter sequence of the microspore specific genes identified as L4, L10, L16 and L19 or portions of such promoter sequences. The preferred promoter essential for the development and / or function of pollen can be used in conjunction with naturally occurring flanking or transcribed coding sequences described herein, or with any other transcribed or coding sequence that is essential for the formation and / or function of pollen.

The promoters essential for the development and / or function of pollen named above, were isolated from a plant of the Brassica napus species. It is believed that it is possible to use these promoters to limit the expression of a given DNA sequence to the formation and / or function of pollen and to a specific period during the formation and / or development of pollen from a plant of a different species. Published scientific literature has clearly shown that plant genes are universal and that specific promoter fragments in plant tissue retain

-3939 its function in other species. For example, the wheat endosperm promoter fragments function to give appropriate seed-specific expression in tobacco (Simpson, J. et al., EMBO Jour. 4: 2723-2729, 1985) and the alcohol dehydrogenase promoter. (Adh-1) from maize (Zea mavs) can be used in conjunction with other promoter fragments to give appropriate expression in tobacco (Ellis, JG, et al., 1987, EMBO J. 6: 11-16). Furthermore, the corn transposition element, Ac, is active in tobacco and other plant species (Tayklor, et al., 1989, Plant Mol. Biol. 13: 109118), providing further evidence of the versatility of the structure and function of the plant gene. These examples demonstrate the tissue-specific role of the same promoters in widely divergent species (monocotyledons to dicotyledons). Research studies have shown the role of the equivalent promoter in more closely related species within the same family or between families such as Solanaceae and Brassicaceae. In a preferred application, the promoter isolated from the anther-specific L10 gene discussed above is the promoter used to regulate expression of the antisense gene.

In a further preferred application, the tobacco anther-specific TA39 promoter is employed to regulate expression of the antisense gene. The anther-specific TA39 promoter is from a genomic clone of a commercially available genomic collection of Nicotiana tabcum L. var NK 326.

-4040 (Clonetech). The collection was applied to a probe with TA39 of specific cDNA in anther obtained from r. Robert B. Goldberg of UCLA. Isolation of TA39 and its expression characteristics have been reported (Koltunow, AM, Truettner, J., Cox, KH, Wallroth, M. and Goldberg RB Different temporal and spatial gene expression patterns occurr during anther development. Plant Cell. 2: 1201-1224, and Goldberg, RB, Beals, TP, and Sanders, PM 1993. Anther Development: Basic Principles and Practical Applications. Plant Cell 5: 1217-1229 (1990).

However, it is appreciated that refinements in function may be required for individual plants or species to maximize or modulate control of the appropriate time or level of expression to carry out aspects of the invention. Accordingly, there is provided herein, method for modification of promoters to modify or improve function in several plants of different origin.

It is understood that there is no high degree of DNA sequence homology between the native promoter essential for the development and / or function of pollen from clones L4, L10, L16 and L19. Test data reveals that the timing and expression level of these genes is not identical in pollen, but that they all overlap in activity at some point.

It is expected that any number of different promoters essential for the development and / or function of pollen can be used to regulate the expression of a DNA sequence in a

-4141 recombinant DNA molecule of the invention. In order to determine which promoter essential for the development and / or function of pollen can be used to regulate the expression of an antisense gene (or other sequences that will be discussed later with respect to other modalities of the invention), it is necessary to have consideration of a number of factors.

The promoter essential for the development and / or function of the pollen used must be a promoter (or a modified form of the promoter) that is active at the appropriate time to produce sufficient levels of transcribed RNA to carry out the invention. The promoter essential for the development and / or function of pollen derived from pollen-specific clones described herein (mainly those derived from microspore genes identified as L4, L10, L16, and L19) that are readily active in the development of microspores such that gene expression occurs both during and after meiotic or mitotic division of pollen stem cells.

Thus, the activity of these promoters is not limited by segregation.

When a promoter essential for pollen development and / or function is used to inactivate a gene that is essential to pollen formation and / or function, as discussed above, it can be difficult to predict, a priori, which potential promoter for the development and function of pollen or modified promoter construction will effectively block the function of such a gene. I know

-4242 prefers to use a promoter essential for the development and / or function of pollen that exhibits a development pattern similar to the gene. A convenient method of determining when the target gene for interference is expressed is to isolate RNA from developing microspores at different stages and to analyze this RNA for gene expression through so-called Northern staining analysis. This procedure will allow the determination of the development period in which the gene is expressed. In order to determine the period of development in which a promoter essential for the development and / or function of pollen is expressed to a gene, a similar series of analyzes can be carried out using a reporter gene such as β- as a probe. glucuronidase bound to the promoter essential for the development and / or function of pollen, or a native gene of the same plant species from which a promoter essential for the development and / or function of pollen is isolated and whose expression is regulated through the promoter essential for the development and / or function of pollen. When the promoter essential for the development and / or function of pollen is isolated from one plant and used in a different plant species, the preferred method is to use a reporter gene linked to the promoter to determine the exact control of development time that the promoter has on that particular species of plant.

It is understood that the activity of a promoter essential for the development and / or function of pollen, if it is intended for the same or a different species, can be modified in structure

-4343 to change or alter activity in a plant. Changes that are contemplated include, but are not necessarily limited to: addition or deletion of sequences, orientation of upstream or downstream sequences and the inclusion of introns or parts of the coding sequence of the specific gene in pollen. The above modification may serve to increase expression or improve regulation of expression at the target stages of development.

It should be noted that the identification of a promoter region (including constitutive promoter, capable of being induced and essential for the development and / or function of pollen) is usually defined by function rather than by a fixed DNA sequence. Two hundred (200) nucleotide bases or less of a promoter sequence may be sufficient to maintain promoter function. It should also be recognized that some upstream DNA sequences can be arranged in opposite orientations and still retain or demonstrate improved promoter function. Furthermore, enhancer-type DNA sequences, which are usually small conserved DNA sequences ranging in size from less than 10 nucleotides to considerably larger numbers of nucleotides, can also be inserted into the promoter regions to enhance expression.

It may also be desirable to include some intron sequences in the promoter constructs, since the inclusion of intron sequences in the coding region may result in

-4444 improved expression result. In this way, it may be advantageous to bind the DNA sequences to be expressed to a promoter sequence containing the first intron and exon sequences of a polypeptide, which is unique to cells / tissues of a plant essential for the formation and / or function of pollen.

Additionally, regions of a promoter can be linked to regions of a different promoter in order to obtain the desired promoter activity. Specific examples of chimeric promoter constructs are the chimeric promoters contained in the 10 vectors PAL1107 and PAL1106.

It is desirable that the promoter essential for the development and / or function of pollen (or constitutive or promoter capable of being induced or whichever is selected to construct a recombinant DNA molecule of the invention) works in such a way that sufficient levels of Antisense RNA to substantially interfere with expression of the target gene. Investigations of the mechanism of antisense RNA inhibition of gene expression in model systems have suggested that levels equal to or greater than equal levels of antisense RNA 20 may be required in order to observe a significant reduction in gene activity. objective. However, in some cases, it is observed that low levels of antisense RNA may have a specific reduction in sense gene activity. Therefore, in some cases if the gene whose object is inactivation through antisense RNA is a gene that

-4545 found essential for the continued development or function of all metabolically competent cells or in all cell types at a low level, an excess of antisense RNA may not be required for inhibition. Furthermore, less than the total reduction in gene activity may be more than enough to interrupt the development of pollen, which is known to be highly sensitive to many stressful conditions. Therefore, it is suggested that the promoter essential for the development and / or function of the pollen that is used, performs certain aspects of this invention. It has been chosen based on the observation that the essential promoter for the development and / or function of pollen works to cause the expression of any of the sequences adjacent to it that will be transcribed at the time that the period that the pollen is parallel or overlaps. target gene is inactivated and expressed and that the levels of antisense RNA expressed from the antisense gene are of sufficient levels to inhibit the expression of the target gene, usually at medium levels greater than, or equal to, sense RNA levels.

By using an essential promoter for the development and / or function of pollen to regulate the expression of the antisense gene, it is possible to interfere with, for example, a normal microspore development in any given plant, without first having to isolate from the plant's genomic DNA a gene which is essential for the development of the microspore. In this way, a male sterility plant can be produced where the target gene for

-4646 interference is a gene that is essential for the development and / or function of all metabolically competent cells / tissues. To produce a male sterility plant, such a gene is specifically interfered with using a recombinant RNA molecule containing a promoter essential for the development and / or function of pollen to regulate transcription of the antisense gene.

Furthermore, a male sterility plant can be produced by developing a plant, which has a recombinant DNA molecule that comprises an antisense gene to a gene that confers on the cells of a plant resistance to a chemical agent or to a physiological stress of existence naturally or artificially induced and an essential promoter for the development and / or function of pollen. The transcription product of the antisense gene will interfere with the expression of the target gene in cells / tissues essential for the formation and / or function of pollen. The gene confers stress resistance on the rest of the plant.

In order to interfere with the expression of a gene, it is preferred that the antisense gene and the target gene be expressed at approximately the same time. Coincident expression can be achieved in a variety of ways using combinations of constitutive, organ-inducing, and organ-specific promoters (eg, a promoter essential for pollen development and / or function as described above). However, co-incident expression can be easily achieved by regulating the

-4747 expression of the antisense gene with the same promoter that controls the target gene, thus causing both to be transcribed in the same time frame. The concept of regulating gene expression using antisense genes is described in Weintraub, H. et al., Antisense RNA as a molecular tool for genesis analysis, Reviews Trends in Genetics, Vol. 1 (1) 1986.

It is preferable that the recombinant DNA molecule containing an antisense gene of the invention also contains one or more selection marker genes, which encode a selection gene product that confers resistance to plant cells / tissues. chemical agent or physiological stress, such that plant cells transformed with the recombinant RNA molecule or plants containing such transformed plant cells can be readily selected using a selective agent. Any effective selective agent for which a resistance gene has been identified can be used to select transformed plant cells. The selective gene used to select plant cells transformed with the recombinant DNA of the invention is said to fall into two broad or mutually exclusive categories, a chemical agent and a physiological strain. Examples of selective agents that can be used to select plant cells are herbicides, antibiotics, toxic substances and plant diseases, in particular, glufosinate, glyphosate, tabtoxinin-B-lactam (TBL), chlorosulfuron, kanamycin, hygromycin, methotrexate, bleomycin and

-4848 phosphinothricin. Preferred selection marker genes that can be used in the present invention to facilitate selection of transformed cells include the gene encoding the enzyme hygromycin phosphotransferase that confers hygromycin and enzyme neomycin phosphotransferase (NPT II) resistance to which confers resistance to kanamycin. Highly preferred resistance markers include genes for resistance to herbicidal agents such as glufosinate (their trade names are Basta, Liberty) glyphosate (trade name is Roundup) and TBL.

Examples of selective agents that can be used to screen for male sterile plants produced using a recombinant RNA molecule of the invention include a chemical agent such as a herbicide which can be used on a commercial scale and therefore may provide an economic benefit during the additional plant development season and separate from the use of the herbicide as a selective agent of a plant transformed through recombinant DNA. Resistance genes include any genes that allow the transformed cell to survive treatment with the selective agent. This includes, but is not limited to, a detoxifying agent, interference with the transport of the selective agent, overproduction of the target molecule through the selective agent. Examples of herbicides for which a resistance gene has been identified that may be useful as selective agents for plants are glyphosate whose trade name is Roundup.

-4949 (described in Comai, L., Facciotti, D., Hiatt, WR, Thompson, G., Rose, RE, Stalker, DM, 1985, Nature, Vol. 317, pages 741744), chlorosulfuron (described in Haughn, GW, and Somerville, CR, 1986, Mol. Gen. Genet., Vol. 210, pages 430-434) and phosphinothricin (Murakamí T, et al., Mol. Gen. Genet. 205: 42-50, 1986). Another herbicide for which a resistance gene or DNA fragment containing a resistance gene was reported is TBL (Kinscherf et al., 1991, J. Bact. 173: 4124-32, and from yeast, Marek, ET and Dickson, RC, 1987, J. Bact. 169: 2440-8).

Genes that encode proteins or polypeptides that can provide a distinctive phenotypic characteristic to plant cells, such as a gene that encodes a protein providing distinguishable color or morphology to plant cells, and genes that encode proteins which confer tolerance to Strains such as the metallothione gene can also be used as selection marker genes to facilitate selection of transformed plants.

For a general reference on selective agents see Weising, K., Schell, j., And Kahl, G. Ann. Rev. of Genetics, 22: 421477 (1988).

The selection marker gene can be expressed in particular cells or constitutively in the entire plant depending on the nature of the selection gene product. The expression of the selection marker gene is regulated through one or more selection marker promoters. The promoters

-5050 selection gene markers can be isolated using techniques known in the art and include constitutive promoters such as CaMV 35S.

To facilitate the selection of plant cells transformed with a recombinant DNA molecule or male sterile plants containing a recombinant DNA molecule, it is preferred that a selection marker gene be contained in the recombinant DNA molecule, most preferably linked to the DNA sequence contained in the recombinant DNA molecule. When the recombinant DNA molecule contains more than one DNA sequence, it is preferred since each DNA sequence can be linked to a different selection marker gene. Some embodiments of the invention are contemplated that at least two recombinant DNA molecules be used to produce a male sterile plant and in such embodiments, it is preferable that each recombinant DNA molecule contain a selection marker gene, preferably linked to the sequence DNA.

In a hybrid seed production scheme, where there are alternating rows of male sterility plants and male fertility plants, it is advantageous to make the final selection of male sterilities in the field together with male fertility donors. Therefore, it is desirable if the appropriate male fertility donors are pre-transformed to resistance to the selective gene to avoid having to

-5151 selectively apply the selective gene to the rows of male sterility plants.

It should be understood that a plant sought to become male sterile does not need to be transformed with a selection marker gene that confers resistance to a chemical agent or to naturally or artificially induced physiological or chemical stress if this gene is native to the plant.

The recombinant DNA molecule containing an antisense gene of the present invention preferably also contains a termination sequence and / or other transcriptional regulatory signals. Examples of termination sequences, which can be used in the recombinant DNA molecules of the invention, are the terminators not found in pRAJ-221 (Clonetech Laboratories, Palo Alto, CA).

A detailed description of the isolation and identification of genes, which are only expressed in cells / tissues of a plant essential to the formation and / or function of pollen and promoters essential for the development and / or functions of pollen, is set out below:

To isolate genes, which are only expressed in plant cells / tissues that are essential to pollen formation and / or function and promoters essential for pollen development and / or function, a genomic collection of DNA can be constructed from plant from DNA isolated from fresh young leaves according to normal methodology (Molecular Cloninq, to ​​Laboratorv

-5252

Manual Maniatis, T., Fritsch, EF, and Sambrooks, J., Coid Spring Harbor Laboratory, Coid Spring Harbor, New York, 1982) and sieved with probes derived from various tissues, one of which can be made from specific RNA to cells / tissues of a plant that are essential to the formation and / or function of pollen (hereafter referred to as pollen-specific RNA). The other probes must be made from RNA from different tissues in order to represent genes expressed in plant tissues that cannot be expected to include genes that are expressed in plant cells / tissues that are essential to formation and / or function of pollen. Examples include, but are not limited to, tissues such as leaves, roots, seeds, stigma, stem, and other organs. Some genes are expressed in all tissues and only a few in a restricted number of tissues, compared to many plant tissues, it is possible to isolate genes expressed exclusively in plant cells / tissues that are essential to formation and / or function of pollen.

Pollen-specific RNA can be isolated from plant cells / tissues that are essential to the formation and / or function of pollen that are early in the last stage of uninucleate. Although it is possible to isolate pollen-specific RNA at other stages, isolation of pollen-specific RNA can be technically difficult in the early stages of development; older cells may have limited nuclear gene activity; and, promoters may not be suitable to be used as

-5353 described in the present invention. For a detailed description of isolation of pollen-specific RNA see Example 1A herein.

Pollen-specific RNA can be labeled for the purpose of detection. It is desirable to make radioactive cDNA using the pollen-specific RNA and AMV reverse transcriptase in the presence of alpha and random hexanucleotide primers [³²P] -dCTP. The probes are used for hybridization to nitrocellulose plate elevations of plates containing the clones from the genomic library. Clones that can be identified as strong hybridization only to the pollen-specific cDNA and not to the cDNA from any other tissue examined are chosen. These clones are plaques purified and developed for DNA isolation. Alternative techniques for DNA and RNA manipulation as well as for recombinant DNA, development and isolation of clones can be found in normal laboratory manuals, such as Molecular Cloning, A Laboratorv Manual (Maniatis. T., Fritsch, EF, and Sambrook , J., Coid Spring Harbor Laboratory, New York, 1982).

For applications where the Brassica napus L4, L10, L16, or L19 genomic DNA sequence are used to carry out certain aspects of the invention, the preferred method of obtaining a gene that is essential for the formation and / or function of pollen is to synthetically produce a homologous DNA sequence according to normal methodology (see Gait, MJ, Ed., {1984} Oligonucleotide synthesis, a practical approach, pp 1-22,

-5454

IRL Press, Oxford, UK), tag the sequence for the purpose of detection and use the tagged sequence to sort a genomic collection of Brassica napus produced according to the methods described.

The identity of the promoter and coding region of a given genomic clone is determined through the design of restriction maps and hybridization analysis. This can be accomplished by hybridizing cDNA probes made from pollen-specific RNA with restriction fragments of nitrocellulose immobilized DNA clones. Restriction endonuclease fragments containing both the coding region and the DNA regions on both sides of the coding region are isolated by subcloning into appropriate vectors. Once isolated, techniques such as S1 mapping and DNA sequencing are conveniently used to obtain exact coding regions and restriction sites within the subcloned DNA. This analysis is easily accomplished once the polarity with respect to gene transcription is known.

In order to determine the transcription polarity of a gene, individual restriction fragments can be subcloned into commercially available vectors, such as pGEM3, pGEM4 or pGEM3Z, pGEM4Z (available from Promega Biotech, Madison, Wisconsin, USA). Using these vectors, it is possible to generate RNA probes with single chain structures, which are complementary to one or the other structures of

-5555 DNA duplex strand in a given subclone. These specific chain structure probes are hybridized to mRNA, in order to establish transcription polarity. Among these probes, those probes that hybridize to, and therefore are complementary to, mRNA can be isolated. Using this information, it is possible to clearly determine from which DNA strand structure of the double stranded strand genomic DNA molecule the mRNA has been transcribed.

In order to delineate and isolate the promoter DNA sequences of a given genomic clone, series of vector pGEMs can be used for one-way deletion of sequences from the individual subclones in hybridization-protection experiments. Detailed descriptions of these experimental procedures can be found in a number of laboratory manuals and in the manufacturers' technical notes, supplied with the pGEM series of vectors. These experiments will clearly establish the promoter and coding regions of the pollen-specific genomic clones.

The sequence of individual deletions in pGEM vectors can be determined through dideoxy sequencing of plasmid minipreparations as described in the manufacturer's technical notes. Elimination of subclones that are deleted very close to the start of transcription or specific restriction fragments spanning the promoter region or promoter region and the start of transcription are chosen for

-5656 construction of genes that are expressed only in the development of pollen-bearing plant microspores. Usually, the promoter fragment is inserted upstream of a terminator, such as the terminator found in pRAJ-221 (available from Clonetech Laboratories, Palo Alto, CA) and specific restriction fragments, which are to be transcribed into antisense RNA. and are inserted between the promoter and the terminator sequences. The entire construct is verified by combining sequencing and restriction digests. The thus constructed and verified antisense gene can be inserted into T-DNA-based vectors for transformation of plant cells. T-DNA vectors containing a selectable marker are preferred. It should be understood that the antisense gene can be constructed in a variety of ways depending on the selection of vectors, restriction enzymes, and individual genes used. For example, it may be desirable to insert restriction fragments that are intended to be transcribed into antisense RNA into an ATN-T-based vector to which a promoter and terminator structure have previously been added. Alternatively, it is possible to insert a promoter fragment upstream of a coding and terminator region that has been previously added to a T-DNA based vector. Furthermore, it may be desirable in some cultures not to insert the antisense gene into a T-DNA-based vector, but rather into a vector suitable for direct DNA consumption. They can be used

-5757 promoters other than promoters essential for pollen development and / or function, and bind to gene-specific restriction fragments and terminators providing that these promoters function in cells and / or tissues essential to pollen formation and / or function .

In accordance with a second embodiment of the invention, a method is provided for the production of a male sterility plant, transforming a plant with a recombinant DNA molecule comprising a promoter essential for the development and / or function of pollen as described above, and a DNA sequence encoding a gene product that renders a cell / tissue essential to pollen formation and / or function susceptible to a chemical agent or physiological strain. Such a recombinant DNA molecule can be introduced into a plant where all the cells of the plant in which the recombinant DNA molecule is not expressed are resistant to the chemical agent and / or to physiological stress.

In accordance with a third embodiment of the invention, a method for the production of a male sterility plant is provided, transforming a plant with a recombinant DNA molecule comprising a promoter essential for the development and / or function of pollen as written above, and a DNA sequence encoding a protein or polypeptide which is cytotoxic to a cell which is essential to pollen formation and / or function. By substance, which is cytotoxic to a cell, you want

-5858 imply a substance that when the asset disrupts the normal function of a cell / plant tissue preferably leads to cell death. Any cytotoxic substance that is known to be encoded by one or more identifiable DNA sequences can be used within the scope of this embodiment of the invention, including, but not limited to, ricin, abrin and diphtheria toxin. In this way, the DNA sequence can encode any substance which is cytotoxic to a cell that is essential to the formation and / or function of pollen, including genes that encode ricin, abrin, and diphtheria toxin.

In addition to DNA sequences encoding cellular toxins such as ricin, abrin, and diphtheria toxin, the recombinant DNA molecule of the present invention can encode other cytotoxic substances including degradation or destructive enzymes such as ribonuclease, dsDNA, ribozymes, lipase. or protease, substances that interrupt or destabilize cytoplasmic integrity such as polylysine or polyproline, substances that interrupt or destabilize proteins or polypeptides that are essential to certain biosynthetic pathways in plant cells or interfere with the expression of such proteins or polypeptides.

According to a fourth embodiment of the invention, the recombinant DNA molecule can comprise one or more DNA sequences, which encode a gene product which makes a non-toxic substance toxic to a cell / tissue that is essential. to the formation and / or function of pollen. In particular,

-5959 Any identifiable DNA sequence encoding gene products which can be converted to a cytotoxic non-toxic substance can be employed within the scope of the invention. Examples of such DNA sequences include the following:

a) DNA sequences encoding indole acetamide hydrolase (lamH), which converts naphthalenacetamide to plant growth regulating alpha naphthalenacetic acid (NAA), which is toxic to pollen grain development, or converts indolacetamide to acid Indoleacetic (IAA), which is a regulator of plant growth. A source of the laMH enzyme is the bacterium Aqrobacterium tumefaciens (Inze. D., et al., 1984, Mol. Gen. Genet. 194: 265-74).

b) DNA sequences encoding the enzyme methoxynine dehydrogenase (MDH) which converts a non-toxic 2-amino4-methoxy-butanoic acid (methoxynine) to toxic methoxyvinylglycine. One source of MDH is Pseudomonas aeruqinosa bacteria (Margraff, R., et al., 1980, Experimenta 36: 486).

c) DNA sequences encoding the enzyme rhizobitoxin synthase which converts 2-amino-4-methoxybutanoic acid to rhizobitoxin (2-amino-4- [2-amino-3-hydroxypropyl] -trans-3-butane! acid) . A source of the enzyme is Rhizobium iaponicum bacteria (Owens, LD, et al., 1973, Weed Science 21: 63-66).

d) DNA sequences that encode the specific betaglucuronidase enzyme in the tissue, which, for example, converts

-6060 a non-toxic glucuronic acid analog conjugated to a toxic substance such as chloramphenicol (Gluc-Camp) or glyphosate (N [phosphomethyl] glycine) to cytotoxic substances such as chloramphenicol and glyphosate, respectively. Other toxic substances can be conjugated to glucuronic acid to provide a non-toxic analog of glucuronic acid.

When the non-toxic substance does not naturally exist in the plant cell into which the recombinant DNA molecule of the fourth embodiment of the invention is to be introduced, the non-toxic substance can be introduced into the plant cell by spraying, irrigation or other similar means. The non-toxic substance can be produced in the cell by providing the recombinant DNA molecule, which has a DNA sequence that encodes a gene product which makes the non-toxic substance cytotoxic to the cell, or in a second molecule of Recombinant DNA, a second DNA sequence that encodes the non-toxic substance. The non-toxic substance can also be introduced into the cell by providing in the recombinant DNA molecule that it has a DNA sequence that encodes a gene product that makes the non-toxic substance toxic to the cell, or in a second DNA molecule. Recombinant, a second DNA sequence that encodes a gene product that converts a substance, which is endogenous to the cell, to the non-toxic substance. For example, a cell may contain a recombinant DNA molecule that has a DNA sequence that

-6161 encodes lamH (which converts indolacetamide to cytotoxic levels of indolacetic acid), and a recombinant DNA molecule that has a DNA sequence that encodes lamS. lamS converts tryptophan, which is generally endogenous to plant cells, to indolacetamide, which in turn is converted through lamH to cytotoxic levels of indolacetic acid. In the above examples, where the introduction of more than one DNA sequence may be desired, the sequences may be introduced into separate recombinant DNA molecules, or, in a preferred case, may be linked together into one molecule.

The recombinant DNA molecule that comprises one or more DNA sequences which encode a gene product which makes a non-toxic substance cytotoxic to a cell / tissue that is essential to the formation and / or function of pollen, additionally contains one or more promoters to regulate the expression of the DNA sequences. If there is more than one DNA sequence in the recombinant DNA molecule, the expression of the DNA sequences can be regulated through an identical promoter or the expression of each DNA sequence can be regulated through a different promoter. Preferably, the expression of the DNA sequences is regulated through a promoter essential for the development and / or function of pollen, as described above, so that the gene product, which makes a non-toxic substance cytotoxic selectively interferes with the function and / or development of essential cells at

-6262 pollen formation and / or function. When the non-toxic substance is introduced into the plant cell providing a second recombinant DNA molecule that has a second DNA sequence that encodes a second gene product, which converts a substance that is endogenous to the cell to a non-toxic substance. Toxic, the expression of the DNA sequence and the second DNA sequence can be regulated, either through a promoter essential for the development and / or function of pollen, a promoter capable of being induced or a constitutive promoter as long as there is selective interference with the function and / or development of cells essential to the formation and / or function of pollen. Preferably, the expression of the DNA sequence encoding a gene product that makes a non-toxic substance cytotoxic, is regulated through a promoter essential for the development and / or function of pollen, and the expression of the second DNA sequence encoding a gene product that converts a substance, which is endogenous to the cell to a non-toxic substance, is regulated through a constitutive promoter, a promoter capable of being induced or a promoter essential for the development and / or function of pollen, although very preferably through a promoter essential for the development and / or function of pollen. For a more detailed discussion of the aforementioned promoters, reference is made to the discussion of promoters with reference to recombinant DNA molecules containing antisense genes. A preferred promoter could be the microspore-specific Bp10 promoter (Albani, D., Sardana, R.,

-6363

Robert, LS, Altosaar, I., Arnison, PG, and Fabijanski, SF,) a Brassica napus gene family that shows sequence similarity to ascorbate oxidase is expressed in developing pollen. Molecular characterization and analysis of promoter activity in transgenic tobacco plants. Plant J. 2: 331-342 (1992) or the anther specific TA39 promoter described above.

In accordance with a fifth embodiment of the invention, the recombinant DNA molecule comprises one or more DNA sequences, which encode a gene product which can be converted to a substance that is cell / tissue cytotoxic essential for formation and / or pollen function.

The recombinant DNA molecule of the second, third, fourth, and fifth embodiments of the invention also preferably further contains one or more selection marker genes as described above and termination sequences and other transcriptional regulatory signals as described above. The preferred selection marker could be a glufosinate or glyphosate or TBL herbicide resistant gene, described above.

The following description sets forth in general terms, the steps that can be employed to produce plants, which are male sterile and carry the male sterile trait, using the recombinant DNA molecules of the invention. It should be understood that these various

-6464.

Stages can be accomplished through a variety of different procedures. In the following description of procedures, we describe ways to accomplish these steps. Preferred procedures can be detailed below and in the examples. However, it is contemplated that other variations will be apparent to those skilled in the art.

The recombinant DNA molecules of the invention can be used to produce a plant, which is male sterile or carries a male sterility trait, according to the following preferred steps:

(1) Introduce one or more recombinant DNA molecules, as contemplated herein, to carry out the invention, to the genome of one or more plant cells, of a plant selected from those pollen-producing plant species, which they are capable of being genetically transformed, preferably cells from which complete differences can be conveniently recovered;

(2) Selecting a plant cell in which the recombinant DNA molecule is stably incorporated; and (3) Regenerate from the selected plant cell a plant which is male sterile or carries the male sterile trait.

A male sterility plant can preferably be produced through the stages before

-6565 mentioned, using the following recombinant DNA molecules of the invention in step (1):

a) A recombinant dna molecule that comprises a DNA sequence that encodes an RNA, which, when produced in a plant cell that is essential to the formation and / or function of pollen, interferes with a gene which is essential to the formation and / or function of pollen, and an essential promoter for the development and / or function of pollen, or a constitutive promoter;

b) A recombinant DNA molecule that comprises a DNA sequence, which encodes an RNA, which when produced in a plant cell that is essential to the formation and / or function of pollen, interferes with a gene which it is essential to the development and / or continuous function of the metabolically competent cells of a plant and an essential promoter for the development and / or function of pollen; and

c) A recombinant DNA molecule comprising the DNA sequence encoding a protein or polypeptide, which when produced in a plant cell that is essential to pollen formation and / or function, is substantially cytotoxic to the cell or makes a non-toxic substance toxic to cells, where the non-toxic substance is introduced into the cell by converting an endogenous substance to the cell to the non-toxic substance, and an essential promoter for the development and / or function of pollen.

-6666

A plant bearing a male sterility trait can preferably be produced through the aforementioned steps, using the following recombinant DNA sequences of the invention in step (1):

(a) A recombinant DNA molecule that comprises a DNA sequence that encodes an RNA which, when produced in a plant cell that is essential to the formation and / or function of pollen, interferes with a gene which is essential to the formation and / or function of pollen and a promoter capable of being induced;

(b) A recombinant DNA molecule that comprises a DNA sequence encoding a protein or polypeptide that causes a plant cell, which is nominally resistant to a chemical agent or physiological strain, sensitive to such a chemical agent or physiological strain and a specific pollen promoter; and (c) A recombinant DNA molecule that comprises a DNA sequence that encodes a protein or polypeptide that when produced in a plant cell that is essential to the formation and / or function of pollen, causes a non-toxic substance. is substantially cytotoxic to the cell, where the non-toxic substance is externally applied to the cell, or where the non-toxic substance is introduced into the cell by converting an endogenous substance to the cell from the non-toxic substance and a promoter capable of being induced controls the expression of the protein and polypeptide, which converts the endogenous substance to the non-toxic substance. toxic.

-6767

A plant bearing a male sterility trait can be made male sterile by exposing the plant to a sterile acting agent, which, for example, is an inducer in the case of (a) above, a chemical agent or physiological strain in the case of (b) above, or a non-toxic substance, which is capable of being made cytotoxic (i.e., a cryptocytotoxic substance) to a cell that is essential to pollen formation and / or function in the case of (c ) above.

A recombinant DNA molecule can be introduced into a plant cell by any of a variety of known methods, preferably by first inserting the recombinant DNA molecule into a suitable vector and then using the vector to introduce the recombinant DNA molecule into a cell. of plant. As described above, a recombinant DNA molecule may also contain a selection marker gene, which encodes a selection gene product, which confers on a plant cell resistance to a chemical agent or physiological strain, or confers a Phenotypic characteristic distinguishable from cells, so that plant cells transformed with the recombinant DNA molecule can be easily selected using a selective agent. The thus selected transformed plant cells can be induced to differentiate into plant structures, which will eventually produce the entire plants. The selection agent can preferably be an herbicide such as glufosinate, glyphosate or TBL and the marker of

-6868 selection can be any gene that confers herbicide resistance to the plant cell. This approach has the added benefit that the resulting plant can also be resistant to herbicide.

The use of cauliflower mosaic virus (CaMV) (Howell, SH, et al., 1980, Science 208: 1265) and geminal viruses (Goodman, RM, 1981, J. Gen. Virol. 54: 9) as vectors, has been suggested but the greatest successes reported have been with Agrobacteria sp. (Horsch, RB, et al., 1985, Science 227: 1229-1231). Methods for using Agrobacterium-based transformation systems have now been described for many different species. Generally, strains of bacteria harboring modified versions of the naturally occurring Ti plasmid are used so that DNA is transferred to the host plant without subsequent tumor formation. These methods involve insertion into the Ti plasmid limiters, the DNA will be inserted into the plant genome linked to a selection marker gene, to facilitate the selection of transformed cells. Bacteria and plant tissues are grown together to allow transfer of foreign DNA to plant cells then transformed into plants that are regenerated in the selection media. Any number of different organs and tissues can serve as targets for Agrobacterium-mediated transformation as specifically described for members of Brassicaceae. These include thin cell layers (Charest, PJ, et al., 1988, Theor. Appl. Genet. 75: 438-444), hypocotyls (DeBlock, M., et al., 1989, Plant

-6969

Physiol, 91: 694-701), leaf discs (Feldman, KA, and Marks, MD, 1986, Plant Sci. 47: 63-69), stems (Fry J., et al., Plant Cell Repts. 6: 321-325 (1987), cotyledons (Moloney MM, et al., Plant Cell Repts. 8: 238-242 (1989) and embroids (Neuhaus, G., et al., Thoer. Appl. Genet. 75: 30- 36 (1987) However, it is understood that it may be desirable in some cultures to select a different tissue or transformation method.

It is also understood that successful transformation and recovery of a plant containing these recombinant sequences does not always result in appropriate pollen-specific expression. The transformation procedure results in the random insertion of the foreign DNA, so that position effects can occur and suppress the activity of any of the introduced DNA. Thus, it is advisable to generate a number of individual transformed plants with any recombinant construct in order to recover individuals free from any of the limiting position effects. It is also preferred to select plants that contain more than one copy of the introduced recombinant DNA molecule, so that high levels of expression of the recombinant molecule are obtained.

It is known that the number of plant species that have been successfully and genetically transformed continues to represent a modest percentage of a total number of plant species that are of potential commercial interest. It is true, however, that

-7070 the number of species that have been transformed has steadily increased and there is reason to expect that transformation systems can be developed for any ongoing crop of interest. Routine transformation was initially accomplished with species from two plant families: Solanaceae and Brassicaceae. Examples of species of commercial interest from these families that have been transformed include: tobacco, Nicotiana tabacum L. tomato, Lvcopersicon esculentum Mili, potato, Solanum tuberosum L., and petunia, Petunia hybrida (Solanaceae): Canola / Colza, Brassica napus L., cabbage, broccoli, cabbage, etc., Brassica olerácea L., mustards, Brassica júncea L .. Brassica niqra L., and Sinapis alba L. (Brassicaceae).

Recently, transformation has been reported from commercially important species from other families such as beet Beta vulqaris, (Chenopodiaceae). cucumber, Curcubita sp. (Curcurbitaceae), cotton, Gossypium sp., (Malvaceae), sunflower (Helianthus annuus and lettuce Lactuca sativa, (Asteraceae Compositae). And pea, Pisum sativum, soybean, Glvcine max v alfalfa, Medicaqo sp (Fabaceae = Lequminoseae). The transformation it has also been achieved with three species such as poplar, Populus sp. (Salicaceae) and walnut, Juqlans niqra, (Juglandaceae).

Transformation with monocotyledonous species has not progressed as rapidly, as these species are generally not very susceptible to Agrobacterium-mediated transformation. However, the progress that has been remarkable includes

-7171 asparagus, Asparagus officinalis; gladiolas, Gladiolus sp., (Lilaceae): corn, Zea mays and rice, Oryza sativa (Poaceae). The recent discovery that transformation with Agrobacterium can be achieved by infecting germination seeds without the requirement of cell culture generation (Chee, PP, et al., Plant Physiol, 91: 1212-1218 (1989) opens new horizons for species they can be difficult to regenerate. Furthermore, widely scattered studies on the use of particle guns to transfer DNA-coated microprojectiles to plant cells of species that are not readily susceptible to other methods remain highly compromised. It is expected that the present invention can be carried out with any of the previous species and with any other species that is capable of being genetically transformed.

The most widely used and generally successful methods for introducing foreign DNA into plants rely on the use of an infectious agent, such as the Agrobacterium tumefaciens Ti plasmid, as a vector for delivery of the foreign DNA as described above. However, it may be possible to use other methods as well. Other methods that have been developed involve mechanical means such as direct consumption of DNA, liposomes, electroporation (Guerche, P. et al., 1987, Plant Science 52: 111-16) and microinjection (Neuhaus, G., et al., 1987, Theor. Appl. Genet. 75: 30-36). Recently, the possibility of using microprojectiles and a pistol or other

-7272 devices for forcing DNA-coated small metal particles into cells has received considerable attention (Klein, TM et al., 1987, Nature 327: 70-73). To date, success with these and other mechanical methods has not been widely reported. The method of choice will depend on the particular plant species and will be apparent to those skilled in the art. In those plant species, where a successful transformation has not yet been demonstrated, it is anticipated that any new transformation methods developed can be used to insert the recombinant DNA molecules of the invention and the current method of inserting the genes will have very little or no effect on the functioning of the crop production systems described above.

It is also possible to produce plants, which have male sterility or carry the male sterility trait by fusing cells of a plant cell line containing cells having one or more recombinant DNA molecules, of the invention with cells of non-plant species. they can be transformed through normal methods. A fusion plant cell line is obtained that carries a genetic component for both plant cells. Fused cells carrying the recombinant DNA molecule can be selected and in many cases regenerated to plants that have male sterility or that carry the male sterility trait.

-7373

It is contemplated that some embodiments of the present invention may require that a plant cell be transformed with a recombinant DNA molecule containing at least two DNA sequences or that it be transformed with more than one recombinant DNA molecule. The DNA sequences or recombinant DNA molecules in such modalities can be physically linked, being in the same vector, or physically separated in different vectors. A cell can be simultaneously transformed with more than one vector as long as each vector has a unique selection marker gene. Alternatively, a cell can be transformed with more than one vector sequentially allowing an intermediate regeneration step after transformation with the first vector. Furthermore, it may be possible to perform a sexual cross between individual plants or plant lines containing different DNA sequences or recombinant DNA molecules. Preferably, the DNA sequences or the recombinant molecules are linked to or located on the same chromosome. It may then be possible to select from the progeny of the cross, plants containing both DNA sequences and recombinant DNA molecules.

When at least two recombinant DNA molecules are necessary to interfere with the function and / or development of a plant cell that is essential to the formation and / or function of pollen (for example, a first recombinant DNA molecule 25 contains a first DNA sequence that encodes a first

-7474 gene product, which converts a non-toxic substance to a cytotoxic substance and the non-toxic substance is introduced into the cell through a second recombinant DNA molecule, which contains a second DNA sequence that encodes a second product gene which converts an endogenous substance to a plant cell to the non-toxic substance), Recombinant DNA molecules can be linked to avoid segregation of DNA sequences required to produce the desired effect. Alternatively, the male sterility plant can be produced by crossing two plant lines, each containing one of the recombinant DNA molecules, where the plant lines are otherwise isogenic. Each plant line is preferably made homozygous for the respective recombinant DNA molecules to ensure that all progeny receive a copy of each of the recombinant DNA molecules. Furthermore, in some hybridization schemes discussed below, it is preferable to have the respective recombinant DNA molecules located on the same chromosome pair in each line. Selection of a pair of chromosomes containing a first recombinant DNA molecule in one plant line can be predetermined, randomly, and the other plant line can be adjusted so that a second recombinant DNA molecule is located in the same pair. of chromosome. For example, to produce a plant line, which is homozygous for a first recombinant DNA molecule, a pair of chromosomes, in which

-7575 locates a second recombinant DNA molecule, is identified in a plant line, a transformed cell in which the first recombinant DNA molecule is incorporated into the identified chromosome pair, is then selected, a plant is regenerated from the transformed cell, the plant is purified, and a plant is selected, which is homozygous for the trait encoded by the first recombinant DNA molecule, and the number of homozygous plants is increased, purifying by isolation.

Methods for identifying the chromosome pair on which a recombinant DNA molecule is located and methods for producing homozygotes are discussed in detail below.

As noted above, it may be desirable to produce plant lines, which are homozygous for a particular gene. In some species, this is rather easily accomplished through the use of anther culture or isolated microspore culture. This is especially true for the cultivation of oilseed Brassica napus (Keller and Armstrong, Z. Pflanzenzucht 80: 100-108, 1978). Using these techniques, it is possible to produce a haploid line that carries the inserted gene and then doubles the chromosome number either simultaneously or through the use of colcycin. This gives rise to a plant that is homozygous for the inserted gene, which can be easily analyzed for whether it carries a gene inserted into a selection marker gene suitable for detection of plants carrying that gene. Alternatively, the

-7676 plants can be self-fertilized, leading to the production of a seed mix consisting of, in the simplest case, three types, homozygous (25%), heterozygous (50%), and null (25%) for the inserted gene . Although it is relatively easy to classify null plants 5 from those containing the gene, it is possible to practice homozygous classification from heterozygous plants through southern staining analysis where much staining is put on the load of exactly equivalent amounts of DNA from the mixed population, and classify heterozygotes through the signal intensity of a probe specific for the inserted gene. It is advisable to verify the results of the southern staining analysis allowing each independent transformant to self-fertilize, since additional evidence for homozygosity can be obtained through the simple fact that if plant 15 was homozygous for the inserted gene, all plants Subsequent from the self-fertilized seed will contain the gene, whereas if the plant was heterozygous for the gene, the developed generation of the purified seed will contain null plants. Therefore, with simple purification, 20 homozygous plant lines can be easily selected which can also be confirmed through southern staining analysis.

Two techniques can be used to produce plant lines, which carry genes that secrete in a similar way, or are on the same chromosome or a group of chromosome pairs.

You can use the simple crossover strategy in two

-7777 transformants that are homozygous for a single inserted gene are crossed to produce the F1 seed. The F1 seed progeny plants (F1 plant generation) can be crossed with a recipient plant and the segregation of the two inserted genes is determined (F2 plant generation). For example, when the lamH and lamS genes are the inserted genes, the F1 plants grown from the F1 seed will be male sterile. If the original transformants are homozygous for an individual inserted gene, when they are crossed with an untransformed plant to produce the F2 seed, the F2 plants will be 100% male sterile if the two transformants originally used for the production of the F1 seed carried the lamH and lamS genes on the same chromosome or in the same linker group. If the genes are in separate linking groups or on different chromosomes, a variable degree of male sterility will be seen, in theory 25% of plants will be male sterile if the genes secrete completely and independently of each other. This approach allows selection of production lines for homozygous transformed plant lines containing the lamS and lamH genes, which will secrete substantially 100% in the hybrid seed sold for commercial use.

An alternative strategy may make use of the extensive genetic maps available for many commercially developing crops and the many easily classifiable markers that are known to most linking groups or

-7878 chromosomes. In some cases, the link groups and chromosomes can be equivalent, while in others, there is more than one link group assigned to each chromosome. When there is a marker for each chromosome, identifying the chromosome into which the recombinant gene has been inserted is relatively simple. A cross is made between each of the individual transformant and a container plant that allows the marker to be viewed.

If there are markers that can be classified that have been located on each of the chromosomes in the plant, and the markers can be classified in the generation produced by this crossing, the segregation of the inserted gene can be located with the marker, thus establishing the chromosomal location of that gene. Therefore, this allows the chromosome and more importantly the linking group with which the inserted gene is secreted. Many crops, such as corn, tomato, and various cereal crops have extensive genetic maps that allow identification of the chromosome that contains the inserted gene. It is contemplated that more detailed chromosome maps are made, especially with the use of RFLP (restriction fragment length polymorphism) maps, the determination of gene inserted to particular chromosomes, will be easily performed for the majority of commercial crop species.

As a means of confirmation, or in plant species where chromosomal markers are not known, it is possible

-7979 use a technique called pulse-field electrophoresis (originally described by Schwartz and Cantor, Cell, 37: p67; 1984) to determine whether different transformed plants contain genes inserted on the same chromosome. Pulse-field electrophoresis is a technique that can separate large pieces of DNA, even chromosomal size, into a pattern of reproduction on a gel. When this is done, it is possible to process this gel so that the chromosome points can be analyzed through southern staining techniques, locating the inserted gene on the chromosome point. When the entire population of primary transformants is analyzed in this way, it is a simple task to select the two transformants that carry the inserted genes at the same point on the chromosome.

As mentioned above, after the recombinant DNA molecules are introduced into the genome of a plant cell, a plant cell that has a recombinant DNA molecule stably integrated into its genome is selected. This selection step can be facilitated by incorporating a selection marker gene into the recombinant DNA molecule, so that plant cells transformed with the recombinant DNA molecule can be easily selected using a selective agent as described above. A selection marker gene is neomycin phosphorotransferase (NPT II), which confers resistance to kanamycin and the antibiotic G-418. Cells transformed with this gene

-8080 selection marker can be selected by in vitro kanamycin treatment by phosphorylation using techniques described in the literature or by testing for the presence of the mRNA encoding the NPT II gene through Northern staining analysis in the RNA through the tissue of the transformed plant. A preferred selective agent and selective marker could be a herbicide and its herbicide resistance gene, as described above.

The expression of the recombinant DNA molecules of the invention in transformed plant cells can be verified using Northern staining techniques. Single stranded structure RNA probes, which are homologous to transcripts of the coding sequences of a recombinant DNA molecule, can be used to detect the presence of the recombinant DNA molecule in a plant or tissue cell, of so that the expression of the coding sequence can be found out. It is preferred to use agarose gel electrophoresis to separate transcripts according to size under denaturation conditions. In the case where the expression of the specific gene in the pollen of the coding sequence is sought to achieve this, it is advisable to test the expression of the coding sequence in the cell type, whose function and / or development will be interfered with. with, and in other tissues, such as leaves, roots, etc., so that gene expression

Tissue-specific -8181 recombinant DNA molecule in such cell types can be verified.

The presence of the stably integrated recombinant DNA molecule in the genome of the plant cell can also be ascertained using Southern staining techniques. In this procedure, total cellular or nuclear DNA is isolated from the transformed plant or plant cell, and is preferably digested with a restriction enzyme, thereby giving rise to discrete fragments. These discrete fragments can be detected in the nuclear or total DNA of the transformed plant or plant cells, using normal gel electrophoresis, followed by nucleic acid hybridization techniques.

The formation of microspores in plants, which contain the recombinant DNA molecules of the invention and which are male sterile, is first verified by visual microscopic examination of the anther structure. As flower maturation occurs, anther formation is expected to be represented or completely inhibited so that no degree of mature pollen is formed or released.

It will be appreciated that plants produced using the method of producing male sterility plants of the present invention can exhibit varying degrees of male sterility. This may be the result of the nature of the transformed plant cell or the opportunity to place the recombinant DNA molecule in the genome of the plant cell.

-8282

The following description of the invention sets forth in general terms the steps that can be employed to increase the number of male sterile plants and plants bearing the male sterile trait and to produce restorative plants, hybrid seeds, hybrid fertile seeds restored, and hybrid seeds that produce a mix of male fertility and male sterility plants. It is understood that these various types can be accomplished through a variety of different procedures. In the following description of the 10 procedures, we describe ways to achieve these steps. Preferred procedures can be detailed later and in the examples. However, it is contemplated that other variations will be apparent to those skilled in the art.

As mentioned above, the invention relates to hybrid seeds having a genome comprising one or more of the recombinant DNA molecules of the invention to produce plants, which are male sterile or carry the male sterility trait, and to seeds of plants developed from hybrid seed. The invention is also directed to a mixture of fertile and sterile hybrid seeds and to plant seeds developed from such hybrid seed. The invention is also directed to hybrid seeds having a genome comprising one or more of the recombinant DNA molecules of the invention, to produce a male sterility plant and corresponding restorer gene product to restore

-8383 fertility in the seed and in plant seeds developed from the hybrid seed.

To produce a hybrid seed on a commercial scale from a plant, which has male sterility or carries the male sterility trait, the number of such plants must be increased or maintained and crossed with a suitable male fertility source line.

A plant that carries the male sterility trait can be more easily maintained since such a plant will self-pollinate rather than be treated with a sterility driving agent that makes the plant substantially sterile. For example, in a plant that contains a recombinant DNA molecule that has a DNA sequence that encodes a gene product that makes a cell susceptible to a chemical agent or physiological strain and that has a promoter essential for development and / or or function of pollen, the plant will be made substantially of male sterility, treating the plant with a driving agent, mainly, the chemical agent or physiological stress. Accordingly, plants, which carry the male sterility plant, are preferably maintained by purification, selecting from the purification progeny a plant that is homozygous for the male sterility trait, and increased the number of plants that are homozygous. for the male sterility trait by purifying in isolation a number of generations. Selecting a plant

-8484 which is homozygous for the male sterility trait, can also be performed through the procedures discussed above.

A plant, which is substantially male-sterile, is preferably maintained by crossing the male-sterile plant with a suitable male-fertility plant, obtaining a seed from the plants, resulting in the form of the crossing, developing seed plants, and selecting the plants that are male sterile among plants developed from seeds. This procedure can be repeated a number of generations until the desired number of male sterility plants is obtained.

A plant that carries the male sterility trait can also be maintained through the latter method.

To facilitate selection of male sterility plants in maintained plants the recombinant DNA molecules used to produce the male sterility plant should preferably comprise a selection marker gene, preferably the selection marker gene is linked to a DNA sequence contained in the recombinant DNA molecule as described above. Such a selection marker may be a herbicide resistance gene.

A male fertility plant suitable for the purpose of crossbreeding with a male sterility plant to increase the number of male sterility may be, but is not

-8585 is limited to, a plant from the same breeding line from which the male sterility plant is derived. In some cases, referred to below, maintenance of the male sterility line can be produced simply by purifying in isolation.

According to another scheme, the number of male sterility plants can be increased through clonal propagation using their tissue explants, or other in vitro propagation techniques.

When the cost is guaranteed, and maintenance cannot be easily accomplished as discussed above, transformed plant cells can be grown into cultures according to routine methodology to produce a cell line. A cell line can be regenerated according to routine methodology to increase and maintain the male sterility cell line. Routine methods for growing cell lines and regenerating transformed plants from cell lines are described in normal plant and tissue culture books. (Plant Tissue and Cell Culture, Green, CE, Somers, DA, Hackett WP, and Biesboer, DD Eds, 1987, alan R. Liss, Inc., New York, Experiments in Plant Tissue Culture, Dodds. JH and Roberts, LW Eds, 1985, Cambridge University Press, or Cell Structure and Somatic Cell Genetics of Plants, Vasil. IK Scowcroft, WR, and Frey KJ, Eds., 1984, Academic Press, New York, Handbook of Plant Cell Culture, Vol 1-4, Evans, DA, Sharp, WR, Ammirato,

-8686

PV, and Yamada, Y. Eds. 1984-1986, Macmillan, New York, Biotechnoloav in Aariculture adn Forestry, VoI 1 and 2, Baja], YPS Ed., 1986, Springer-Verlag, Berlin, or Plant Propaqation bv Tissue Culture - Handbook and Diretory of Commercial Laboratories, George EF, and Sherrington, PD, 1984, Eastern Press, Reading).

Hybrid seed production can be accomplished by pollinating male sterile plants with pollen derived from selected male fertility plants. Pollination can be by any means, including, but not limited to, manual, wind, or insect pollination, or mechanical contact between the male fertility and male sterility plant. For hybrid seed production on a commercial scale in most plant species, wind or insect pollination is preferred. Plant selection for pollen donation is determined through normal crossing of different plants with subsequent progeny analysis and selection of lines with the best combining capacity and superior agronomic traits. Restoration of fertility in hybrids can be accomplished using the methodology detailed below.

For certain crops of interest, such as vegetables, this may only be the leaves, stems, or roots of the plant that are to be commercially sold. Therefore, although the recombinant DNA molecules, which cause the male sterility of the plant to be inherited and expressed in the plant

-8787 hybrid, it is not necessary to overcome or restore male fertility in the seed of the hybrid plant. However, for other crops, the commodity of trade may be the seed or fruit produced by the hybrid plant. Thus, for optimal commercial utility of the hybrid, it may be desirable to produce a hybrid seed that is fertile.

The invention contemplates a variety of recombinant DNA molecules that can be used to produce a male sterility plant. In a scheme where any recombinant DNA molecule consists of a single gene or linked genes, the gene can be segregated as a unit to produce a mixture of a sterile hybrid seed and a fertile hybrid seed. Similarly, unbound recombinant molecules can segregate to produce a mixture of fertile and sterile seed except that, as discussed below, where the molecules are located on different chromosomes of the same chromosome pair. In crossover species, plants that develop from the fertile seed present may allow full pollination of hybrid male sterilization plants. Therefore, it may be preferable to adopt a scheme to produce hybrid seeds, which are fully fertile. However, in species where they are hybrid plants of male sterility, of relatively weak crossing, they will not be fully pollinated, thus reducing seed production. Therefore, when the comfort of trade is the seed produced

-8888 For the hybrid plant, it may be desirable to produce hybrid seeds, which are fully fertile.

The invention contemplates methods for restoring fertility in hybrid plants produced according to the methods of the invention. Specifically, the invention contemplates a method of restoring hybrid plants produced in accordance with the methods of the invention by incorporating into a male plant, a recombinant restorative DNA molecule containing a restorer gene encoding a restorer gene 10 product, the which compensates for a gene function that has been compromised by a gene product encoded by a first recombinant DNA molecule of the invention, or which negates the effect caused by a gene product encoded by a first recombinant DNA molecule of the invention, as discussed below. The male plant in this way can act as a restorative plant. The restorative plant can be homozygous for the gene encoding the restorative trait to ensure that all progeny inherit a gene. The homozygous restorative plant can be maintained by purification in isolation to produce a restorative line.

Selection of a plant that is homozygous for the restorative trait can be performed, as discussed above, by conducting culture of anther or microspore isolated from the genetically transformed plant bearing the restorative trait, or

-8989 preferably purifying the plant in isolation before selection.

The expression of a restorative gene can be regulated through any promoter that is active during the transcription period of a first DNA molecule, which encodes a gene product, which substantially interferes with function and / or development. from an essential cell to the formation and / or function of pollen. It is preferred that the expression of the restorative gene be regulated through the same promoter used to regulate the expression of the first recombinant DNA molecule or any promoter that is highly active in tissues that are essential to the formation and / or function of pollen.

The methods for restoring fertility in hybrid plants, which have been integrated into the genome of its 15 recombinant DNA molecules from cells of the invention, are discussed below.

The action of a protein or polypeptide, which is substantially cytotoxic to a plant cell, which is essential to the formation and / or function of pollen, and which is encoded through a DNA sequence contained in the Recombinant DNA molecule, which is integrated into the genome of cells of a hybrid plant, can be negated by regulating the expression of the DNA sequence encoding the polypeptide protein. For example, a gene antisense to the gene encoding the protein or polypeptide may be incorporated into a line of male origin.

-9090

In particular, when the recombinant DNA molecule has a DNA sequence that encodes a protein or polypeptide, which is substantially cytotoxic to a plant cell, which is essential to the formation and / or function of pollen, the plant Male sterility can be crossed with a suitable male fertility plant that has previously been transformed with a restorative recombinant DNA molecule. The restorative recombinant DNA molecule may contain a restorative DNA sequence, which is in the antisense orientation to that of the DNA sequence encoding the cytotoxic protein or polypeptide and a promoter that controls the restorative DNA sequence, which activates transcription of the restorative DNA sequence, approximately at the time of transcription of the DNA sequence. The restorative gene product inhibits expression of the DNA sequence encoding the protein or polypeptide in the hybrid plant. The promoter that controls the expression of the restorative DNA sequence is preferably the same promoter essential for the development and / or function of pollen that controls the expression of the cytotoxic protein or polypeptides. Constitutive and inducible promoters can also be used advantageously to control the expression of the restorative DNA sequence.

A cytotoxic method for producing male sterile plants may involve the synthesis of protein or polypeptides capable of substantially interfering with the function and / or development of a cell / tissue that is essential for formation and / or function.

-9191 of pollen. Methods for restoring fertility in male sterile plants produced through cytotoxic methods involve the synthesis, in cells / tissues that are essential to the formation or function of pollen, of restorative gene products that specifically block, neutralize, or destroy proteins. or polypeptides. If the recombinant DNA molecule integrated into the genome of a hybrid plant has a DNA sequence, which encodes a cellular toxin, a restorative plant that contains DNA sequences that encodes a detoxification molecule, can be used to restore fertility. . If the recombinant DNA molecule embedded in the genome of a hybrid plant has a DNA sequence that encodes a destructive enzyme, a restorative plant that contains DNA sequences that encode a specific enzyme inhibitor can be used to restore fertility. If the recombinant DNA molecule integrated into the genome of a hybrid plant has a DNA sequence that encodes a cytoplastic disruption molecule, a restorative plant that contains DNA sequences that encode a specific peptidase can be used to restore fertility. Judicious selection of gene expression regulatory mechanisms for the male sterility-producing agent could allow the use of gene expression mechanisms as restorative gene sequences. An example could be the use of the LexA mechanism (Brenn, R. and Ptashne, MS, Reqlation of Eukaryotic gene expression. Patent # 4,833,080, 1989).

-9292

Specific examples of cytotoxic proteins and polypeptides and their restorative gene products are proteinase enzyme trypsin and soybean or cowpea trypsin inhibitor; ribonuclease and a ribonuclease inhibitor; or a starch-degrading enzyme such as alpha-amylase and an alpha-amylase inhibitor.

Fertility in a hybrid plant that contains a recombinant DNA molecule containing an antisense gene in the genome of its cells can be restored as follows. An antisense gene contained in a recombinant DNA molecule that will be incorporated into the genome of a plant can be selected to contain sequences of a sense gene, which are transcribed, but not transported, in an antisense orientation. The transcribed but not transported sequence may include a 5 'non-transported leader sequence, intervention sequences and a 3' non-transported sequence, or any substantial fragment of these sequences. It is understood that the sequences or their fragments can be of natural existence or strange sequences.

In this way, any hybrid plant produced from a cross with a plant containing such an antisense gene can be restored by crossing with a restorative plant that has been transformed with a modified form of the initially target gene, which does not contain the complementary regions. to the antisense gene and therefore does not undergo antisense regulation.

-9393

As described above, a plant bearing a male sterility trait can be produced by integrating various recombinant DNA molecules, of the invention, into the genome of a plant cell and regenerating the plant from plant cell 5. The development and / or function of cells that are essential for the formation and / or function of pollen in the plant are interfered with only after the plant has been exposed to a sterility driving agent such as an inducer, a toxic agent or a cryptocytotoxic substance. Thus, restoration is inherent in development plants produced from hybrid seeds in the absence of the sterility driving agent.

When more than one recombinant DNA molecule of the invention is used to produce a male sterility plant, the recombinant DNA molecules can be inserted into the same pair of chromosomes in separate isogenic plant lines. The respective lines are preferably homozygous for the respective recombinant DNA / gene molecules before crossing the lines to produce a male sterility line 20. When a first and a second recombinant molecule are integrated to the same chromosome in isogenic plant lines, a crossing of these lines results in the first and second recombinant DNA molecules that are located on separate chromosomes of the same chromosome pair in the plant. 25 male sterility. Consequently, when a plant of

-9494 male sterility is crossed with a suitable male fertility plant from a different line, both chromosomes from the chromosome pair are segregated to a separate F1 progeny with the result that the first and second recombinant DNA molecules are not expressed in the same plant. In this way, the F1 hybrid seed is fully fertile. If the two recombinant DNA molecules are integrated into different chromosomes in the male sterility plant, then a portion of the F1 hybrid seed will be male sterile, since there is a 25% probability of co-segregation of the chromosomes that contain both. Recombinant DNA molecules to the male sterility plant. This latter approach may be advantageous with respect to cross species. When F1 male fertility plants are crossed, a portion of the F2 seed will inherit both chromosomes containing the first and second recombinant DNA molecules, and will consequently be male sterile. When the seed is for convenience of trade, it is advantageous for seed producing companies to use a hybrid seed production scheme, where the safety of F1 hybrid seed is emphasized. Crossing over into F1 hybrid plants results in partial male sterility in F2 generation, thereby reducing the seed production of F2 plants, which is commercially desirable. An example of this method is as follows: a first line of male sterility plant, incorporating in its genome a recombinant DNA molecule that

-9595 has a lamH gene that encodes lamH, which converts non-toxic lamH to toxic levels of IAA, can be crossed with a second plant line that has a genome incorporating a second DNA molecule that has a lamS gene, which converts Tryptophan 5 to IAM.

In a preferred method of the invention for producing a hybrid seed, a first line of male sterility plant that has a genome incorporating a recombinant DNA molecule that has a first DNA sequence that encodes a protein or polypeptide that causes a substance nontoxic is substantially cytotoxic to a plant cell which is essential for pollen formation and / or function and an essential promoter for pollen development and / or function, it is crossed with a second plant line, which contains a second recombinant DNA molecule that has a second DNA sequence that encodes a second gene product that converts a sequence which is endogenous to a plant cell to the non-substance toxic. Preferably, the first and second plant lines used in this method are isogenic and each line carries a homozygous site 20 for the first DNA sequence or the second DNA sequence.

Most preferably, the first and second DNA sequences are located on the same chromosome pair of the plant lines, so that at any junction of the two lines, a single chromosome pair contains both the first and second sequences of DNA. The first line of plant is made of sterility

-9696 male exposing the first plant line to the non-toxic substance. The protein or polypeptide encoded by the recombinant DNA molecule incorporated into the genome of the first plant line will make the non-toxic substance toxic in plant cells, which are essential for the formation and / or function of pollen, producing thus a line of male sterility plant. The male sterility plant line also preferably has a selection marker gene linked to the first DNA sequence encoding the protein or polypeptide, which renders a non-toxic substance cytotoxic to facilitate harvesting of the seeds that have cells. that contain the first and second DNA sequences.

When the first male sterility plant line and the second plant line cross, the first male sterility plant line produces seeds that have cells that contain the DNA sequence encoding the non-toxic substance (eg AMI) and the DNA sequence that encodes the protein or polypeptide (lamH) which makes the non-toxic substance toxic (for example, lamH converts AMI to a toxic level of IAA). Seed that has cells that contain the first and second DNA sequences will produce male sterility plants, which can be pollinated with the male fertility line to produce the commercial hybrid seed. If the first and second DNA sequences are located on the same chromosome or in the same linking group, the DNA sequences

-9797 will fully segregate into F1 hybrid seed and the hybrid seed will be substantially male fertile.

Advantage is taken in the aforementioned preferred method of the fact that most plant species produce, per plant, many hundreds of seeds. In Brassica oilseed, for example, a plant, under normal conditions, can produce thousands of seeds. Using the method described above, a thousand-fold increase per seed can be expected per unit area sprayed with the non-toxic substance. That is, for example, when two isogenic lines are produced that carry the lamS and lamH genes, the first stage of pre-production involves the use of NAM to cause male sterility in the plant line that carries only the lamH gene. When the pollinated cross with the pollen of the plant that contains the lamS gene, you can expect hundreds of seeds per unit area, each seed capable of growing in a male sterile plant. When these seeds are planted and crossed with a male fertility plant, thousands of seeds can be expected per unit area. Therefore, if one acre of the plant line carrying the lamH gene and the pollinator carrying the lamS gene were to be planted, this acre might need to be sprayed with NAM. However, from this acre, enough seeds could be obtained to develop 1,000 acres of male sterility plants and pollinators, and from these 1,000 acres, enough hybrid seeds with restored fertility could be obtained to plant

-9898 one 1,000,000 acres of hybrid cultivation. The amount of handling required to produce this hybrid seed is reduced compared to conventional methods due to the pre-production amplification step employed. If the lamH gene is linked to a herbicide resistance gene, fields can be planted randomly to ensure high cross-pollination regimes and the herbicide can be used to annihilate pollination plants after flowering. This method, therefore, allows sufficient hybrid seed production over methods where the hybrid seed is harvested directly after the first cross pollination.

A particularly preferred embodiment of the above method is described in more detail below, with reference to Figures 17 and 18. As illustrated in Figure 17, the method employs two plant lines, which are homozygous, respectively, for the lamH gene (plant line A2) and lamS genes (plant line A1) and otherwise isogenic. These genes are located on the same pair of chromosomes in the plant. Consequently, plants produced from a cross between these two isogenic lines will contain lamS and the lamH gene, respectively, on different chromosomes from an individual pair of chromosomes. This will ensure that the two genes will be segregated when this plant is crossed with a male fertility plant. A two-stage procedure is used to produce the hybrid seed. The first stage involves a pre-production of a line of

-9999 isogenic male sterility, the second step is the same production of the hybrid seed. To accomplish the first stage, the following approach is used: The two isogenic lines A1 and A2 are plants in rows as shown, and when flowering begins, the rows are sprayed with NAM (naphthalenacetamide). This chemical is a non-toxic version of the plant growth regulator NAA, and the action of the lamH gene converts NAM to NAA. Under the control of the promoter essential for the development and / or function of pollen, the lamH gene is only expressed in the pollen of the A2 line, and as such NAA is only made in the pollen of the A2 line. Since NAA is a regulator of plant growth, normal anther and microspore development is disrupted, leading to male sterility in plant line A2 when treated with NAM. Plants that contain the lamS gene under the control of the promoter essential for the development and / or function of pollen (plant line A1) are not affected by NAM, since they are unable to convert NAM to NAA, therefore these plants they remain fully male fertile and can be crossed by pollination with A2 plants, which are now male sterile after NAM treatment. In line A2, the seed is produced and contains both the lamH and lamS genes under the control of promoters essential for the development and / or function of pollen (A2 / A1 plant seed). The seed produced in line A2 (plant seed A2 / A1) is harvested. This harvest can be done by harvesting specific rows. Alternatively, the gene

-100100 lamH (A2) can be linked to a gene for herbicide resistance, so that the herbicide can be used for variation of plant line A1. The herbicide application occurs after flowering and will annihilate the A1 plants, so that only the seed that has the A1 / A2 genotype is produced. Seed harvested from such a field will produce substantially 100% male sterility plants. The cross produces plants that express both the lamS and lamH genes only in pollen. This leads to the conversion of tryptophan, a 10 amino acid normally found in plant cells, to IAM (indole acetamide) through lamS activity and finally to IAA (indolacetic acid) through lamH activity. The IAA molecule is a plant growth regulator not normally found in substantial amounts in the plant's developing or anther pollen grains. Since IAA and the precursor lam are small molecules that can be transferred from cell to cell through diffusion or active transport, altered levels of growth are seen throughout the anther. This altered growth regulatory level leads to an abnormality in pollen and anther development, producing a male sterile plant. This plant can be pollinated with a male fertility line leading to a commercial hybrid seed. For hybrid seed production, the isogenic male sterility line can be planted in rows along a suitable male fertility plant, and the seed

-101101 hybrid produced in the male sterility plant can be harvested. If the lamH gene is linked to a herbicidal resistance gene, harvesting of the hybrid seed is facilitated by using the herbicide to kill the pollinator plants after cross-pollination. Then the entire field can be combined. All the seeds produced will therefore be hybrid. If the two genes (lamS and lamH) are located on the same chromosome or in the same linking group, these two genes will be completely segregated in the F1 hybrid seed, since the plants will contain either the lamS gene or the lamH gene, but not both, the seed produced through this hybrid cross will be substantially 100% male fertility. Therefore, the plants developed from the seed of this cross will be fully fertile and will establish normal seed levels. The F2 seed resulting from the harvest of this field, however, will contain a variable degree of male sterility, since in theory 12.5% ​​or 2 out of 16 of the plants developed from this F2 seed will contain both lamS and lamH genes. , as illustrated in Figure 7. Therefore, particularly in poorly crossing species, the generation of F2 will not be attractive to the plant since up to 12.5% ​​of the plants will not produce seeds. Similarly, in crossing species, which are not fully pollinated, a corresponding reduction in yield will be seen.

It is contemplated that as a variation on the particularly preferred method above, a number of

-102 102 different ways to produce the toxic molecule specifically in pollen. In all approaches, at least one stage in the production of the cytotoxic molecule has to occur specifically within the pollen cells or anthers. For example, it is possible to use a lamS gene constitutively expressed in a plant and subsequently cross that plant with a plant that contains the lamH gene under the control of a promoter essential for the development and / or function of pollen, so that lam is produced in all the cells of the plant, but the growth regulating IAA is produced only in the pollen cells due to the action of the specific lamH gene in the pollen. Conversely, it is possible to have lamH constitutively expressed in the plant, and to cross this plant with a plant that contains a promoter essential for the development and / or function of pollen, activating the lamS gene. In this situation, the growth regulator IAA is only produced in pollen cells. Care must be taken that in this case, NAM cannot be used to induce transient male sterility in the plant containing the lamH gene, since the application of NAM could be lethal to the plant. In this case, then it is preferred that manual pollination of the combination of these genes be performed. With respect to these methods, the preferred embodiment of the present invention places both the lamH gene and the lamS gene under the control of promoters essential for development and / or pollen function and preferably using the same promoter essential for development and / or pollen function

-103103 or a promoter essential for the development and / or function of pollen, whose expression substantially overlaps that of the other of each one independently activating the expression of these two genes. Furthermore, by binding the IAMM gene to a selectable agent such as a herbicide, the production of the hybrid seed is greatly facilitated. Any number of genes can be used to carry out the invention as long as the simultaneous production of two or more of the enzymatic or synthetic activities specifically in the pollen leads to the production of a substance which is toxic or inhibitor of the development of normal pollen, or specifically interfere with the development of anther or pollen. This implies that one or more of these activities may be constitutive in the plant, but that of the final combination of all enzyme activities will be limited to pollen. It can also be seen that natural or surface media such as sterility can be induced in plants. Specifically, one embodiment of the method uses a plant line that carries a lamS gene for control of a promoter capable of being induced and a lamH gene under the control of a promoter essential for pollen development and / or function. These genes are preferably linked, but may be unlinked. When grown under inductive conditions, the plant becomes male sterile, and can be pollinated through a suitable male fertility plant. The suitable plant can also carry a lamS gene under the control of a promoter essential for the development and / or function of pollen in a way

-104104 that the progeny of this cross will be male sterility. These plants can then be crossed with a male fertility plant, producing a hybrid seed. This seed could, depending on the location and number of inserted genes, carry varying degrees of male sterility. It is also contemplated that a suitable male fertility plant is one that carries one or more copies of an antisense gene that is capable of selectively inhibiting the expression of one or more of these genes, so that 100% of fertility is recovered in plants developed from hybrid seed.

Another particularly preferred embodiment of the invention uses lamS and lamH genes to produce an origin of male sterility A as described herein, with two genes introduced at Origin A as a recombinant DNA molecule, i.e. linked. This could have the advantage that the two genes could not segregate in T1, as could be the case with the two genes that are introduced in isogenic lines, on the same chromosomal pair. Furthermore, it is proposed that the recombinant DNA molecule of this particularly preferred embodiment contains, linked to the lamS and lamH genes, a gene that serves as a selection marker for transformation. It is preferred that the selection marker is a herbicide resistant gene to glufosinate (its trade names are Basta and Liberty) or glyphosate (Roundup) or TBL as discussed above. Origin A could be kept in a homozygous state for the gene that

-105105 produces the male sterility trait (lams and lamH in this modality), and for the resistance marker (herbicide resistance in this modality) through gamete transformation methods as described above. Origin A is maintained by crossing with an isogenic, but untransformed, origin A and spraying herbicide, as shown in Figures 20 and 21.

Origin B could be the masculine origin. In a preferred case, this may be a plant line produced through transformation with the herbicidal resistance gene, using methods described above and being homozygous for the herbicidal resistance gene. As described in Figure 21, the hybrid seed could be 100% herbicide resistant.

In an alternative embodiment, if a male fertility seed is desired, Origin B could be produced using a recombinant DNA molecule containing both male fertility restoration and herbicide resistance genes, as described in Figure 20. The seed The resulting hybrid is 100% herbicide resistant and 100% male fertility.

Of course, origin A can be produced by crossing isogenic lines, each containing one of the two genes, ie lamS and lamH, bound to the herbicide resistance marker. Both the plant carrying the lamS gene and the plant carrying the lamH gene are made homozygous for the genes by anther transformation or cell culture methods of

-106106 microspore or through purification of the plant as described above. This is illustrated in Figure 22a. Origin A is selected and maintained as described above and as illustrated in Figure 22b. Origin A intersects with Origin B, produced as before. The hybrid seed is 100% herbicide resistant, and if Origin B contains a male restoration gene, then the hybrid seed is 100% male fertility. The resulting plant of the crossing and the hybrid seed is 100% resistant to herbicide and is either 100% male sterile or 100% male fertility, as determined through the recombinant DNA molecule that was chosen to be introduced into the genome of origin B. DNA sequences that can serve to restore male fertility to a hybrid plant, where an origin is made of male sterility, were discussed above. A preferred restorer gene could be an antisense gene for either gene 1 and / or gene 2 (lamS and lamH genes, described above). The use of an antisense gene was as previously described. Another preferred male fertility system in the hybrid plant could use depressing mechanisms of gene expression, especially the lexA system discussed above. A transactivator can also be converted to a repressor to create a dominant negative inhibitor, which can inhibit the expression of target promoters linked to gene 1 and / or gene 2. In another preferred embodiment, the judicious construction of promoter sequences from

-107107 manner containing recognition sites for binding through a transactivator or transdominator could allow greater control of lamS and / or lamH expression. If the transdominator sequences themselves may be able to be induced through some externally applicable chemical, the plant with the male sterility genes, such as a fertile plant, could be propagated to utilize the need for such a plant as male sterility in hybrid crossing.

To produce a male sterility origin A for the hybrid through the use of the linked gene 1 and gene 2 scheme, described above, it may be essential that the expression of at least gene 1, or gene 2, or both , is regulated using a promoter active only in cells involved in the formation / structure of pollen. Such promoters were previously described. A preferred case could employ the anther specific promoters described above, B10 supplied from the L10 Brassica gene and / or the anther specific gene from tobacco, TA39. The herbicide resistance gene can be constitutively expressed.

In Figure 1, a schematic representation of the production of the antisense PAL1302 vector is shown. A plasmid containing the GUS gene (beta-glucuronidase, described in Jefferson, RA, Plant Molecular Biology Reporter, 1987, 5: 387-405) in the antisense orientation flanked by the CaMV 35S promoter and the termination signal was obtained from the vector pBI221.1

-108108 (available from Clonetech Laboratories, Palo Alto, CA, USA). The GUS coding sequence found between the CaMV 35S promoter and nos of vector pBI221.1, was excised and digested with the restriction enzymes Smal and Sstl. The Sstl site was made from shaved ends using the Klenow fragment of DNA polymerase I and the shaved end vector and the GUS coding sequence were relegated. A plasmid (pPAL303) containing the Gus coding sequence inverted with respect to the transcription direction of the CaMV 35S promoter was identified.

The binary PAL1302 vector containing the antisense GUS gene was constructed using vector pVU1011 (obtained from The Plant Breeding Institute, Cambridge, UK). pVU1011 contains the hygromycin phosphotransferase coding sequence flanked through the CaMV 35S promoter and is not inserted into the Agrobacterium Bin19 binary vector polylinker described by

Bevan, M., Nucí. Acids Res. 1984, 12: 871 1-8721. The vector pVU1011 can confer both hygromycin and kanamycin resistance to transformed plant cells. Insertion of the antisense GUS nos ter fragment of the CaMV35S promoter into 20 pVU1011 was accomplished in such a way that it inactivated the NPTII gene of this vector and was performed as follows. A small Sphl-Pstl restriction fragment containing the right limiter (RB), the NOS promoter, and the start of the NPT II coding sequence from pVU1011 was first subcloned into the Sphl and Pstl sites preceding the CaMV 35S promoter from pPAL303 to the form pPAL306. The digestion

-109109 from pPAI_306 with Sphl and EcoRI released a fragment consisting of the RB, the NOS promoter, the start of the NPT II coding sequence and the CaMV 35S promoter, the antisense GUS noster construct. This fragment was then ligated to the Sphl sites of pVU1011 by adding the elongation pGEM-4Z cut (Promega biotech, Madison, Wl, USA) with EcoRI and Sphl to provide a small polylinker fragment as a bridge between the Sphl site of pVU1011. and the EcoRI site of the insert from pPAL306 respectively. Insert orientation was verified and a binary vector (PAL1302) possessing a reconstructed RB fragment and an NPT II gene inactivated through the CaMV 35S promoter, insertion of the antisense GUS noster gene, was identified. This vector can only confer hygromycin resistance to plants and carries the antisense GUS gene.

In Figures 2a-d the orientation of the genes contained within four specific microspore clones from Brassica napus are 5 'to 3'. As shown, the 5 'region corresponds to the promoter region and is identified with a small arrowhead. The 3 'region delineates the end point of gene transcription. The L4, L10 and L19 genes were used for the isolation of the microspore-specific promoter fragments and for the isolation of the microspore-specific coding regions. Non-transcribed regions are identified as a single thin line, while clones that are transcribed are delimited by a box area.

-110110

Within this box area, the portion of the transcribed DNA representing the exon regions is delimited by being filled with black, while the intron sequences are left unfilled. The approximate regions of the sequenced DNA for clones L10, L16 and L19 are shown through the underline. The identified restriction sites are those that are relevant to the constructs detailed below. The right and left arms of the lambda cloning vectors are not shown.

In Figures 3a-d, the complete DNA sequence of clone L4 is shown along with the DNA sequence of the portions of clones L10, L16 and L19 that are identified in Figures 2a-d. In Figure 3a, clone L4, nucleotide 1 in the full sequence is at the EcoRI site furthest to the left, while nucleotide 8579 is at the first nucleotide of the EcoRI site furthest to the right. The start of gene 1 transcription in clone L4 is nucleotide 235. The 5 'and 3' intron binding sites are identified in the bold type. The start ATG codon is also shown as the stop stop codon. The deduced amino acid sequence of the proteins encoded through these genes is also shown. The end of transcription for gene 1 is approximately nucleotide 1427. The second gene in the L4 clone is probably non-functional due to an insertion and deletion that occurs in the region of the promoter and the first exon. This gene was not used for constructs. The third gene in clone L4 has a transcriptional start in number

-111111 position 6298 in the DNA sequence and transcription ends at approximately nucleotide 7490. The ATG start codon, intron binding sites and termination stop codon are all as identified above. Vectors were constructed from this clone using promoter fragments from both genes 1 and 2, as well as promoter fragments from genes 1 and 2 that contained the first exon and intron sequences and a short portion of the second exon to each of the genes. The specific promoter fragment constructs are detailed below.

In Figure 3b, the nucleotide sequence of the region of clone L10 delimited in Figure 2b is shown. The initiation of transcription is at nucleotide 1. In this sequence, the ATG start codon is at nucleotides 45-47, the first exon ends at nucleotide 315, the second exon starts at nucleotide 476 and expands towards nucleotide 1586. The third exon begins in 1673 and extends approximately to nucleotide 1989. The precise end of transcription was not determined. The deduced amino acid sequence is also shown. For some promoter constructs, the region from clone 5 'to the sequenced portion was used. Specific construction details are listed below.

In Figure 3c, the nucleotide sequence of L16 is shown. Clone L16 shows considerable homology to clone L10 specifically in the portions of the two clones that encode

-112112 for the protein sequence. The intron sequences between the two clones differ considerably. The L16 clone does not contain a 5 'promoter region and as such was only used as a source of coding sequences for the 5 antisense RNA constructs. Nucleotide 1 outlines an EcoRI site that occurs in a DNA coding region that is homologous to the first exon of L10. By homology, this coding region extends to nucleotide 124, where the first intron is located. This intron, which is located in the same relative position as the first intron of clone 10, is larger than the intron in clone L10 and extends to nucleotide 688. Nucleotide 689 is the start of the second exon and this exon, which shows strong homology to the second exon of clone L10, extends to nucleotide 1793. There is

<td colspan="6">a second intron at this point and this intron extends</td><td>toward</td><td>the</td>

<td>15 nucleotide 1909.</td><td>The</td><td>third exon</td><td>start</td><td>in 1910 and</td><td>I know</td><td colspan="2">extends</td>

<td>approximately</td><td>to the</td><td>nucleotide</td><td> 2210.</td><td>I also know</td><td colspan="2">shows</td><td>the</td>

deduced amino acid sequence for specific regions of the clone that show considerable homology to clone L10. The precise nucleotide where transcription stops, has not been determined.

In Figure 3d, the nucleotide sequence of the region of the L19 clone delimited in Figure 2d is shown. The start of the transcription is located at position 1 in the sequence. The ATG start codon is at nucleotides 136-138 and the first 25 intron starts at nucleotide 1201. This intron ends at

-113113 nucleotide 1338, where the second exon starts. The end of transcription occurs at approximately nucleotide 2074. The deduced amino acid sequence is also shown.

In Figure 4, the DNA sequences of the three cDNA clones that are homologous to the genes contained in clone L4 are shown. The DNA sequence of these three cDNA clones as well as the sequence of the correctly linked transcribed regions of the Bp4A and Bp4C genes in the L4 genomic clone are aligned, only the nucleotide differences within these clones are shown. Nucleotides that are conserved between sequences are only shown in the sequence above. The asterisks shown in Figure 4 mark the 5 'end of the cBp401, cBp405 and cBp408 cDNA clones.

In Figure 5, the partial nucleotide sequence of a cDNA clone that is homologous to the coding region of clone L10 is shown. This cDNA clone is approximately 1.3 Kb in length and has EcoRI sites at the 5 'and 3' ends of the cDNA sequence that were aggregated through synthetic linkers in the cDNA cloning procedure.

In Figure 6, the nucleotide sequence of the cDNA clone corresponding to the coding region of clone L19 is shown. Identified in this sequence is the EcoRV site present at the 5 'end of the L19 transcribed region. A portion of the poly A end is shown. EcoRI sites that were added as linkers in the procedure are not shown.

-114114 cloning cDNA; these sites are present adjacent to the 5 'and 3' ends of the cDNA clone.

The construction of six vectors containing the promoter and promoter fragments of the L4 clone is described in Figure 7 (a, b, c, d, e). The first stage in the construction of these vectors was achieved by first subcloning the EcoRI-Sstl fragment (nuc. 1-2132) containing the first gene from the L4 clone (235 base pairs of promoter / exon / intron / second exon) in a vector commercially available pGEM-4Z (Promega Biotech, Madison, Wl, USA), using the EcoRI-Sstl sites of the polylinker of this vector. This plasmid was named pPAL0402. The 2.7 Kb EcoRI fragment from clone L4 containing the third gene (Bp4C) was then cloned into the EcoRI site of pGEM 4Z, leading to a plasmid named pPAL0411. Plasmid pPAL0402 was then digested with EcoRI and the 2.7 Kb EcoRI fragment from pPAL0411 (nuc. 5859-8579) containing all three gene numbers (Bp4C) from clone L4 that was added to it. The clones were recovered and contained this 2.7 kb EcoRI fragment inserted in both orientations relative to the promoter region of the first gene. A clone that contained this third gene fragment in one orientation, so that the promoter of the third gene was opposite to the promoter in the first gene, was chosen and named as pPAL0403. Plasmid pPAL0403 contains the entire third gene of the L4 clone so oriented that it has the promoter region immediately adjacent to the promoter region of 235 pairs of

-115115 base of the first gene in pPAL04303. This plasmid, pPAL0403 was digested with Ddel, producing a fragment of approximately 1.9 kb. Ddel sites are located at nucleotides 303 and 7366. Due to the orientation of these fragments, Ddel digestion produces a 1.9 Kb fragment. This 1.9 Kb fragment contains a copy of the third gene (Bp4C) oriented, so that the transcription direction of this third gene is from right to left, fused to the 235 base pair promoter fragment from the first gene of the clone L4 (Bp4A) which was transcribed from left to right, ending at the Ddel site which is located 67 base pairs downstream of the main transcription start site and precedes that ATG start of the translational codon through 2 nucleotides. This 1.9 Kb Ddel fragment was shaved with the klenow fragment and cloned into the Xbal site of the previously end-shaved pGEM 4Z polylinker region with the klenow fragment. The resulting plasmid pPAL0408 was recovered and subsequently digested with Sal I and Sstl, which released the cloned Ddel fragment delimited by the left side (nucí 7366) Salí and on the right side (nucí 303) of this construction and contains a portion of the pGEM 4Z polylinker containing the following unique sites: BamHI, Smal, Kpnl and Sstl restriction enzyme sites. The Salí-Sstl fragment was cloned into the Salí-Sstl sites of PAL1001. PAL1001 is a Bin19 binary vector (described by Bevan, M., Nucleic Acids Res., 1984, 12: 871 1-8721) to which the polyadenylation signal of

-116116 we ter as a 260 bp Sstl-EcoRI fragment isolated from plasmid pRAJ221 (available from Clonetech Laboratories, Palo Alto, CA USA) at the Sstl-EcoRI sites of the Bin19 polylinker region. This nos ter was identified as a dotted box. The binary transformation vector resulting from the insertion of the Salí-Sstl fragment from pPAL0408 to PAL1001 was named PAL1107. Construction details are shown in Figure 7a. This vector has a copy of the third oriented gene, so that the transcriptional orientation of this third gene is from right to left, fused to the 235 base pair promoter fragment from the first gene of clone L4, which was transcribed from left to right, followed by a polylinker with unique sites for DNA insertion consisting of: BamHI, Smal, Kpnl and Sstl followed by signal nos. This vector has the characteristic that additional 5 'non-coding sequences were placed upstream of the 235 base pair core promoter on Bp4A, but these additional 5' sequences were in opposite orientation. The provision of these sequences in this orientation does not affect the specificity of the 235 base pair promoter nucleus pollen.

In addition to this vector, similarly structured vectors were made, which contained essentially the same type of gene promoter array but contained the intron of the first gene (Bp4A) from the L4 clone. Intron sequences in plant genes were shown in some cases and play a role

-117117 important in gene expression. This vector-containing intron was constructed by doing a deletion run on clone pPAL0402. pPAL0402 was first digested with Pstl and Smal. Exonuclease III was used to unidirectionally digest DNA as shown (Figure 7b). After treatment with S1 nuclease and repair with Klenow, the plasmid was relegated and clones that had different portions of coding regions of the Bp4A gene digested outside of them were recovered. Elimination subclones were sequenced. One was chosen for the vector constructions. This was named as 23B deletion. This subclone represented a deletion that has the majority of the second exon of the Bp4A gene removed, but contains the intron binding site and the first exon of the Bp4A gene. This subclone contains a portion of the L4 clone that extends from nucleotide 1 to nucleotide 1166. To this subclone was added the 2.7 Kb EcoRI fragment from pPAL0411 containing the third gene of L4 (Bp4C) in such an orientation that the direction of transcription of the third gene is from right to left (as in PAL1107, pPAL0408 ) fused to the 235 base pair promoter region from the first gene of clone L4, which is oriented to transcribe from left to right, followed by the first exon of gene 1, all the intron of gene 1 and 33 nucleotides of the second exon of the Bp4A gene of clone L4. This plasmid containing the 23B deletion and the 2.7 Kb EcoRI fragment containing the third gene fragment was named pPAL0406. This plasmid was

-118118 digested with Hindlll, which produces a fragment containing a small portion of the third gene promoter, as well as the entire promoter of the first gene, the first exon, intron, and a portion of the second exon. This Hindlll fragment was inserted into the Hindlll site of PAL1001, resulting in vector PAL1106 (deletion 23B derived). This vector has, in the following order, a portion of the promoter of the third gene promoter in clone L4, the entire 235 base pair promoter of the first gene in the clone.

L4, followed by the first exon, the intron and a portion of the second 10 exon of gene 1 from clone L4, followed by a polylinker containing the following unique cloning sites: Salí, Xbal,

BamHI, Smal, Kpnl and Sstl and the nos ter polyadenylation signal. The construction is shown in Figure 7b.

Additional constructs were made with the promoter regions of the genes contained in clone L4, in order to provide a number of suitable vectors that are useful for the specific expression of gene sequence pollen. All three genes within the L4 clone (Bp4A, Bp4B, p4C) show very accurate DNA homology and this is very evident between the first gene (Bp4A) and the third gene (Bp4C). The second gene (Bp4B) is a homologous copy that has undergone sequence changes that appear to have led to inactivation. The extensive similarity between the first, second and third genes in the L4 clone was also maintained in the promoter region, such that outside of the first 235 nucleotides of the first and third regions of

-119119 gene promoter There are only 5 nucleotides that differ from each other. Downstream of the TATA box in these two promoters, the only difference between them is the presence of an additional nucleotide at the start of transcription. For example, the comparison of promoter 1, Bp4A, partially represented by: ...... TATGTTTtAAAA ... with Promoter 3, Bp4C, partially represented: ........ TATGTTTAAAA ..... shows that the underlined transcribed region and the unique lowercase nucleotide difference. However, within the sequence of the first gene 10 there is a nucleotide change that introduces a Ddel site (nucí.

303) in the 5 'untranslated leader sequence upstream of the ATG start codon that is not present in the untranscribed third gene leader sequence in clone L4. Chimeric promoter constructs were made which used this Ddei site and the first 15 to combine with sequences from the third promoter of the gene. The region of the first promoter used for these constructs consisted of the sequences contained between the site

SnaBI (nucí 21) near the TAT box towards the Ddel site located immediately upstream of the ATG start codon in the first 20 gene (nucleotide 303 is the first nucleotide in the Ddel recognition sequence). The other region of this chimeric promoter (5 'from the TATA box) was a fragment extending from the EcoRI site of the third promoter (nucleotide 5858) to the SnaBI site near the TATA box (nucleotide 6273). Thus,

-120120 To facilitate the construction of these specific vectors in pollen, the following reconstructions were performed.

The EcoRI to Ddel fragment spanning the promoter region of the first gene in clone L4 was isolated by first cutting pPAL0402 with Ddel, shaving with Klenow, and then cutting with EcoRI. The 235 base pair fragment corresponding to this region was cloned into the EcoRI-Smal sites of pGEM 4Z. This plasmid (pPAL0422) was then cut with EcoRI and Snabl. A DNA fragment containing the EcoRI to SnaBI portion of the promoter for gene 3 in clone L4 was isolated by digestion of pPAL0411 with EcoRI and SnaBI. This released an EcoRI of approximately 415 base pairs (nuc. 5858) to the SnaBI fragment (nuc. 6273) that represents the majority of the 5 'region of the L4 clone gene 3 promoter (the SnaBI recognition site is 2 pairs of bases downstream of the TATA box). The EcoRI-SnaBI fragment was used to replace the shorter EcoRI-Snabl fragment removed from the first promoter subclone (pPAL0422) by reconstructing a promoter fragment of approximately 550 base pairs. This plasmid is called pPAL0421. This chimeric promoter fragment contains 415 base pairs of the promoter of the three genes in clone 4, followed by approximately 99 nucleotides the first gene promoter / untranslated leader sequence.

For the construction of a pollen specific cassette vector, the following plasmids were first constructed. The first

-121121 constructed plasmid contained the nos ter polyadenylation signal with a polylinker in front of the nos ter signal. This was achieved by first isolating nos pRAJ22 as a Satl Ecorl fragment and this fragment was cloned into pGEM 4Z using the Sstl and EcoRI sites in the polylinker. This subclone is named as pPAL.001. To pPALOOl, a fragment encoding neomycin phosphotransferase (NPT II) derived from plasmid pRAJ162 was added, in the antisense orientation as follows: Plasmid pRAJ162 contains the NPT II gene from transposon TN5 inserted as a Sali fragment and linked through of a polylinker on plasmid pUC-9 (which was obtained from the Plant Breeding Institute, Cambridge, UK). PRAJ162 was digested with Hindlll and Smal. The DNA fragment containing the NPT II gene was isolated through elution from an agarose gel. pPAL.001 was digested with Hindlll and Smal and the NPT II gene fragment was inserted. The resulting plasmid was named pPAL002 and had such restriction site orientation and NPT II gene and we ter as follows: HINDIII, Pstl, Sali, 3 'and NPT II the 5' end and 3 'end coding sequence , Salí, BamHI, Saml, Kpnl, Sstl, nos ter, EcoRI. pPAL002 was cut with Hindlll and the site was shaved end through the use of the klenow fragment. pPAL0421 was digested with Hincll and pVLIII, both of which leave shaved ends, and the promoter fragment was ligated to Hindlll cut shaved end pPAL002. Plasmids were obtained and contained the promoter in both orientations with

-122122 in relation to the signal nos. A plasmid was chosen in the proper orientation (antisense 57NPT II promoter / nos ter) and was named pPAL0419. PPAL0419 stains the following DNA fragments: A small (approximately 130 bp) pGEM 4Z containing the SP6 promoter, the 550 base pair promoter, the NPT II gene in the antisense orientation relative to the promoter, followed by the signal of polyadenylation we ter. This complete promoter / NPT ll / nos ter construct can be cut through EcoRI. pPAL0419 was digested with EcoRI, and the NPT II promoter from structure nos ter was cloned into Bin19 using the EcoRI single site in the Bin19 polylinker. The resulting transformation vector was named PAL1419. In addition to the antisense NPT II gene, the vector contains a constitutive NPT II gene under the control of the nos promoter. This vector, therefore, confers resistance to kanamycin in all cell types with the exception of pollen cells where the expression of the gene from the constitutive promoter is inhibited through the antisense RNA produced from the construction of promoter / NPT ll / nos ter contained in PAL1419.

In order to provide promoter sequences that can be used with additional gene constructs, plasmid pPAL0419 was digested with Sali. This digestion removes the NPT II coding region and this Salí digested pPAL0149 was relegated giving rise to pPAL0420. pPAL0420 represents the essential promoter for the development and / or function of pollen followed

-123123 by a polylinker for gene insertion having the following unique sites: Hincll, Pstl, Sali, BamHI, Smal, Kpnl, Sstl, followed by the polyadenylation signal from nos. The entire promoter / polylinker / nos ter construct can be conveniently excised as a single EcoRI fragment. The details of this construction are shown in Figure 7c.

For additional promoters essential for pollen development and / or function, the following approach was used. The intact L4 clone in the lambda cloning vector was digested to complete with Sstl and Hhal restriction enzymes. The resulting fragments were separated by gel electrophoresis and a 2.65 Kb fragment containing the promoter / first exon / intron / second exon region of the three gene in clone L4 and corresponds to nucleotides 4565 to 7216 in the sequence of clone L4, was isolated. This fragment was made from klenow shaved ends and cloned into the previously described binary transformation vector PAL1001. PAL1001 is first cut with Hindlll and made from shaved ends with Klenow. Clones containing this fragment were recovered (promoter / first exon / intron / second partial exon). A clone was chosen that contained this fragment in the proper orientation so that the direction of transcription was toward nos in PAL1001. This vector was named PAL1421. This vector contains approximately 1.9 Kb of the promoter region upstream of gene 3 in clone L4 followed by the first exon, the complete intron and 15 bases of the

-124124 second exon of the gene of three followed by a polylinker containing the following unique sites: Salí, Xbal, BamHI, Smal, Kpnl, Sstl, and finally the polyadenylation signal nos. A variant of this vector was constructed by digesting PAL1421 with EcoRI and isolating the fragment of this clone that contains the promoter polylinker sequences but contained less of the region upstream of the promoter. This fragment was re-cloned into PAL1009. PAL1009 is a Bin19 derived from the vector from which most of the polylinker was removed. This vector was constructed by digesting Bin19 with Hindlll and Sstl, and making these shaved end sites with Klenow and relegating such that a vector was recovered and contained a single EcoRI single site for fragment insertion. PAL1009 was digested with EcoRI, and the EcoRI fragment from PAL1421 containing the shorter promoter / exon / intron / second exon / polylinker / nos ter structure was added thereto. This gave rise to vector PAL1422, a vector that is essentially the same as PAL1421 with the exception that it is a minor 5 'promoter region. It should be noted that both PAL1421 and PAL1422 contain the intron of the third gene. For constructs where the presence of the intron is not desired, the intron sequences from PAL1421 were removed by first digesting PAL1421 with EcoRI and replacing the promoter / exon / intron / second exon / polylinker / noster structure with the promoter / polylinker / we ter from pPAL0420

-125125 using EcoRI, such that the longest 5 'promoter region was rebuilt in the binary transformation vector. The resulting vector was named PAL1423. The important aspect of this construction is shown in Figure 7d.

In Figure 7e, a schematic diagram of the relationship of the aforementioned vectors is presented. It should be noted that the vectors underlined in the Figure fall into three categories: 1) vectors containing 5 'upstream promoter regions that are substantially derived from the upstream of the Bp4C 10 gene (pPAL0420, PAL1420, PAL1423), 2) promoter constructs containing 5' upstream promoter regions and sequences of intron from the Bp4C gene (PAL1422, PAL1421) and 3) promoters which contain a chimeric 5 'upstream region where a portion of the 5' DNA sequence is inverted relative to arrangement, which appears in the genomic clone and uses the Bp4A promoter fragment as a core promoter structure (PAL1107, PAL1106). It should be noted that the operation of each of these constructs may vary from plant species to plant species and it may be desirable to test a number of these promoter constructions when carrying out certain aspects of the invention.

The construction of pollen specific vectors using the promoter regions of the L10 and L19 clones was conducted as follows. The construction of pollen specific vectors 25 depicted in Figure 8 uses promoter regions

-126126 of clone L10. The Start of transcription of the L10 clone is located at nucleotide 1. The ATG start codon is located at nucleotides 4547. The promoter region of this clone was excised by first subcloning the EcoRI-Xbal fragment of the clone that covers the entire region of promoter a portion of the first exon (the Xbal site is nucleotide 358 in the DNA sequence). This subclone (pPALIOEX) was then digested with Hincll and Ndel. The Ndel site is located immediately downstream of the ATG start codon at nucleotide 60 and the Hincll site is located at nucleotide number -399.

io Digestion with these two enzymes releases a DNA fragment from

459 nucleotides, which contains 62 nucleotides of untranslated transcribed leader sequence, and 397 nucleotides from the promoter region

5'. The Ndel site in this fragment was made from shaved ends through the use of klenow, and this fragment was subcloned to the Hincll site 15 of the pGEM 4Z polylinker. Clones were recovered in both orientations and the clone containing the fragment in orientation: Hindlll, Sphl, Pstl, Hincll, 62 base pair promoter leader fragment (Ndel blunt / Hincll, not cut with Hincll or Ndel) Xbal, BamHI, Smal, Kpnl, Sstl, EcoRI was chosen and named 20 pPAL1020. To add additional upstream regions, the Hincll-Hincll fragment that is approximately 1 Kb in length and is immediately upstream of the Hincll site at position 391 in the DNA sequence was isolated from pPALIOEX through digestion with Hincll and gel elution of this fragment.

The Hincll fragment was cloned into the Smal site of pGEM 4Z. The

-127127 clones, which contained the fragment in both orientations, were recovered and a clone was chosen that contained the fragment in the following orientation: Hindlll, Sphl, Pstl, Hincll, Salí, Xbal, BamHI, the Hincll fragment in the same orientation as in genomic clone 5, which went from right to left, 5'-3 '(as in the Hincll / Smal insert that was not cut with any enzyme), Kpnl, Sstl, EcoRI. This subclone (pPALIOHc) was digested with Κηρ I, made from shaved end through the use of klenow, then it was digested with EcoRI. The promoter / untranslated leader sequence of pPAL1020 was added to this cut subclone by digesting pPAL1020 with Hincll and EcoRI, and adding this promoter fragment to the cut pPALIOHc. The resulting subclone contained a reconstructed promoter region from clone L10 differing from the intact region only by the Kpnl-filled site used for binding of the two promoter fragments. This construction was named pPAL1021. This vector contains in the following order: Hind111, Pstl, Sphl, Hincll, Salí, Xbal, BamHI, the approximately 1 Kb Hincll fragment linked to the Hincll-Ndel promoter fragment followed by Xbal, BamHI, Smal, Kpnl, Sstl and EcoRI .

This subclone allows convenient removal of the promoter region from the L10 clone, so that the promoter can be easily used in cassette transformation vectors. The summary of this construction is shown in Figure 8. The promoter region of pPAL1021 was used for the construction of a pollen-specific cassette transformation vector by performing

-128128 the following constructs: Plasmid pPAL1021 was digested with Ncol and PstI. The plasmid was klenow treated and relegated. This procedure effectively removed the portion of the polylinker that was 5 'to the promoter in pPAL1021. This plasmid was then digested with Hindlll and Sstl, and cloned into the Hindlll and Sstl sites of PAL1001, giving rise to PAL1121. PAL1121 has the following order: the promoter essential for the development and / or function of the pollen of clone L10 (approximately 1.1-1.2 Kb), followed by a polylinker with the following unique sites: Xbal, BamHI, Smal, Kpnl, Sstl, followed for us ter. The construction of this is summarized in Figure 8. The L10 promoter isolated here is also referred to as the Bp10 promoter. Also see Albani, D., Sardana, R., Robert. LS, Altosaar, I., Arnison, PG, and Fabijanski, SF, A family of the Brassica napus gene showing sequence similarity with ascorbate oxidase is expressed in developing pollen. Molecular characterization and analysis of promoter activity in transgenic tobacco plants, Plant J. 2: 331-342 (1992).

The L19 clone promoter region was also used for the construction of specific vectors in pollen. The construction of these vectors is as shown in Figure 9. Clone L19 has a specific gene for individual pollen contained therein. The start of transcription of this gene is located at position 1 in the DNA sequence. The ATG start codon is located at nucleotide position 136-138. Only the intron is located at nucleotides 1202-1387, the translational codon of

-129129 termination is located at nucleotides 2024-2026. The end of the transcript is located approximately at nucleotide 2074. The entire EcoRI fragment of this clone was subcloned into PGEM 4Z using the EcoRI site located on the polylinker. The resulting clone was named pPAL1901. The promoter region of this clone was excised as a single fragment by digestion of pPAL1901 with BamHI and EcoRV, and a 2177 base pair fragment corresponding to the promoter region was isolated. This fragment covers nucleotide -2200 (BamHI) to nucleotide 156 (EcoRV). This promoter fragment contains a 5 'upstream 2KB region of the promoter in clone L19, 134 5 base pairs of the promoter region of clone L19 were also used for the construction of pollen specific vectors. The construction of these vectors is as shown in Figure 9. Clone L19 has a specific gene for individual pollen contained therein. The start of transcription in this gene is located at position 1 in the DNA sequence. The ATG start codon is located at nucleotide position 136-138. Only the intron is located at nucleotides 1202-1 387, the termination translation codon is located at nucleotides 2024-2026. The end of the transcript is located at approximately nucleotide 2074. The entire EcoRI fragment of this clone was subcloned into PGEM 4Z using the EcoRI site located on the polylinker. The resulting clone was named pPAL1901. The promoter region of this clone was excised as a single fragment through the

-130130 digestion of pPAL1901 with BamHI and EcoRV, and a 2177 base pair fragment corresponding to the promoter region was isolated. This fragment covers nucleotide -2200 (BamHI) to nucleotide 156 (EcoRV). This promoter fragment contains about 2 Kb of the 5 'upstream region of the promoter in clone L19, 134 base pairs of the 5' untranslated leader sequence and 20 base pairs of the translated sequence. The BamHI site in this fragment was shaved ends through the use of klenow and cloned into PAL1001. This step was accomplished by cutting PAL1001 with HindIII, causing this site to have shaved ends through the use of klenow and inserting the BamHI-EcoRV fragment of shaved ends in such an orientation that the promoter was oriented 5 'to 3' with respect to the polylinker / polyadenylation signal we ter. This vector was named PAL1920 and contained within it in the following order: the promoter from clone L19 containing 134 base pairs of the 5 'untranslated leader sequence, 20 base pairs of the translated sequence fused to the polylinker containing a forming site of HindIII inactivated through Sphl, Pstl, Salí, Hincll, Xbal, BamHI, Smal, Kpnl, Sstl of shaved ends (the only 20 cloning sites are underlined), the polyadenylation signal from us. This vector is suitable for insertion of DNA sequences that will be transcribed into pollen cells. The summary of this construction is shown in Figure 9. A plasmid for the future construction downstream of the TA39 promoter was constructed.

-131131 as follows, the TA39 promoter was cut and isolated from pPHP1481 as the Pstl fragment.

This was ligated into pPHP585 (i.e., BlueScript SK +) cut with Pstl and Notl, along with the Notl-Pstl fragment from pPHP1428, and classified for construction containing both PStl and Pstl-Notl fragments. The classification employed in addition to Pstl-Notl, Pvull, and Hindlll digestions. The resulting plasmid was pPHP1428. That portion of pPHP1494 incorporated into the lamS construct pPHP5840 (Figure 25) and the 10-lamH construct, pPHP5841 (Figure 26) was the Hindlll-BamHI fragment containing the TA39 (883) and θ 'promoter.

In Figure 10, the restriction map of a Brassica napus (HP101) genomic clone containing a constitutively expressed gene is shown, and fragment 15 of this clone containing a 5 'promoter region is identified along with a portion of a transcribed sequence. This fragment was isolated by first cloning the small 2.5 Kb EcoRI fragment into pGEM 4Z, and obtaining a subclone that had this fragment inserted in the orientation indicated relative to the 20 pGEM 4Z polylinker. This clone was named pPAL0101 and was deposited on January 26, 1990 at the American Type Culture Collection (ATCC), 12301 Parklawn Drive, Rockville, MD, 20852, USA as pPAL0101 / alpha of E. coli strain DH5 under the accession number. ATCC 68210. This strain of E. coli was grown in normal E. coli (LB) medium with 25 100 micrograms per ml of ampicillin. This subclone, pPAL0101,

-132132 was then digested with EcoRI, treated with the klenow fragment, then digested with BamHI, which released the indicated promoter / transcribed region. This fragment was cloned into Hincll-BamHI cut from pGEM 4Z, resulting in subclone pPALHP101. The 5 subclone can be used for the isolation of promoter sequences in vector constructs that use a constitutive promoter to synthesize specific antisense RNA in pollen.

In Figure 11, a schematic representation of the production of a polylysine encoding gene is shown. In this construct, the pGEM 4Z cloning vector was used as a container for a synthetic oligonucleotide containing an ATG start codon and a polynucleotide consisting of only one nucleotide was added to it. Subsequently, this gene, depending on the nucleotide used, is a gene that predominantly has a codon, and codes for a protein that is composed of a polyamino acid. Construction was performed as follows: To provide an ATG start codon in a favorable start context, a synthetic oligonucleotide 20 was constructed and pGEM 4Z inserted between the Hindlll and Sphl sites. This nucleotide had the sequence: 5'-ACGTGGATCCAAGATGACATG-3 '. The resulting subclone then had the DNA sequence (restriction site for an entered BamHI site underlined, and the ATG start codon is in bold) AACGT GGATCC AAG ATG ACA TGC 25 GCA ACA TGG in the 5 'region, so which was a start codon

-133133 of ATG in a favorable initiation context and a BamHI site upstream of this site for excision of the coding sequence. This subclone was digested with Pstl, divided between two aliquots, one was designed with T residues using terminal transferase and TTP, one was designed with A residues using terminal transferase and dATP. The two designed plasmids were mixed together, phenol-chloroform extracted, ethanol precipitated and resuspended. The plasmid mixture was cut with Sstl and relegated. The clones that were recovered were either clones containing poly A of the classification chain structure or poly T. The clones were cut with BamHI, the size of the insert determined by gel electrophoresis and sequenced to determine if the clone encoded for poly-lysine (poly A) or poly-phen Halan in (poly T, poly U in the MRNA). A clone encoding poly-lysine was chosen and was approximately 300 nucleotides, and was named pPAL pLys. This clone was cut with Xbal, made from klenow shaved ends, and a universal translation terminator (available from Pharmacia PL Biochemicals, Montreal, Canada) was added thereto, completing the construction of the gene. This construction is shown in Figure 10.

In Figure 12, a scheme is shown for the production of antisense gene specific to the intron region of clone L19 and the production of an intron-free version of gene number L19. For the first stage in this, a restriction fragment was isolated

-134134 of clone L19 which is substantially all intron. This fragment was isolated using the restriction enzyme Ddel which cuts at a number of sites in the genomic clone, but the sites at nucleotides 1186 and 1348 give rise to a restriction fragment that is substantially intron sequences, having only about 16 nucleotides on the 5 'side of the intron that are included in the final transcript, and 10 nucleotides on the 3' side of the intron that are included in the final transcript. This Ddel fragment was isolated through gel electrophoresis, all 10 ends were shaved and cloned into a Smal section of pGEM 4Z. Clones were obtained in both orientations and the clone that contained the intron region in orientation: Hindlll, Sphl, Pstl, Salí, Xbal, BamHI (initial Smal) from intron 3 'end, intron sequences, 5' end of intron , (Smal Initiator), Kpnl, Sstl, EcoRI was chosen.

This clone was named pPAL1914 and was digested with BamHI and Sstl, and inserted into PAI1920 previously cut with BamHI and Sstl, creating the vector PAL1954.

To create a specific restorer vector, the

CDNA from clone 19 of cDNA (Figure 6) to the promoter region as follows: The cDNA clone was digested with EcoRV and Smal as shown in Figure 11. The EcoRV-Smal fragment was added to this cut vector from clone L19. The clones that were recovered that contained the reconstructed 5 'region of the promoter and the coding sequence, but carried coding sequences that lack the intron, most regions of

-135135 encoding is derived from the cDNA clone. This clone was digested with EcoRI, but made from klenow shaved ends and then re-digested with Smal: A DNA fragment representing the entire coding region and a portion of the promoter region was isolated and cloned to PAL1920 cut with Smal, leading to a reconstruction of the promoter region and the coding region devoid of an intron. This vector was named PAL1955.

In Figure 13, the production of clones containing the coding sequences of a protein 10 functionally related to the ricin A chain protein isolated from Ricinus communis is detailed. This was accomplished by first isolating a ricin homologous genomic clone from a Ricinus zanzabarenis DNA genomic library constructed in the lambda gt10 vector using standard protocols. The collection was classified with a DNA probe that corresponds to the N-terminal leader sequence of the ricin gene. The probe sequence was obtained from the published sequence of a Ricinus communis ricin gene (Halling, et al., Nucí. Acids Res 13: 8019-33, 1985). A genomic clone containing the leader sequence was isolated and a portion of the A chain was isolated and named RIC 1B. This clone contained the promoter region, the 5 'untranslated region, the N-terminal leader sequence, and the coding region that extends to amino acid 191 (11e) in the published sequence (Halling, et al., Nucí Acids Res 13: 8019-33, 1985). The difference between the published sequence and RIC 1B was 25 that of the published nucleotide sequence in the region of 1 le

-136136

191 and was: (1le 191 underlined) ACG AGA ATT CGG that code for amino acids: Arg 1 le Arg while in RIO 1B the nucleotide sequence is ACG AGA ATT CGG, which codes for the same amino acids Arg He Arg), the only difference being that the last Arg is encoded through CGG in RIC 1B while the published sequence is encoded by AGG. This individual nucleotide substitution has the effect of introducing an EcoRI site at 1 le 191. The RIC 1B clone therefore lacked amino acids present after 1 le 191 since the clone was isolated as a single EcoRI fragment. This truncated version of ricin was used for the construction of the N-terminal deletions of ricin A chain as follows: Clone RIC 1B was digested with Hincll and EcoRI. The fragment was cloned into the EcoRI-Smal pGEM 4Z cut. The resulting clone pPALRIB was digested with Pstl and BamHi. This cleavage clone was digested with Exo III nuclease, treated with S1 nuclease and the klenow fragment, and then relegated. The subclones were obtained and had several portions of the deleted 5 'region, and some of these deletions were sequenced. One deletion, called pPAL-Ridefin had most of the N-terminal leader DNA sequence removed and had 5 'Hindlll and Sphl sites to this region such that the DNA sequence was as follows: AAGCTT GCATGC GCA ACA TGG .. .. where the first six nucleotides code for a Hindlll site found in pGEM 4Z, the following six nucleotides code for the Sphl site in pGEM 4Z and the following three triplets code for the

-137137 amino acids -20, -19, -19 .... (Ala, The Trp ...) in the published sequence (Halling, et al., Nucí. Acids Res 13: 8019-33, 1985). Therefore, this subclone had an deletion that removed the first 15 amino acids from the ricin A chain leader sequences. To provide an ATG start codon in a favorable start context, the synthetic oligonucleotide was constructed and inserted into the subclone between the Hindlll and Sphl sites. This nucleotide had the sequence: 5'-ACGTGGATCCAAGATGACATG-3 '. The reconstructed con (pPAL Rictr) therefore had the DNA sequence (the restriction site for an introduced BamHI site is underlined, and the ATG start codon in bold) AACGT GGATCCAAG ATG ACA TGC GCA ACA TGG in region 5 'such that there is an ATG start codon in a favorable start context and a BamHI site for excision of the coding sequence. The pPAL Rictr clone was digested with EcoRI, the end was filled with klenow and dephosphorylated with alkaline phosphatase. The universal translational terminator purchased from Pharmacia-PL Biochemicals (Montreal, Canada) was added to the vector to provide a stop codon. The coding region for this clone was isolated through digestion with BamHI and pVUH, releasing the coding region and a small portion of pGEM 4Z, and this fragment can be cloned to the BamHI and Smal sites of transformation vectors. This DNA fragment codes for a version of a ricin A chain in which a Cterminal portion has been removed. It should be noted that a number of deletions C

-138-138 terminal and N-terminal ricin A chain has been tested in vitro for toxicity, and these reports have concluded that half of the N-terminal ricin A chain is sufficient for in vitro cytotoxicity (eg, see: Sudan et al., Nucí Acids Res, 17: 1717-32, 1989). In order to obtain a complete coding region for the ricin A chain, a synthetic version of the rest of the A chain was synthesized using the published sequence. This synthetic portion of the gene extended the DNA sequence to nucleotide 1182 in the published sequence and has EcoRI sites at both ends that allowed this EcoRI site fragment to bind at amino acid numbers 190-192. This reconstructed version of the gene also had a BamHI site after the stop codon, so that the DNA sequence of the gene at the 3 'end was as follows: (nucleotide 1181 are marked with *) CCT CCA * TAA GGATCC GAATTC which encodes for amino acids: Pro Pro termination, followed by BamHI and EcoRI, the EcoRI site is used for insertion of the synthetic portion of the gene, the BamHI site is used for excision of the entire ricin A chain sequence, since a BamHI site Synthetic is at the 5 'end of the coding region, too. The clone was named pPAL Riccom.

In Figure 14, the results of antisense RNA inhibition of beta-glucuronidase gene activity are shown in a histogram of GUS gene activity found in transgenic plants containing a sense GUS gene and

-139139 were re-transformed with a vector containing an antisense GUS gene. GUS activity levels were expressed as a percentage of the GUS activity found in GUs plus original TTR-48 plant. Tissue was obtained from young (Y), 5 medium (M) and old (O) leaves and the activity of GUs was analyzed spectrophotometrically. Lanes 1-5, samples from plants transformed from TTR-1 to TTR-5 (TTR48 / PAL1302), lane 6, samples from TTR-48 plant; lane 7, samples of TTR-88 (TTR48 / pVU1011), and lane 8, samples of non-transformed tobacco.

Figure 20 illustrates hybrid seed production using a variation of the binary cryptotoxicity method and a herbicide resistance gene. Genes 1 and 2 can be the lamS and lamH genes. The herbicide resistance gene can be any resistance gene against any herbicide. Origin A is produced with the three genes G1, G2 and H being linked by virtue of the introduction of a recombinant DNA molecule, through the transformation of a plant cell. It is maintained by crossing with the untransformed isogenic strain and spraying herbicide resistance. Origin B is homozygous for a male sterility restoration gene such as antisense for gene 1 and / or gene 2 and for a linked herbicidal resistance gene. The resulting hybrid seed is essentially 100% herbicide resistant and restored with respect to male fertility.

Figure 21 illustrates the production of hybrid seeds 25 with origin A being produced and maintained as shown in

-140140

Figure 20. Origin B is homozygous for herbicide resistance. The hybrid is essentially 100% herbicide resistant.

Figure 22a illustrates the use of the binary cryptotoxicity method for the production of an herbicide resistant origin A of male sterility. Gene 1 and gene 2 come from different transformed lines, each homozygous for the recombinant DNA molecule, with either G1 or G2, or both, being linked to a herbicidal resistance gene. The resulting Origin A is essentially 100% male sterile with G1 and G2 on 10 different chromosomes. Origin A is resistant to herbicide.

Figure 22b illustrates the method used to maintain origin A shown in Figure 22a.

Figure 22c illustrates the crossing of origin A shown in Figure 22a with origin B. Origin B is the result of a transformation where the restorative gene (antisense sequences, or lexA, for example) and herbicide resistance are bind into the recombinant DNA molecule. The result of crossing origin A and origin B is essentially a hybrid that is restored with respect to male fertility and is herbicide resistant. If restoration to male fertility is not desired, Origin B is transformed with a recombinant DNA molecule that contains the herbicide resistance gene, but not the gene responsible for fertility restoration.

Example 1

-141141

In this example, antisense RNA was used specifically for the inhibition of gene activity in plants. A tobacco plant expresses the Beta-glucuronidase gene under the control of the CaMV 35S promoter was produced through transformation of a non-transformed control tobacco culture, N. tabaccum, cv. Delgold. To achieve this, tobacco leaves less than 20.32 cm (8 inches) in length were surface sterilized through exposure to ethanol for 5-6 seconds, then subsequent exposure to 1% sodium hypochlorite for a few minutes, usually 5-10 minutes, or until the cutting edge of the petiole turned white, then rinsed several times in sterile distilled water. Leaf segments of approximately 0.5 to 1.0 square centimeters were co-cultured for two days with Aqrobacterium tumefaciens GV 3101 carrying the Ti plasmid pMP90 to promote vir in trans functions (described by Koncz, C. and Schell J., 1986, Mol. Gen. Genet. 204: 383-396) carrying the binary vector pBI121.1 in an activation induction medium. This vector is a derivative of Bin19 which contains the GUS gene activated through the CaMV 35S promoter and terminated through us and is available from Clonetech Laboratories, Palo Alto, CA., USA. The transformed tobacco cells were selected in an activation induction medium containing 0.8% agar, MS salts, vitamins B5, 3% sucrose, 1 mg per liter of benzyladenine, 0.1 mg per liter of alfanaftalenacetic acid, (NAA) 300 pg / ml of kanamycin and 500 pg / ml of

-142142 carbenicillin (essentially as described by Horsch et al., 1985, Science, 227: 1229-31). The regenerated shoots were then transferred to a root induction medium consisting of a B5 medium with 2% sucrose, 500 pg / ml carbenicillin, and 0.5 mg / l 5 each of NAA and indolacetic acid (IAA). Following selection in kanamycin, a tobacco transformant that exhibited relatively high constitutive levels of GUS activity and contained an individual unwilling insert of the 35S CaMV-GUS-noster promoter construct was selected. This plant (TTR10 48, GUS +) was then transformed again with a binary vector

PAL1302, which contains an antisense GUS gene, the construction of which is described in Figure 1. In experiments involving transformation of the TTR-48 tobacco plant with PAL1302, the shoot induction medium contained 20 pg / ml of hygromycin and 300 15 pg / ml kanamycin to ensure selection of plants containing both sense and antisense GUS constructs. Transformants developed to maturity and self-pollinated in the greenhouse.

The leaves of tobacco plants resulting from the re-transformation of TTR-48 (GUS +) with the antisense GUS vector

PAL1302 were analyzed for GUS activity. GUS activity in leaf extracts was analyzed spectrophotometrically. About 0.5 g of the leaf tissue was ground with Polytron in 2 ml of GUs extraction pH regulator (50 mM NaPO4 pH 7.0, 1 25 mM EDTA, 0.1% Triton-X-100, 10 mM B- mercaptoethanol, 1

-143143 mM p-nitrophenylglucuronide, 100 pg / ml bovine serum albumin 0.02% sodium azide) was incubated at 30 ° C for 6 hours and the reaction was stopped through the addition of 0.4 ml 2.5 M 2-amino-2-methyl-1,3-propanediol. The amount of p-nitrophenol produced was calculated by measuring the absorbance at 415 nm. The reaction was stopped containing an identical amount of leaf extracts which was used as a template. Relative enzyme activities in the extracts were calculated and expressed as nanomoles of p-nitrophenol produced per mg of protein per minute. Ten plants were classified and all exhibited a large reduction in GUS activity levels relative to those observed in TTR-48. Five plants (TT-1 to TTR-5) were chosen for a detailed analysis of the effects of antisense RNA inhibition of GUS gene activity. Retransformed tobacco plants TTR-1 to TTR-5 (Figure 14, lanes 1-5) all showed a considerable complete reduction (lane 4) in GUS activity without considering the stage of development of the examined leaves. Comparison of the highest level of GUS activity observed in GUS + original TTR-48 plant (Figure 14, lane 6) with the highest level found in any of the plants transformed with AL1302 (lanes 1-5) shows that the reduction GUS activity was at least 90%. The GUS activity levels found in the control TTR-88 plant (Figure 14, lane 7) were similar to those of the original GUS + plant (lane 6), indicating that the retransformation and regeneration procedure

-144144 suffered by TTR-48 was not responsible for the decline in GUS activity observed in TTR-1 to TTR-5. Western blot analyzes of total protein extracts obtained from young leaves were performed. For leaf protein extraction, approximately 100 mg of tissue was ground in a 1.5 ml Eppendorf tube containing 0.7 ml of the GUS extraction pH regulator described above. An equal volume of SDS PAGE 2X Sample Loading pH Regulator (1.3M Tris-CI pH 6.8, 2% B-mercaptoethanol), 50% Glycerol, 5mM EDTA, 0.1% Bromophenol Blue) is added and samples were incubated at room temperature for 15 minutes. The extract was centrifuged at 12,000 rpm for 2 minutes and the supernatant was frozen at -80 ° C until used. Proteins were resolved on 10% SDS PAGE gels and immediately transferred electrophoretically on a nitrocellulose filter. The GUS protein was then detected in the gels using a rabbit anti-B-glucuronidase antibody obtained from Clonetech Laboratories and an anti-rabbit IgG alkaline phosphatase conjugate kit (Promega Biotech, Madison Wl, USA) according to the instructions of the maker. Equal amounts of protein were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE), transferred to nitrocellulose and related to specific antibodies to the GUS protein. The amount of the GUS protein detected in western staining correlated well with the GUS activity found in the leaves of all tobacco plants examined.

-145145 if they exhibited high levels of activity such as TTR-48 or there was no discernible activity in TTR-3 (Figure 14, lane 3). The reduction in GUS activity in TTR-1 to TTR-5 can thus be directly attributed to lower amounts of GUS enzyme within these plants. Southern staining analyzes were performed to confirm that the sense GUS gene is still present and intact in TTR-1 through TTR-5 and to verify that the antisense GUS construct has been correctly integrated into the DNA of these plants. The original sense GUS gene arrangement in TTR-48 was found to be unaffected by transformation with the vector containing the antisense gene, and it was further found that among the plants selected for this analysis, between 1-3 copies were inserted of the antisense gene in the plant genome, which contained the sense gene. It was also determined that an individual inserted antisense gene can lead to a near total reduction in sense gene activity. Northern staining analyzes on full-sheet RNA were conducted to determine whether the reduction in the amounts of the GUS enzyme observed in TTR-1 or TTR-5 correlated with their GUS mRNA levels. Total RNA was prepared from tobacco leaves by milling 0.5 g of tissue in 2 ml of extraction pH regulator (6 M guanidine hydrochloride, 0.1 M NaAc pH 6.0, 1.0% b-mercaptoethanol) for 10-30 seconds. using a Polytron. The mixture was then centrifuged at 5,000 rpm for 3 minutes', the supernatant was layered in

-146146 an equal volume of 5.7 M CsCI in 0.1 M NaCI, 10 mM TrisCl pH 7.5, 1 mM EDTA and again centrifuged at 35,000 rpm for 16 hours at 15 ° C. The resulting RNA pellet was resuspended in 0.1 M NaAc pH 6.0, 0.1% SDS and extracted with an equal volume of phenol-chloroform (50:50 v / v). The aqueous phase was adjusted to 0.3 M NaAc and the RNA was precipitated with 2 volumes of ethanol. After centrifugation, the pellet was washed in 70% ethanol and resuspended in sterile distilled water. The RNA samples were resolved in the presence of methylmercury hydroxide in 1.3% agarose gels and transferred to a nylon membrane. Membranes were applied to probes with (32p) UTP-labeled sense or antisense GUS RNA transcripts. These transcripts were made from pGEM-GUS, a plasmid obtained by inserting the BamHI-Sstl fragment from pBI221.1 (which contains the entire GUS coding sequence) to the BamHI-Sstl sites from pGEM-4Z. Probes were made by cutting pGEM-GUS with EcoRI and then using T7 RNA polymerase to provide a transcript, which can hybridize to antisense GUs RNA or through Hindlll digestion and transcription with SP6 RNA polymerase giving antisense transcription which can hybridize to the sense GUS mRNA, only. A radiolabelled sense-specific GUS RNA probe demonstrated that the GUS mRNA levels found in TTR-1 through TTR-5 were considerably lower than those observed in TTR-48, the

-147147 original GUS + plant, as stated, unprocessed tobacco does not possess the GUS transcript. Sense mRNA levels correlate well with the amount of GUS protein and the activity observed in these plants. Northern staining analysis using an antisense-specific GUS RNA probe demonstrated the presence of antisense GUS transcription in the transformed plants. The reduced amounts of the GUS protein and the GUS activity observed in TTR-1 through TTR-5 can thus be attributed to the low levels of GUS mRNA found in these plants. Low levels of GUS mRNA were always associated with the presence of antisense GUS RNA. These results clearly indicate that an objective sense gene can be successfully inhibited using antisense RNA.

Example 1A

This example describes a method to isolate microspores in order to obtain genes that are essential to the formation and / or function of pollen. Microspores were conveniently isolated through manual shoot dissection to remove anthers that are subsequently interrupted by moderate grinding in a mortar and grinder in 10% sucrose. The extract was then filtered through a 44 pm nylon mesh and the microspores were collected by centrifugation at 3,000 xg for one minute. The pellet-shaped microspores were resuspended in 10% sucrose, filtered, and

-148148 formed into pellets as before. Other methods can also be used to isolate microspores.

Tissues other than microspores can be disrupted through a variety of methods and the broken tissue used for RNA isolation. The tissue should be broken using a motor-driven homogenizer with 10 ml of a 6M solution of guanidinium HCI, 0.1M Na acetate, pH 6.0, 0.1M beta-mercaptoethanol per gram of tissue. The homogenate was cleared by centrifugation at 5,000 xg and the supernatant was layered over a 6M solution of CsCI in Tris-EDTA pH regulator (TE pH regulator). Centrifugation at 100,000 xg for 12-20 hours at 15 ° C was used to pellet the RNA which was subsequently re-suspended in water and re-precipitated in the presence of 0.3 M Na acetate and two volumes of ethanol . RNA was recovered through centrifugation and resuspended in water. The RNA obtained through such a method can be fractionated through oligo-dT cellulose chromatography to separate the polyadenylated mRNA from the volume of the non-polyadenylated RNA. Microspore RNA was isolated using a tight-fitting motor-driven glass homogenizer to break up the microspores. Homogenization of 300 ul of packed microspores was conducted in 1 ml of 6 M of guanidinium HCI, 0.1 M of Na acetate, pH 6.0, and 0.1 M of beta-mercaptoethanol. The homogenate was centrifuged at 5000 xg and the clarified supernatant was

-149149 layered over a 6M CsCI solution in TE pH buffer. Overnight centrifugation at 100,000 xg was used to pellet the RNA, which was subsequently re-suspended in water and re-precipitated in the presence of 0.3 M Na acetate and two volumes of ethanol. Other RNA extraction methods can be used to obtain the RNA from the described tissues. The normal methodology was used using oligo-dT

<td>cellulose to obtain mRNA total RNA preparations.</td><td>polyadenylated to</td><td>depart</td><td>of this</td>

<td>Example 2</td><td></td><td></td><td></td>

<td>This example is</td><td>refers to the use</td><td>of</td><td>promoters</td>

<td>constitutive to regulate the</td><td>expression of</td><td>RNA</td><td>antisense</td>

<td colspan="2">specific in pollen. The CaMV 35S promoter of</td><td colspan="2">pB1 221 was isolated</td>

as a fragment of Hindlll-Xbal and cloned into PAL1001 previously cut with Hindlll and Xbal. This produced a vector that had the CaMV 35S promoter bound to the nos and between the promoter and the terminator were unique sites for: Xbal, BamHI, Smal, Kpnl and Sstl. This vector was named PAL1007. PAL1007 was digested with BamHI and a 2.4 kb BamHI fragment containing a coding region for clone L16 was added to this digested vector in the antisense orientation. This vector was named PAL1305 and was used to transform Brassica napus. The transformation was performed using the method described in Moloney, MM, et al., (Plant Cell Reports 81989) 8: 238-42) or, the transformation was performed with layers

-150 150 Stem epidermal sterilized on the surface. For this procedure, cells from B. napus L. ssp. oleífera cv. Westar were grown in 'Promix' mixed with 2 g / l of Nutricoate slow-release fertilizer in 20.32 cm (8 inch) containers. Plants were grown in the greenhouse during a photoperiod of 16 (using natural and artificial light). For co-cultivation and regeneration experiments, stem sections from the internodes of three upper stems of plants approximately 1.5 months old (i.e. those with elongated flower spikes and several open flowers) were used. Intact stem sections were surface sterilized for 30 seconds in 70% ethanol and 10 minutes in 1% sodium hypochlorite followed by three rinses in sterile distilled water. For transformation, Aqrobacterium tumefaciens GV 3101 carrying the Ti plasmid pMP90 to provide vir functions in trans and the PAL1110 binary vector was grown in YEP medium (which consists of 10 gm per liter of yeast extract, 10 g per liter of bactopeptone and 5 g per liter of NaCI, pH 7.0, containing 100 ug per ml of kanamycin for the selection of bacterial cells containing the binary vectors). The cells grew from one to two days at 28 ° C. Cells were harvested by centrifugation and resuspended at a density of approximately 10⁶ - 10? cells by me in the liquid EL, which consists of micro and macro MS nutrients and vitamins B5 containing 40 mg / l FeNa-EDTA (obtained from BDH-chemicals) and

-151151

3% sucrose, 10 mg / l benzyladenine and 0.5 mg / l alfanaftalenacetide (NAA) and 18.8 mM KNO³ plus 20.6 mM NH4NO³. The medium was solidified with 0.8% agar (Sigma) when EL medium was used for solid media plates.

The cell suspension was emptied into the bottom of a sterile petri dish and sterilized stems were dissected directly into the bacterial suspensions. The segments were longitudinally sectioned into medium segments and cut into sections of approximately 5 mm. The dissected segments were placed on filter paper discs on a solid EL medium for 3 days of co-cultivation under fluorescent light conditions (60 microinsteins / m2 / sec2) at 25 ° C. After a 2-3 day co-culture, the explants were transferred to the solid EL medium containing 500 ug / ml of carbenicillin, and 100 ug / ml of bekanamycin (Sigma). Sprouts formed in 4-8 weeks, sections were transferred to fresh solid EL medium with carbinicillin and bekanamycin every 3-4 weeks. Sprouts that formed and did not blanch were excised and rooted in PDR medium (B5-containing 2% sucrose and 0.5 mg / l each of NAA and IAA). In some cases, the non-regenerating calluses developed in the selective medium were separated from explants and transferred to the fresh medium to stimulate regeneration. The transformed plants were placed in a humidity chamber, and after two to four weeks they were transferred to the greenhouse. The plants were developed under a

-152152 16 hour photoperiod and allowed to flower. Clonal propagation was used to increase plant lines as well as manual crossing and selection of cross-plant seeding in the medium containing kanamycin. This medium consisted of 0.8% agar, one tenth of the MS salts and 100 ug per ml of bekanamycin without any sucrose in the medium. Sterilized surface seeds were used. The seeds were sterilized on the surface by rinsing in 70% methanol for a few seconds, soaking in 1% sodium hypochlorite for 15 minutes, followed by three times rinsing of sterile distilled water. The seeds were placed on the surface of the agar in sterile dishes and left to empty. Plants which do not carry the kanamycin gene bound to the bleached and dry antisense gene, while those which carry the antisense gene in green and were subsequently transferred to the soil and allowed to flower.

Example 3

In this example, another pollen specific coding region was used with the vector PAL1007. In this case, a 1.3 Kb Kindlll fragment of clone L19 was isolated, its ends shaved, and cloned to the Smal site of pGEM 4Z. This subclone was named pPAL1914 and was then digested with Xbal and Sstl. This fragment was added to the Xbal-Sstl cut of PAL1007, giving rise to a vector named PAL1307. This vector contains the CaMV 35S promoter fused to a region of

-153153 encoding from clone L19 in an antisense orientation. This vector was used to transform Brassica napus as detailed in Example 34.

Example 4

This example relates to the use of promoters capable of being induced to regulate the expression of specific antisense RNA in pollen. In this example, the 1.2 Kb Hindlll PstI fragment of the D. melanoqaster 70 KD 10 heat shock protein promoter was isolated from subclone pPW 229 (Holmgren, R. et al., 1979, Cell 18: 1359-1370 ) and cloned Hindlll-PstI cut to pGEM 4Z. The heat shock promoter was excised as a Hindlll-Smal fragment and cloned into Hindlll-Smal cut to PAL1001. This produced a vector (PAL1009) containing a heat shock promoter followed by a portion of the i polylinker and nos ter. The Smal site was used to clone the 1.3 KB Hindlll fragment from clone L19 and then shaved ends of this fragment. The clone containing this fragment in the antisense orientation relative to the heat shock promoter was named PAL1403. This vector was used to transform Brassica napus as shown in Example 34. In addition, since there is a single EcoRI site at the nos-end in this construct, a selection marker gene was added to this construct, beta-glucuronidase. of enzyme driven to 25 through the CaMV 35S promoter using the unique EcoRI site for

-154154 insertion of this gene. This vector, which is the same as PAL1403, except that it now contains a convenient gene for transformation classification was named PAL1408 and was used for the transformation of Brassica napus as described in Example 34.

Example 5

This example relates to the use of a promoter essential for the development and / or function of pollen, to specifically express specific antisense RNA in pollen in pollen cells. For this, vector PAL1107 was used for the production of antisense RNA from the 4F cDNA clone described in Figure 4. To construct this antisense vector, the EcoRi fragment of the 4F cDNA clone was isolated, made from shaved ends with klenow and cloned to the Smal site of PAL1107. A vector that contained the cDNA clone in the antisense orientation (as determined by restriction enzyme analysis) was chosen. This vector was named PAL11074F and was used to transform Brassica napus as described in Example 34.

Example 6

For this example, vector PAL1107 was used for the production of antisense RNA from the 2.4 Kb BamHI fragment of clone L16 described in Example 2. To construct this antisense vector, a 2.4 Kb BamHI fragment was isolated to

-155155 from clone L16 and cloned into the BamHI site of PAL1107. A vector that contained this fragment in the antisense orientation (as determined by restriction enzyme analysis) was chosen. This vector was named PAL1007-16CRAS and was used to transform 5 Brassica napus as described in Example 34.

Example 7

For this example, vector PAL1107 was used for the production of antisense RNA from the 10 1.3 Kb Hindlll fragment of clone L19 described in Example 3. To this construction of this antisense vector, the Hindlll fragment was isolated from 1.3 Kb from the L19 clone and its shaved ends were made and cloned to the Smal site of PAL1107. A vector that contained this fragment in the antisense orientation (to be determined by restriction enzyme analysis) was chosen. This vector was named

PAL1107-19CRAS and was used to transform Brassica napus as described in Example 34.

Example 8

For this example, vector PAL1107 was used for the production of antisense RNA from the cDNA clone relative to the L10 clone described in Figure 5. To construct this antisense vector, the EcoRI fragment of the cDNA clone was isolated, made its ends shaved with klenow and isolated to the Smal site of PAL1107. A vector was chosen that contained a cDNA clone in the

-156156 antisense orientation (as determined through restriction enzyme analysis). This vector was named PAL110710G and was used to transform Brassica napus as described in Example 34.

Example 9

For this, the vector PAL1107 was used for the production of antisense RNA from the cDNA clone related to the L19 clone described in Figure 6. To construct this antisense vector, the EcoRI fragment from the cDNA clone was isolated, Their shaved ends were made with klenow and cloned to the Smal site of PAL1107. A vector was chosen that contained the cDNA clone in the antisense direction (as determined through restriction enzyme analysis). This vector was named PAL110719 and was used to transform Brassica napus as described in Example 34.

Example 10

In this example, the promoter essential for the development and / or function of pollen in vector PAL1421 was used for the production of antisense RNA using the cDNA clone homologous to the gene contained in L10. This was accomplished by excising the cDNA clone from the EcoRI cloning vector and making this fragment have klenow shaved ends. This shaved-ended cDNA fragment was cloned into the Smal site of PAL1421. The

-157157 clones were recovered and contained the cDNA insert in both orientations, and one was chosen containing the insert in the antisense orientation relative to the PAL1421 promoter, resulting in the formation of a PAL1492 binary transformation vector. PAL1492 was used to transform Brassica napus as described in Example 34.

Example 11

In this example, the promoter essential for the development and / or function of pollen in vector PAL1121 was used for the production of antisense RNA using the cDNA clone homologous to the gene contained in L10. This was accomplished by excising the cDNA clone from the EcoRI cloning vector and making this fragment have klenow shaved ends. This shaved-ended cDNA fragment was cloned into the Smal site of PAL1121. The clones were recovered and contained the cDNA insert in both orientations, and one was chosen containing the insert in the antisense orientation relative to the PAL1121 promoter, resulting in the formation of a binary transformation vector PAL1110. PAL1110 was used to transform Brassica napus as described in Example 34.

Example 12

In this example, the promoter essential for the development and / or function of pollen from clone 19 was used for the expression of the

-158158

Specific antisense RNA in pollen. The transformation vector PAL1920 was digested with Smal. To the Smal site, the DNA fragment corresponding to the cDNA clone homologous to clone L10 was added through digestion of the vector containing this clone with EcoRI and making its ends shaved. This shaved-ended cDNA fragment was cloned into the Smal site of PAL1920. The clones were recovered and contained the cDNA insert in both orientations and one was chosen to contain the insert in the antisense orientation relative to the pPAL1920 promoter, resulting in the formation of the PAL1921 binary transformation vector. PAL1921 was used to transform Brassica napus as described in Example 34.

Example 1 3

<td>In</td><td>this</td><td>example,</td><td>I know</td><td>did</td><td>a</td><td>RNA</td><td>antisense</td>

<td colspan="2">specifically to the</td><td>intron region</td><td>of the</td><td colspan="2">clone L19 and</td><td>a gene</td><td>restorer</td>

<td>specific it</td><td>did,</td><td colspan="2">which lacks</td><td>of the</td><td>intron</td><td colspan="2">of clone L19, the</td>

<td>building</td><td colspan="2">of these vectors</td><td>I know</td><td colspan="2">presents</td><td>in the</td><td>Figure 12.</td>

<td>PAL1954 se</td><td>used</td><td colspan="2">to transform</td><td colspan="2">Brassica</td><td>napus</td><td>as in the</td>

Example 34 to make a male sterility line. To create a specific restorative plant line, vector PAL1955 was used to transform Brassica napus as in Example 34.

Example 14

-159159

In this example, the highly active promoter fragment from clone HP101 was used to synthesize antisense RNA to the intron region of clone 19. For this example, the subclone containing the intron region of clone 19 detailed in Figure 5 12, pPAL1914 was digested with Hindlll and BamHI. The promoter fragment from clone pPALHP101 was added to this construct as a Hindlll-BamHI fragment giving rise to clones containing the promoter fragment in the antisense orientation relative to the intron. This clone contained the promoter in an 10 orientation so that transcription of the promoter could cause the production of an RNA, a portion of which could contain antisense RNA homologous to the intron region of clone L19. This clone was named pPAL19HP. The pPAL19HP clone was digested with Hindlll and Sstl, and the vector PAL1001 was cloned using the Hindlll and Sst 15 sites of PAL1001, creating PALHP19. PALHP19 was used to transform Brassica napus as in example 34. It should be noted that the vector PAL1955 (see example 13 and Figure 12) can be used for fertility restoration in plants bearing PALHP19.

Example 1 5

The promoter essential for the development and / or function of the pollen contained in the subclone pPAL0420 was used for the construction of an antisense RNA gene under the control of a promoter essential for the development and / or function of the pollen as

-160160 follows: A DNA fragment encoding an Arabadopsis polyubiquitin was isolated from a plasmid, which contains a polyubiquitin gene that has 5 copies of the ubiquitin monomeric protein obtained from University of Wisconsin, Madison, Wl, USA and is described in Burke et al., Molecular and General Genetics, in the press, The BamHI-BglII fragment containing 3-5 copies of the polyubiquitin gene was isolated and this fragment was inserted into pPAL0420 using the individual BamHI site of the PAL0420 polylinker. This gave rise to a plasmid containing a promoter essential for the development and / or function of pollen from clone L4 followed by a DNA fragment containing the ubiquitin coding sequences in the antisense orientation followed by the polyadenylation signal nos . This antisense promoter / gene construct was excised from pPAL0420 by digestion with EcoRI. The EcoRI fragment containing the promoter antisense gene was inserted into the EcoRI site of the Bin19 polylinker, resulting in the formation of a binary transformation vector named PAL1479. PAL1479 was used to transform Brassica napus as described in Example 34.

Example 16

Vector PAL1479 was used to transform tobacco as described in Example 34.

-161161

Example 17

The promoter essential for the development and / or function of the pollen contained in subclone pPAL0420 was used for the construction of a specific antisense RNA gene in pollen, as follows: A DNA fragment encoding pine actin was obtained from J. Kenny-Byrne, Petawawa Cañada. Two clones were obtained, Pac 1-A and Pac 2, clone Pac 1-A is described in: Canadian Journal of Forestry Research (1988) 18: 1592-1602, and the second clone Pac 2 (the sequence being closely homologous to 10 that in Pac 1-A and the nucleotide sequence of which having been submitted for publication). An Sphl fragment was isolated from Pac 2 containing the complete pine actin coding sequence. This fragment has contains a small amount of the 5 'and 3' uncoded region. This 15 Sphl fragment was cloned into the unique Sphl site of pGEM 4Z. From this subclone, an Xbal fragment containing only the coding region was isolated and this Xbal fragment was cloned to the unique EcoRI site of pGEM 4Z. The clone that was chosen was oriented so that the 5 'end of the gene was next to the BamHI site in the polylinker and the 3' end of the gene was next to the Sali site in the polylinker. This plasmid was named pPAL PAC. The actin coding region was isolated from pPAL PAC by digestion with BamHI and Sali. This BamHI-Sali fragment was cloned into pPAL0420 using the BamHI and Sali sites contained within the polylinker of pPAL0420. This gave

-162162 emergence to a plasmid containing the promoter essential for the development and / or function of pollen from clone L4 followed by a DNA fragment containing the actin coding sequence in the antisense orientation followed by the polyadenylation signal of we ter. This antisense promoter / gene construct was excised from pPAL0420 through digestion with EcoRI. The EcoRI fragment containing the antisense promoter gene was inserted into the EcoRI site of the Bin19 polylinker, resulting in the formation of a binary transformation vector PAL1498. PAL1498 was used to transform Brassica napus as described in Example 34.

Example 18

The vector PAL1498 was used to transform tobacco plants as presented in Example 34.

Example 1 9

In this example, tobacco that was previously transformed to hygromycin resistance was transformed with an antisense gene that specifically blocks hygromycin resistance in pollen cells by virtue of the fact that the antisense gene is under the control of an essential promoter for the development and / or function of pollen in vector PAL1106. The transformed tobacco that was resistant to neomycin was obtained using the vector PAL1302, which is described in Example 1. The

-163163 selection of the hygromycin resistant tobacco plant cells was through a culture with PAL1302 and the selection of 50 ug per ml of hygromycin. Southern staining analysis demonstrated the presence of 5-6 copies of the same sense hygromycin phosphotransferase gene in a plant. This plant, named as TTR-122, was re-transformed with a vector named PAL1107A. PAL1107A is vector PAL1106 to which was added the 0.8 Kb BamHI hygromycin phosphotransferase fragment isolated from PAL1302 and inserted into PAL1106 in the antisense orientation relative to the promoter essential for pollen development and / or function in PAL1106. The resulting vector was named PAL1107A. The plants obtained from this transformation were resistant to both hygromycin and kanamycin in the leaf tissue and were shown through southern staining analysis that contained the antisense gene. These plants were left to grow in the greenhouse and were self-fertilized. The clonal propagation of these plants was used as a preliminary increment of individual plants, and these clonally propagated plants were used for the preparation of a male sterility plant line. For example, a plant that contained an individual copy of the antisense gene and was derived from the TTR-122 transformation, was planted in a sand-soil mixture, and allowed to grow in a greenhouse. This plant was named as TTR-203. Measurement of the hygromycin phosphotransferase activity of this plant

-164164 demonstrated high activity on the leaves, petals, stigma, pistil, and anther walls, but very low levels of pollen. TTR-122 showed high levels of hygromycin phosphotransferase activity in leaves, petals, stigma and pistil and front walls and in pollen. This demonstrated that the antisense gene was effective in blocking the expression of the sense gene only in pollen. Northern staining analysis confirmed the presence of antisense transcription specifically in TTR-203 pollen and also demonstrated low levels of the sense mRNA gene. When the first flower buds appeared, TTR-203 was watered three times a week with a hygromycin solution (250 ug per ml), thoroughly saturating the sand-soil mixture. This irrigation was continued for approximately four weeks, and the period in which the main flowering occurred. The flowers produced during this time in plants containing the antisense gene were male sterile. Anther and pollen formation was inhibited and mature pollen development failed. Female fertility was not affected through this treatment since manual pollination could be used to pollinate the female portion of the male sterility flowers. Pollen that formed from a hygromycin-resistant plant was used for the production of hybrid seeds. Watering of the plants with hygromycin was stopped, and normal watering was resumed. The flowers that formed on the plants after hygromycin irrigation stopped were male fertility and self-seeded.

-165165

Example 20

The pollen specific vector PAI1107 was used for the production of male sterility plants. The TTR122 plant described above was retransformed with a vector named PAL1107HYGAS. The transformation was conducted as described in Example 34. PAL1107HYGAS is the vector PAL1107 to which was added the 0.8 Kb BamHI hygromycin phosphotransferase fragment isolated from PAL1302 and inserted into PAL1107 in the antisense orientation relative to the promoter essential for the development and / or function of pollen in PAL1107. The plants obtained from this transformation were resistant to both hygromycin and kanamycin in the leaf tissue and showed through the southern stain that they contain the antisense gene. These plants were left to grow in the greenhouse and were self-fertilized. The clonal propagation of these plants was used as a preliminary increment of individual plants, and these clonally propagated plants were used for the production of male sterility plant lines as in Example 19.

Example 21

For the production of male sterility plants, vector PAL1419 was used to transform tobacco as presented in Example 34. Vector PAL1419 contains the promoter essential for the development and / or function of pollen from clone L4

-166166 by controlling the expression of the NPT II gene oriented in the antisense orientation in relation to the promoter essential for the development and / or function of pollen. This vector also contains a constitutive version of the NPT II gene in the sense orientation driven by the nos promoter. The vector, therefore, can confer kanamycin resistance in all plant cells, except for pollen cells, where the expression of the sense gene is inhibited through the expression of the antisense gene, which is specifically expressed in pollen. The tobacco plants were obtained following the transformation of the PAL1419 vector. These plants were rooted and allowed to set in bloom. The plants were watered with kanamycin while they were blooming.

Example 22

The PAL1419 vector was used to transform Brassica napus as presented in Example 34.

Example 23

The vector PAL1419 was used for the transformation of petunia leaf discs.

Example 24

In this example, the complete ricin A chain (pPAL Riccom) described in Figure 13, was inserted into the PAL1420 vector as the BamHI fragment. Vectors were recovered which

-167167 contained the ricin gene in both the sense and antisense orientation. The vector that contained the ricin A chain in the sense orientation was recovered and named PAL1420RIC. A vector that contained the gene in the antisense direction was 5 named PAL1420RICAS. PAL1420RIC and PAL1420RICAS were used for the transformation of Brassica napus as described in Example 34, giving rise to plants that carry either the sense or antisense copy of the ricin A chain gene under the control of the promoter essential for development and / or pollen function 10 in PAL1420.

Example 25

In this example, the truncated version of the ricin A chain (pPAL Rictr) described in Figure 13 was inserted into vector PAL1420 as a BamHI fragment. Vectors were recovered and contained the truncated ricin gene in both the sense and antisense orientation. A vector that contained the ricin A chain in the sense orientation was recovered and named PAL1420tRIC. A vector that contained the 20 gene in the antisense orientation was named PAL1420tRICAS.

PAL1420tRlc and PAL1420tRICAS were used for the transformation of Brassica napus as in example 34, giving rise to plants that carry either sense or antisense copies of the truncated ricin chain gene A under the control of the promoter essential for development and / or function in PAL1420.

-1684

168

Example 26

PAL1420RIC and PAL1420RICAS were used for the transformation of tobacco plants as in Example 34, giving rise to plants bearing either the sense or antisense position the sense copy of the ricin A chain gene under the control of the essential promoter for the development and / or function of pollen in PAL1420.

Example 27

The PAL1420tRIC and PAL1420tRICAS vectors were used for the transformation of tobacco plants as in Example 34, giving rise to plants that carry either the sense or antisense copy of the ricin chain gene to truncated under the control of the promoter essential for development and / or function of pollen 15 in PAL1420.

Example 28

For this example, vector PAL1423 was used to express the polylysine gene contained in the pPAL pLys subclone. The coding region for pPLA pLys was isolated through BamHI digestion and cloned in the sense orientation in BamHI cut PAL1423, giving rise to PAL1487. PAL 1487 was used to transform tobacco as in Example 34.

-169169

Example 29

For this example, the vector PAL1920 was used to express the polylysine gene contained in the pPAL subclone pLys. The coding region for pPAL pLys was isolated through BamHI digestion and cloned in the PAL1920 sense orientation of the BamHI slice, giving rise to PAL1987. PAL1987 was used to transform tobacco as in Example 34.

Example 30

In this example, a promoter essential for pollen development and / or function was used to synthesize a protein molecule that is destructive to cell function and development, primarily protease trypsin. The cDNA sequence encoding trypsin has been described by Stevenson et al., Nucí Acids Res, 1986, 14: 8307-30. The pMPt9 from the cDNA clone was obtained and used for the production of a modified trypsin molecule where the N-terminal amino acid residues were removed to give a protein that consisted only of the active protease form of trypsin and differed in form. mature in that a methionine residue existed at the N-terminal position of the mature protein, replacing the isoleucine found at the position in the active mature protein. This was accomplished through digestion with pMPt9 with Fokl and PstI, and retrieving a fragment spanning nucleotides 81 to 835, nucleotides after 835 being G: C-terminus used for cDNA cloning, and the

-170170 treating this fragment with klenow and cloning to Smal of shaved ends cut from M13mp19RF, and isolating a single stranded structure phage clone that was used for site specific mutagenesis to change the 5 isoleucine codon in the nucleotides 84-86 in the published sequence from ATT to ATG, introducing an initiation codon where the isoleucine codon was presented. The recovered mutated gene was excised with Salí and Sstl, and inserted into PAL1421 and named PAL1456. The PAL1456 vector was used to transform tobacco 10 as in Example 34. It should be noted that a restorative gene can be made by inserting the trypsin inhibitor in an analogous way using an L4 clone derived from the promoter and transformed to the line of origin male. Expression of the trypsin inhibitor in the hybrid will specifically block the activity of the trypsin enzyme. A number of cDNA sequences can be found and

<td>Genomic DNA for soy</td><td>and</td><td colspan="2">other inhibitors</td><td>of</td><td>trypsin, by</td>

<td>example see: Jofuku, KD,</td><td>and</td><td>goldberg,</td><td>RB, The</td><td colspan="2">Plant Cell 1989,</td>

<td> 1:1079-1093.</td><td></td><td></td><td></td><td></td><td></td>

<td>20 Example 31</td><td></td><td></td><td></td><td></td><td></td>

<td>In this example,</td><td>, I know</td><td>uses a</td><td>promoter</td><td colspan="2">essential for the</td>

<td>development and / or function of</td><td colspan="2">pollen for</td><td>synthesize</td><td>the</td><td>lamH enzyme</td>

specifically in pollen cells. The enzyme has activity that can cause the production of NAA from NAM, the NAA substance functioning as a plant hormone that is

-171171 Substantially toxic for developing pollen grains, while the NAM precursor is relatively non-toxic. For this example, the lamH gene was inserted into vector PAL1423. The lamH gene was isolated from pPCV311 as described in Figure 19 and cloned as a BamHI-Sstl fragment at the BamHI-Sstl sites in PAL1423, creating PAL1424. This vector has the lamH gene (T-DNA gene 2) under the control of a promoter essential for the development and / or function of the pollen of the L4 clone. PAL1424 was used to transform tobacco as presented in Example 34.

Example 32

The PAL1424 vector was used to transform Brassica napus as presented in Example 34.

Example 33

In this example, vector PAL1107 was used for the production of the specific GUS enzyme (beta-glucuronidase) in tissue. The gene for this enzyme is available from Clonetech Laboratories, Palo Alto, CA, USA. The gene was inserted into PAL1107 20 as a BamHI-Sstl fragment and was used to transform tobacco as in Example 34. The plants produced had detectable GUS activity only in the development of pollen cells, and not in any other tissue tested. It should be noted that the application of the non-toxic analog of glucuronic acid to which a toxic molecule such as glyphosate has been conjugated, can be

-172172 applied to these plants and cleavage of the toxic portion of glucuronic acid could occur only in pollen cells. This provides an example of an enzyme that can be used for the production, in a specific form in tissue, of a toxic substance from a non-toxic analog.

Example 34

This example describes methods used to transform tobacco and Brassica napus.

For the transformation of tobacco, the cultivation of tobacco, N. tabaccum, cv. Delgold was used. To achieve this transformation, tobacco leaves less than 20.32 cm (8 inches) in length were sterilized on the surface through exposure to ethanol for 5-6 seconds, then subsequent exposure to 1% sodium hypochlorite for some minutes, usually 510 minutes, or until the cutting edge of the petiole turned white, then rinsed several times in sterile distilled water. Leaf segments of approximately 0.5 to 1.0 square centimeters were excised from the sterile leaves and cocultivated in a two day shoot induction medium with Agrobacterium tumefaciens GV 3101 carrying the Ti plasmid pMP90 to provide vir functions in trans (described by Koncz, C. Schell, J., 1986, Mol. Gen. genet. 204: 383-396), carrying the binary vector of interest. The vector is usually a derivative of Bin19, which contains the NPT II gene activated by the synthase promoter of

-173173 nopaline and terminated by nos ter for the selection of plant cells with kanamycin. Bin19 is available from Clonetech Laboratories, Palo Alto, CA., USA. Transformed tobacco cells were selected in a 5-shoot induction medium containing 0.8 $ agar, MS salts, vitamins B5, 3% sucrose, 1 mg per L of benzyladenine, 0.1 mg per I of alfanaphthalenacetic acid (NAA) 300 pg / l kanamycin and 500 pg / ml carbenicillin (essentially as described by Horsch et al., 1985, Science, 227: 1229-31). The regenerated shoots 10 were then transferred to a root induction medium consisting of medium B5 with 2% sucrose, 500 pg / ml carbenicillin and 0.5 mg / l NAA and indolacetic acid (IAA). The rooted transformants were transferred to a humidity chamber containing high humidity, after which the humidity was gradually reduced and the plants were subsequently transferred to the greenhouse.

For the transformation of Brassica napus, the binary vector containing the Agrobacterium GV 3101 strain carrying pMP90 to provide vir functions in trans, was used. The transformation was performed either using the method described by Moloney, MM, et al., (Plant Cell Reports (1989) 8: 238-42) or, the transformation can be performed with sterilized stem epidermal layers on the surface. For this procedure, seeds of B. napus L. ssp. oleífera cv. Westar in 'Promix' mixed with 2 g / l of the 25 'Nutricoate' low-release fertilizer in 20.32 cm (8 inch) containers. Plants are

-174174 developed in the greenhouse under a photoperiod of 16 (using natural and artificial light). For co-cultivation and regeneration experiments, stem sections from 3 internode stems of plants with humidity of approximately 1.5 months (i.e. those with elongated flower spikes and several open flowers) were used. Intact stem sections were surface sterilized for 30 seconds in 70% ethanol and 10 minutes in 1% sodium hypochlorite followed by three rinses in sterile distilled water. For the transformation of Aqrobacterium tumefaciens GV 3101 carrying the Ti plasmid pMP90 to provide vir functions in trans and the binary vector of choice was developed in a YEP medium (which consists of 10 gm per liter of yeast extract, 10 g per liter of bactopeptone and 5 g per liter of NaCI, pH 7.0, containing 100 ug per ml of kanamycin for the selection of bacterial cells containing the binary vectors). The cells grew from one to two days at 28 ° C. Cells were harvested via centrifugation and resuspended at a density of approximately 10θ-10? cells by me in liquid EL, which consists of micro and macro MS nutrients and vitamins B5 containing 40 mg / l FeNa-EDTA (obtained from BDH-chemicals) and 3% sucrose, 10 mg / l benzyladenine and 0.5 mg / l of alfanaftalenacético acid (NAA) and 18.8 mM of KNO³ plus 20.6 mM NH4NO³. The medium was solidified with 0.8% agar (Sigma) when EL medium was used for solid media plates.

-175175

The cell suspension was emptied into the bottom of a sterile petri dish and sterilized stems were dissected directly into the bacterial suspensions. The segments were longitudinally sectioned into middle segments and cut into sections of approximately 5 mm. The dissected segments were placed on filter paper discs on a solid EL medium for 3 days of co-cultivation under fluorescent light conditions (60 microinsteins / m2 / sec2) at 25 ° C. After a 2-3 day co-culture, the explants were transferred to the solid EL medium containing 500 ug / ml of carbenicillin, and 100 ug / ml of bekanamycin (Sigma). Sprouts formed in 4-8 weeks, sections were transferred to fresh solid EL medium with carbinicillin and bekanamycin every 3-4 weeks. Sprouts that formed and did not blanch were excised and rooted in PDR medium (B5-containing 2% sucrose and 0.5 mg / l each of NAA and IAA). In some cases, the non-regenerating calluses developed in the selective medium were separated from explants and transferred to the fresh medium to stimulate regeneration. The transformed plants were placed in a humidity chamber, and after two - four weeks they were transferred to the greenhouse. Plants were grown under a 16 hour photoperiod and allowed to flower.

Clonal propagation was used to increase plant lines as well as manual crossing and selection of cross-plant seeding in the medium containing kanamycin.

-176176

This medium consisted of 0.8% agar, one tenth of the MS salts and 100 ug per ml of bekanamycin (available from Sigma Chemicals, St. Louis, MO., USA) without any sucrose in the medium. Sterilized seeds were used on the surface. The seeds were sterilized on the surface by rinsing in 70% ethanol for a few seconds, soaking in 1% sodium hypochlorate for 15 minutes, followed by three times rinsing of sterile distilled water. The seeds were placed on the surface of the agar on sterile plates and allowed to empty. Plants which do not carry the kanamycin gene linked to the antisense gene remained green and were subsequently transferred to the soil and allowed to flower.

Example 35

This example describes the isolation of two genes involved in tumor formation in plant tissues after infection with Agrobacterium, the lamS and lamH genes and the Ti plasmid from Agrobacterium tumefaciens strain C58. The isolation of the lamH gene is particularly described. The source of DNA encoding these genes was plasmid pPCV 311. Plasmid pPCV311 is described in Koncz, C. and Schell, J., Molecular and General Genetics, (1986), 204: 383-396, and contains the oncogenic region of the T-DNA plasmid contained in Agrobacterium strain C58. Plasmid pPCV 311 contains a T-DNA region that, when transferred to plant cells, causes tumor formation. This

-177177 T-DNA oncogenic region is completely contained in plasmid pPCV-311. This region of DNA contains four genes, which when expressed in plant cells are sufficient for tumor formation. The approximate coding regions of these four genes and the transcription direction of these four genes are indicated in Figure 15 portions of the pPCV 311 vector are not shown as they do not refer to the following constructs. Furthermore, the oncogenic region of the C58 Agrobacterium strain is located on the T-DNA plasmid within that bacterium, commonly referred to as the wild-type nopaline plasmid. An absolutely identical oncogenic region was also found in wild-type octopine strains that can also be used as a source of genes. The complete nucleotide sequence of an octopin strain oncogenic region is described by Barker et al., Plant Molecular Biology 2: 335-50 (1983). The partial sequence obtained from various gene constructs derived from pPCV 311 was compared to the published nucleotide sequence.

Two pPCV 311 genes, the lamH and lamS genes, commonly referred to as genes 2 and 1, respectively, were isolated. The gene was isolated by first subcloning the indicated Hindlll fragment, a fragment containing the entire coding region of gene 2 and the additional 5 'sequences that were subsequently removed for the construction of a promoter-free version of the gene. The mapped restriction sites in this subclone are

-178178 show in Figure 15 and the subclone is named as pPAL

G2. For the isolation of the coding sequences only pPAL G2 was first divided into two smaller clones and the gene was then reconstructed. The Xbal-Smal and Smal 5 Smal fragments shown in Figure 15 were isolated through gel lusion. and subsequently cloned into the following vectors: The Xbal-Smal fragment was cloned into pGEM 4Z, giving rise to pPAL899. The Xbal-Smal fragment was subcloned into pGEM 7Z, giving rise to pPAL898. The 5 'uncoded sequences of the lamH gene that are present in this subclone were removed as follows: pPAL898 was digested with EcoRI, the EcoRI site in this promoter region of the clone and in this subclone is only the EcoRI site. This digested DNA was then treated with Exonuclease III, and after nuclease S1 and 15 digestion the Klenow fragment of DNA polymerase I. The treated DNA was then cut with PsTI and treated with the klenow fragment in order to make the Pstl shaving site. The linear, digested shaved-end plasmid was then relegated and used to transform E. coli DH5-alpha according to normal protocols. Twenty subclones were chosen, sequenced, and one subclone was chosen which was deleted for nucleotides opposite the ATG start in the translational codon. The ATG start codon was determined by comparing the nucleotide sequence obtained from the deleted subclones to the nucleotide sequence for the octopine strain described by 25 Barker, et al., Plant Molecular Biology 2: 335-50 ( 1983). The

-179179 nucleotide sequences from both the 5 'uncoded region and decoding were nearly identical. This subclone was named pPAL897, the ATG codon is shown in Figure 15, the transcription direction in this case could be from right to left in Figure 15. The plasmid contained the 5 'half of the coding region for the gene. lamH, with the promoter sequences removed.

.The construction of the 3 'half of the lamH gene, contained in the plasmid pPAL898, was carried out as follows. A 3 'region of the gene that contains the polyadenylation signal naturally found in the gene, was isolated through digestion of pPAL898 with the enzymes BamHI and Apal. The digested DNA was treated with the klenow fragment to make it shaved ends and was relegated. This gave rise to subclone pPAL896 which is a plasmid containing the 3 'half of the lamH gene. To reconstruct the intact lamH gene, pPAL896 was digested with Hindlll and Smal, and the 3 'half of the gene fragment was isolated through gel elution. PPAL897 was digested with Smal and Hindlll and the isolated 3 'fragment of pPAL896 was cloned at these sites, reconstructing a promoter-free version of the gene containing the indicated array of restriction sites flanking the gene. This plasmid was named pPAL895 and is shown in Figure 15.

Example 36

-180180

This example describes the isolation and construction of a promoter-free version of gene 1, lamS: the indolacetamide synthase gene from the Ti plasmid of Agrobacterium tumefaciens strain C58, the procedure of which is summarized in Figure 16. The gene was isolated from plasmid pPCV311. The Smal-Pstl fragment containing the 5 'and 3' regions of the lamS gene as well as the coding region, was isolated through gel elution and subcloned into a derivative of pGEM 4Z called pGEM-noEco. PGEM-noEco is a plasmid from which the EcoRI site of pGEM 4Z has been removed by cutting with EcoRI and making shaved ends and relegating, so that the EcoRI site was removed. This fragment was inserted in the orientation shown relative to the polylinker. This subclone was named pPAL889. PPAL889 was digested with EcoRI, and briefly treated with exonuclease III, followed by nuclease S1. The DNA was digested with Smal and treated with a klenow fragment to make it shaved ends. DNA was relegated and clones were recovered. Some of these clones were chosen, sequenced, and one clone was found, which has 5 'sequences deleted, so that only about 15 bases upstream of the ATG start of the translation codon remained. This plasmid was named pPAL888. The Kpnl site at the 5 'end of the gene as well as the Pstl site at the 3' end of the gene were both converted to Salí sites by cutting with Kpnl, filling the end with klenow and adding synthetic Salí linkers, and repeating the addition of the linker

-181181 at the Pstl site, so that the entire gene can be excised as an individual fragment of Salí. This plasmid was named pPAL8887. This plasmid contains the promoter-free version of the lamS gene and contains the array of restriction sites shown in the gene flank as shown in Figure 16.

Example 37

This example relates to the detailed characterization of the coding regions of clone number L4 in a specific microspore clone isolated from a genomic collection of Brassica napus and the construction of a vector containing these genes. Figure 2a is a schematic representation of the restriction map and coding regions of the L4 clone number. The clone contains three different members in the same gene family. These genes are identified as Bp4A, Bp4B, and Bp4C. The first gene (Bp4A) and the third gene (Bp4C) are functional, the second gene has modifications that probably make it non-functional. The restriction map is done diagrammatically since the untranscribed regions are shown as a single line, while the transcribed regions are shown as an area in a box. The second gene (Bp4B) is identified based on sequence homology and is therefore in a box with a dotted line. The 220 annotation refers to a deletion / rearrangement of approximately 220 base pairs affecting the second gene (Bp4B) in this clone. The start of the transcription is located in the

-182182 leftmost side of each area in a box (except for the Bp4B gene) and the exon and intron positions are noted by the exons that are filled in black and the intron positions that are not filled. A small arrowhead is shown above the 5 'untranscribed region of each gene, this arrowhead serves to indicate the promoter region of each gene. Restriction sites are identified such that the number of the first nucleotide of the restriction enzyme recognition site is shown. No restriction sites are shown, only those relevant to the detailed constructions. Genes are represented with the 5 'region on the left side and the 3' region on the right side. The DNA sequence numbering is from left to right, 5 'to 3' in all cases.

In Figure 3a, the DNA sequence of the L4 clone, the sequence orientation and the genes contained within the specific microspore clone from Brassica napus are shown and are 5 'to 3'. In Figure 3a, clone L4, nucleotide 1 in the full sequence is furthest to the left of the EcoRI site, while nucleotide 8579 is in the first nucleotide of the site furthest to the right of EcoRI. The start of gene 1 transcription in clone L4 is nucleotide 235. The 5 'and 3' intron binding sites are identified in bold type. The start ATG codon is also shown as well as the stop stop codon. The deduced amino acid sequence of the proteins encoded for these genes is also shown. The transcription end for the

-183183 gene 1 is approximately nucleotide 1427. As indicated above, the second gene in clone L4 is probably non-functional due to insertion and deletion occurring in the promoter region and first exon. This gene was not used for constructs. The third gene in clone L4 has a transcriptional start at which position number 6298 in the DNA sequence and the transcription ends at approximately nucleotide 7490. This ATG start codon, intron binding sites and end stop codon are all identified as above, vectors were constructed from this clone using promoter fragments of both genes 1 and 2. The fragment constructs of specific promoters are detailed below.

Construction of two vectors containing the promoter and promoter fragments of clone L4 was performed as follows and is shown in Figure 7 (A, B, C, D). The first vector was constructed by first subcloning the EcoRI-Sstl fragment (nuc. 1-2132) containing the first gene from clone L4 (235 promoter / exon / intron / second exon base pairs) into a commercially available vector pGEM-4Z ( Promega Biotech, Madison, Wl, USA), using the EcoRI-Sstl sites of the polylinker of this vector. This plasmid was named pPAL0402. The 2.7 Kb EcoRI fragment from clone L4 containing the third gene (Bp4C) was then cloned into the EcoRI site of pGEM 4Z, leading to a plasmid named pPAL0411. Plasmid pPAL0402 was then digested with EcoRI and

-184184 the 2.7 Kb EcoRI fragment of pPAL0411 (nucí. 5859-8579) containing the three numbers of the gene (Bp4C) from the L4 clone that was added to it. The clones were recovered that contained this 2.7 kb EcoRI fragment inserted in both orientations relative to the promoter region of the first gene. A clone that contained this third gene fragment in one orientation, so that the promoter of the third gene was opposite to the promoter in the first gene, was chosen and named as pPAL0403. Plasmid pPAL0403 contains the entire third gene of the L4 clone oriented in such a way as to have the promoter region immediately adjacent to the 235 base pair promoter region of the first gene in pPAL04303. This plasmid, pPAL0403 was digested with Ddel. Ddel digestion producing a fragment of approximately 1.9 kb. Ddel sites are located at nucleotides 303 and 7366. Due to the orientation of these fragments, Ddel digestion produces a 1.9 Kb fragment. This 1.9 Kb fragment contains a copy of the third gene (Bp4C) oriented, so that the transcription direction of this third gene is from right to left, fused to the 235 base pair promoter fragment from the first gene of the clone L4 (Bp4A) which was transcribed from left to right, ending at the Ddel site which is located 67 base pairs downstream of the major transcription start site and proceeds to the ATG start of the translational codon through 2 nucleotides. This 1.9 Kb Ddel fragment was shaved with the fragment

-185185 klenow and cloned into the Xbal site of the pGEM 4Z polylinker region previously shaved ends with the klenow fragment. The resulting plasmid pPAL0408 was recovered and subsequently digested with Salí and Sstl, which released the cloned Ddel fragment bounded on the left (nucí 7366) Salí and on the right (nucí 303) side of this construct and contains a portion of the pGEM 4Z polylinker containing the following unique sites: BamHI, Smal, Kpnl and Sstl restriction enzyme sites. This Salí-Sstl fragment was cloned into the Salí-Sstl sites of PAL1001. PAL1001 is a Bin19 binary vector (described by Bevan, M., Nucleic Acids Res., 1984, 12: 8711-8721) to which the noster polyadenylation signal was added as a 260 bp Sstl-EcoRI fragment isolated from plasmid pRAJ221 (available from Clonetech Laboratories, Palo Alto, CA USA) at the Sstl-EcoRI sites of the Bin19 polylinker region. Nos ter was identified as a dotted box. The binary transformation vector resulting from the insertion of the Salí-Sstl fragment from pPAL0408 to PAL1001 was named PAL1107. Construction details are shown in Figure 7a. This vector has a copy of the third oriented gene, so that the transcriptional orientation of this third gene is from right to left, fused to the 235 base pair promoter fragment from the first gene of clone L4, which was transcribed from left to right, followed by a polylinker with unique sites for DNA insertion consisting of: BamHI, Smal, Kpnl and Sstl followed by nos.

-186186

This vector has the characteristic that additional 5 'non-coding sequences were placed upstream of the 235 base pair core promoter on Bp4A, but these additional 5' sequences were in opposite orientation. The provision of these sequences in this orientation does not affect the specificity of the 235 base pair promoter nucleus pollen. With this vector, you can also find out the position effects of the transformation procedure since the vector contains an intact copy of the Bp4C gene, with a probe for the expression levels of the Bp4C gene that can give an indication of what expression levels from of the other promoters essential for the development and / or function of pollen in PAL1107, can be expected.

Example 38

This example describes the additional constructs that the promoter regions of the genes contained in the L4 clone have, which may be useful for the specific expression of gene sequence pollen. The three genes within the L4 clone (Bp4A, Bp4B, Bp4C) show very accurate DNA homology and this is very evident between the first gene (Bp4A) and the third gene (Bp4C). The second gene (Bp4B) is a homologous copy that has undergone sequence changes that appear to have led to inactivation. The extensive similarity between the first, second and third genes in the L4 clone was also maintained in the region of

-187187 promoter, so that of the 235 nucleotides of the promoter regions of the first and third genes there are only 5 nucleotides that differ from each other. Downstream of the TATA box in these two promoters, the only difference between them is the presence of an additional nucleotide at the start of transcription (eg promoter 1, Bp4A: ...... TATGTTTtAAAA ... against the

Promoter 3, Bp4C ,: ....... TATGTTTAAAA ..... shows that the underlined transcribed region, the only lowercase nucleotide difference). However, within the first gene sequence there is a nucleotide change that introduces a Ddel site (nuc. 303) into the 5 'untranslated leader sequence upstream of the ATG start codon that is not present in the untranscribed leader sequence. of the third gene in clone L4. Chimeric promoter constructs were made which used this Ddel site and in the first gene linked to the promoter sequences of the third gene. The region of the first promoter used for these constructs consisted of the sequences contained between the SnaBI site (nucí 210) near the TAT box towards the Ddel site located immediately upstream of the ATG start codon in the first gene (nucleotide 303 is the first nucleotide in the recognition sequence for Ddel). The other region of this chimeric promoter (5 'from the TATA box) was a fragment extending from the EcoRi site of the third promoter (nucleotide 5858) to the SnaBI site near the TATA box (nucleotide 6273). Thus,

-188188 To facilitate the construction of these specific vectors in pollen, the following constructions were made.

The EcoRI to Ddel fragment spanning the promoter region of the first gene in clone L4 was isolated by first cutting pPAL0402 with Ddel, shaving with Klenow, and then cutting with EcoRI. The 235 base pair fragment corresponding to this region was cloned into the EcoRI-Smal sites of pGEM 4Z. This plasmid (pPAL0422) was then cut with EcoRI and Snabl. A DNA fragment containing the EcoRI to SnaBI portion of the promoter for gene 3 in clone L4 was isolated by digestion of pPAL0411 with EcoRI and SnaBI. This released an approximately 415 base pair EcoRI fragment (nuc. 5858) to the SnaBI fragment (nuc. 6273) that represents the majority of the 5 'region of the L4 clone gene 3 promoter (the SnaBI recognition site is 2 pairs of bases downstream of the TATA box). The EcoRI-SnaBI fragment was used to replace the shorter EcoRI-Snabl fragment removed from the first promoter subclone (pPAL0422) by reconstructing a promoter fragment of approximately 550 base pairs. This plasmid is called pPAL0421. This chimeric promoter fragment contains 415 base pairs of the promoter of the three genes in clone L4, followed by approximately 99 nucleotides the first gene promoter / untranslated leader sequence.

For the construction of a pollen specific cassette vector, the following plasmids were first constructed. The first

-189189 constructed plasmid contained the polyadenylation signal from nos ter with a polylinker in front of nos ter. This was achieved by first isolating nos pRAJ22 as a Satl-Ecorl fragment and this fragment was cloned into pGEM 4Z using the Sstl and 5 EcoRI sites in the polylinker. This subclone is named as pPALOOl. To pPALOOl, a fragment encoding neomycin phosphotransferase (NPT II) derived from plasmid pRAJ162 was added, in the antisense orientation as follows: Plasmid pRAJ162 contains the NPT II gene from transposon TN5 inserted as a Sali fragment and linked to through a polylinker on plasmid pUC-9 (which was obtained from the Plant Breeding Institute, Cambridge, UK. pRAJ162 was digested with Hindlll and Smal. The DNA fragment containing the NPT II gene was isolated through elution from an agarose gel. pPALOOl was digested with Hindlll and Smal and the NPT II gene fragment was inserted.

The resulting plasmid was named pPAL002 and had such restriction site orientation and the NPT II gene and we ter as follows: Hindlll, PstI, Sali, 3 'and NPT II the 5' end and 3 'end coding sequence , Salí, BamHI, Saml, Kpnl, Sstl, nos ter, 20 EcoRI. pPAL002 was cut with Hindlll and the site was shaved end through the use of the klenow fragment. pPAL0421 was digested with Hincll and pVUlI, both of which leave shaved ends, and the promoter fragment was ligated to Hindlll cut shaved end pPAL002. The plasmids were 25 obtained, they contained the promoter in both orientations with

-190190 relation to us ter. A plasmid was chosen with the appropriate orientation (antisense 5 '/ NPT II promoter / nos ter) and was named pPAL0419. PPAL0419 stains the following DNA fragments: A small (approximately 130 bp) pGEM 4Z containing the SP6 promoter, the 550 base pair chimeric promoter, the NPT II gene in the antisense orientation relative to the promoter, followed by the polyadenylation signal ends. This complete promoter / NPT ll / nos ter construct can be cut through EcoRI. In order to provide promoter 10 sequences that can be used with additional gene constructs, the plasmid pPAL0419 was digested with EcoRI, this digestion removes the coding region of NPII, and this Sali digested with pPAL0419 was relegated giving rise to pPAL0402. PPAL0420 represents the essential promoter for the development and / or function of pollen followed by a pollinator for the insertion of genes that has the following unique ones: Hincll, Pstl, Salí, BamHI, Saml, Kpnl, Sstl, followed by the signal of polyadenylation of nos ter. The entire promoter / polylinker / nos ter construct can be conveniently excised as a single EcoRI fragment. 20 The details of this construction are shown in Figure 7B. This plasmid was used for the construction of an additional promoter essential for the development and / or function of pollen in a binary transformation vector. The intact L4 clone in the lambda cloning vector was digested to complete with the restriction enzymes Sstl and Hhal. The resulting fragments were separated

-191191 via gel electrophoresis and a 2.65 Kb fragment containing the promoter / first exon / intron / region of second partial exon of gene three in clone L4 and corresponds to nucleotides 4565 to 7216 in the sequence of clone L4 , and isolated. This fragment 5 was made from klenow shaved ends and cloned into the binary transformation vector PAL1001 previously described with reference to Figure 7a. PAL1001 was first cut with Hindlll and made from shaved ends with klenow. Clones containing this fragment (promoter / first exon / intron / second partial exon) were recovered. A clone was chosen that contained this fragment in the proper orientation so that the direction of transcription was toward nos in PAL1001. This vector was named PAL1421.

This vector contains approximately 1.9 Kb of the promoter region upstream of gene 3 in clone L4 followed by the first 15 exon, the complete intron and 15 bases of the second exon of gene three followed by a polylinker containing the following unique sites: , Xbal, BamHI, Smal, Kpnl, Sstl, and finally the polyadenylation signal of the term nos. A variant of this vector was constructed by digesting PAL1421 with EcoRI and replacing the promoter / exon / intron / second exon / polylinker / nos _____ter structure with the promoter / polylinker / nos ter structure from pPAL0420 using EcoRI , so that the longest 5 'promoter region in the binary transformation vector was reconstructed. The resulting vector is

-192192 named PAL1423. The summary of this construction is shown in Figure 7D.

Example 39

In this example, a promoter essential for the development and / or function of pollen was used to synthesize the enzyme lamH specifically in pollen cells. The enzyme has activity that can cause the production of NAA from NAM, the substance NAA functioning as a plant hormone that is substantially toxic to the development of pollen grains, while the NAM precursor is relatively non-toxic. For this example, the lamH gene was inserted into vector PAL1423. The lamH gene was isolated from pPCV311 as described in Figure 15 and cloned as a Salí fragment at the Salí site of PAL1423, creating PAL1426. This vector has the lamH gene (T-DNA gene 2) under the control of a promoter essential for the development and / or function of pollen from clone L4 in the sense orientation. PAL1426 was used to transform tobacco as presented in Example 34.

Example 39a

In this example, constructs were made using lamS (gene 1) or lamH (gene 2) that were expressed from the microspore-specific Bp10 promoter or anther-specific TA39. The four constructs are shown in Figure 23 (Plasmid

-193193 pPHP5838, bp10: lamS), Figure 24 (pPHP5839, Bp10: lamH), Figure 25 (pPHP5840, TA39: lamS) and Figure 26 (pPHP5841, TA39: lamH). All of these four constructs are generated using a base structure of the plant transformation vector pALLTKREP (Boutilier, KA, Gines, M.-J., DeMoor, JM, Huang, B., Baszczynski, CL, lyer, VN, and Miki , BK Expression of the BnmNAP subfamily of napin genes coincides with the induction of Brassica microspore embryogenesis, 1994. Plant Molec. Biol. 26: 171 1-1723). pALLTKREP was renamed to Pioneer and will be referred to anywhere as PPHP1741.

Relevant portions of pPHP1741 that are retained on all four Pioneer plasmids include: (1) a broad-scale host bacterial vector, including a bacterially expressed kanamycin resistance gene as the selectable marker, (2) left and right T-DNA limiters derived from Agrobacterium tumefaciens, and (3) among the limiters T-DNA, a chimeric selectable marker gene for plant expression of kanamycin resistance, comprising a CaMV 35S promoter operably linked to the NPTII open reading frame (ORF) and the 3 'untranslated region of CaMV 19S, which encodes signals required for proper processing of the 3' end of the transcript. Observe on the maps the broad host scale origins of replication that are named as OriV and OriT (vegetative origin of replication and transfer origin of replication); the bacterial selectable marker

-194194 is named as SF KanR; the left and right limiters of T-DNA are referred to as LB and RB respectively; the CaMV 35S promoter is referred to as 35SPro; and the CaMV 19S 3 'region is designated as "CT.

In Figures 25 and 26, incorporated between the TA39 promoter and the IAMS or IAMH genes is the untranslated leader of tobacco mosaic virus (TMV), 0 ', (Gallie, DR, Sleat, DE, Watts, JW, Turner, PC, and Wilson, TMA 1987. The 5 'leader sequence of tobacco mosaic virus RNA improves expression of foreign gene transcripts in vitro and in vivo Nucí Acids Res. 15: 3257-3273; Gallie, DR, Sleat, DE, Watts, JW, Turner, PC, and Wilson, TMA1987. A comparison of eukaruotic viral 5'leader sequences as enhancers of mRNA expression in vivo. Nucí. Acids Res. 15: 8693-8711; Sleat, DE, Hull, R., Turner, PC, and Wilson, TMA 1988. Studies on the mechanism of translational enhancing ment by the 5'-leader sequen ce of tobáceo mosaic virus RNA. Eur. J. Biochem. 175: 75-86. 0 's is the designation on the maps in Figures 25 and 26 for the TMV untranslated leader, which is known as 0'.

The constructs also contain the 3 'untranslated region of either the nopaline synthase gene (nos) from the Ti plasmid pT¡T37 from Agrobacterium tumefaciens, or from the proteinase inhibitor II (Pl-ll) gene of Solanum tuberosum (potato) . Both are well known in the art. The reference for the 3 'region is: Bevan, M., Barnes, WM, and Chilton, M.-D. 1983.

-195195

Structure and transcription of the nopaline synthases gene region of T-DNA. Nucí. Acids Res. 11: 369-385. The reference for the Pl-ll 3 'region is: An, G., Mitra, A., Choi, HK, Costa, MA, An, K., Thornburg, RW, and Ryan, CA, 1989. Functional Analysis of the 3 'Control Region of the Potato Wound-inducible Proteinase Inhibitor II Gene. Plan Cell 1: 115-122. Note that Figures 23-26, the 3 'region is denoted as NT and the Pl-ll 3' region is denoted as Pinll.

Figures 27-30 list partial sequence information for plasmids pHP5838, pPHP5839, pPHP5830 and pPHP5841 for which maps are provided in Figures 2326 respectively. The nucleotide numbers indicated in Figures 27-30 are the same numbers on the maps in Figures 23-26. The sequences provided in Figures 27-30, extend from the left limiting region to the right limiting region of each plasmid. That part of the DNA sequence not included is common to all four plasmids and belongs to plasmid pPHP1741 (pALLTKREP, described above).

Example 40

Vector PAL1426 was used to transform Brassica napus as described in Example 34.

Example 41

-196196

In this example, vector PAL1107 was used for the production of the specific GUS enzyme (beta-glucuronidase) in tissue. The gene for this enzyme is available from Clonetech Laboratories, Palo Alto, CA., USA. The gene was inserted into PAL1107 as a fragment of BamHI-Sstl and was used to transform tobacco as described in Example 34. The plants produced had GUS activity detectable only in pollen-developing cells, and in no other tissue tested. The application of a non-toxic glucuronic acid analog to which a toxic molecule was conjugated can be applied to these plants and the excision of the toxic portion of glucuronic acid could only occur in the pollen cells. This provides an example of an enzyme that can be used for the production, in a specific form in tissue, of a toxic substance from a non-toxic analog. Such an analog that can be used is Gluc-Camp, a glucuronic acid analog that is conjugated to chloramphenicol. When activated under glucoronidase, chloramphenicol is produced, inhibiting cell growth and development.

Example 42

In this example, two isogenic plant lines (A1, A2) were produced that carry the genes either lamS or lamH. Tobacco plants were transformed with PAL1426, containing the lamH gene as in Example 39, producing line A2. The lamS gene described in Figure 16 was inserted as a fragment of Salí in

-197197 the vector PAL1423 in the sense orientation, giving rise to PAL1425. PAL1425 was used to transform tobacco as described, and tobacco plants bearing PAL1425 were produced. These lines represented lines A1. Tobacco plants containing both PAL1426 and PAL1425 were purified and homozygous A1 and A2 lines were selected.

Example 43

In this example, PAL1426 (see Example 39) and PAL1425 were used for the transformation of Brassica napus. Homozygous plant lines for genes A1 and A2 were selected as in Example 42.

Example 44

Gene 1 (lamS) and gene 2 (lamH) constructs were introduced to different canola lines through transformation, as in Example 34. Two different anther specific promoters were used for each gene: tobacco TA39 and Bp10 from Brassica constructed as described in Example 39a and Figures 23-26. TO plants were recovered each carrying the gene / promoter combination. TO plants carrying gene 2 were observed completely normal, and were completely male and female fertility. Some of the OTs carrying gene 1 were abnormal, showing very large flowers and not much pollen. The pollen of these plants was probably not fertile. In the

-198198 crosses described above, generally the most normal gene 1 plants were used. Crosses were made between the plants containing gene 1 and gene 2. Eight T1 progenies were identified which are positive by 5 POR for the presence of both gene 1 as gene 2. The following combinations of T1 were obtained: TA39 / gene 1 +

TA39 / gen 2 (2 plants); TA39 / gen 1 + TA39 / gen 2; TA39 / gen 1 +

TA39 / gen 2; TA39 / gen 1 + TA39 / gen 2; BP10 / gen 1 + TA39 / gen 2;

BP10 / gen 1 + TA39 / gen 2 (2 plants); BP10 / gen 1 + BP10 / gen 2 (2 10 plants); BP10 / gen 1 + BP10 / gen 2. T1 progeny were male sterile and female fertile. Male fertility plants showed very low anthers and both lacked pollen and pollen became brown. In some cases, male sterility plants actually produced 15 pods but no seeds were presented. One of the T1 progeny only, when purified produced limited seed. Female fertility was confirmed by applying wild-type pollen to plants that resulted in seed production.

-199199