CA2103572C - Stamen-specific promoters from corn - Google Patents

Abstract

The present invention relates to male flower specific cDNA sequences isolated from corn and stamen-specific promoter sequences of genes corresponding to such cDNA sequences. The invention further describes how the promoters of the invention can be used to direct expression of DNA sequences, such as for instance, a male-sterility DNA
in plants, so as to obtain male-sterile plants.

Description

STAMEN-SPECIFIC PROMOTERS FROM CORN
This invention relates to promoters isolated from corn which can provide <gene e:~pression predominantly or specifically in stamen cc=_lls of a plant, particularly a monocotyledonous plant, and thereby provide little or no gene expression in oth.e:_ parts of the plant that are not:
involved in the production of fertile pollen. The promoters are useful in the production of transformed plants, in which a gene is to be expressed at least. predominantly, and 1C preferably specifically, in t:he ~~tamen cells, preferablir in the anther cells. The promoters are especially useful in the production of male-sterile pla.nt~; and male fertilityr-restorer plants as described in European patent publications 344 029 and 412 911, respectively particularly in the production of hybrids of monocotyl.edc>nous plants, such as corn, rice or wheat.
Summary of the Invention In accordance with this invention are provided:
male flower-specific c:DNA sequences isolated from corn comprising the sequences, SEQ ID no. 1 and SEQ ID no. 2, shown in the sequence listing. Also in accordance with this invention are provided stamen-specific, preferably anther-specific, promoters of the corn genes corresponding to such cDNA sequences, particwlarly the ~:~romoter which controls the expression of the genomic coding sequence corresponding to the cDNA of SEQ TD no. 2 and which i:~ contained within the sequence of nucleotides 1 to 1179 of SEQ ID no. 3 (the "CA55 promoter" or "PCA55")- Each of such promoters can be used in a foreign DNA sequence, preferabl;r a foreign chimaeric DNA sequence, which contains a structural gene, preferably a male-sterility DNA or a male fertilit=y-restorer DNA, under the transcriptional control of the promoter and which can be used to transform the nuclear genome of a cell of a plant, particularly a monocotyledonous plant. Further in accordance with this invention are provided: the male-sterile plant or male fertility-restorer plant which can be regenerated from such a cell transformed with the foreign DNA sequence of this invention; the transformed cell, itself; a culture of such a transformed cell; seeds of such a regenerated plant and its progeny; and a fertility-restored plant and its seeds resulting from crossing such male-sterile and male fertility-restorer plants.
In one aspect, there is described a recombinant anther-specific promoter region, wherein said promoter region has the sequence of SEQ ID No. 5 between nucleotide position 1224 and nucleotide position 2408.
In another aspect, there is described use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA comprising the promoter region described above, operably linked to a heterologous structural gene, for the production of plants in which said heterologous structural gene is to be expressed predominantly in anther cells.
In another aspect, there is described use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA comprising a DNA encoding a ribonuclease operably linked to the promoter region described above, for the production of male-sterile plants, or for the production of hybrid monocotyledonous plants.
In another aspect, there is described use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA comprising a DNA encoding a barnase -2a-operably linked to the promoter region described above, for the production of male-sterile plants, or for the production of hybrid monocotyledonous plants.
In another aspect, there is described use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA comprising a DNA encoding a barstar operably linked to the promoter region described above.
In another aspect, there is described a method for isolating a stamen-specific promoter from corn which comprises performing inverse PCR on genomic corn DNA using a pair of oligonucleotides consisting of a first oligonucleotide complementary to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 2 or complementary to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 3 between nucleotide positions 1120 and 1839; and a second oligonucleotide corresponding to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 2 or corresponding to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 3 between nucleotide positions 1120 and 1839.
Detailed Description of the Invention In accordance with this invention, a male-sterile plant or a male fertility-restorer plant can be produced from a single cell of a plant by transforming the plant cell in a known manner to stably insert, into its nuclear genome, the foreign DNA sequence of this invention. The foreign DNA
sequence comprises at least one male-sterility DNA or male fertility-restorer DNA that is: under the control of, and fused in frame at its upstream (i.e., 5') end to, one of the stamen-specific, preferably anther-specific, particularly tapetum-specific, promoters of this invention, such as the -2b-promoter and optionally the leader sequence of SEQ ID No. 3;
and fused at its downstream (i.e., 3') end to suitable transcription termination (or regulation) signals, including a polyadenylation signal. Thereby, the RNA and/or protein or polypeptide, encoded by the male-sterility or male fertility-restorer DNA, is produced or overproduced at least predominantly, preferably exclusively, in stamen cells of the plant. The foreign DNA sequence can also comprise at least one marker DNA that: encodes a RNA and/or protein or polypeptide which, when present at least in a specific tissue or specific cells of the plant, renders the plant easily separable or distinguishable from ot~ier plants which do not contain such RNA and/or protein or polypeptide at least in the specific tissue or specific cel:L~~; is under the control of, and is fused at its 5' end to, a second promoter which is capable of directing expression of the marker DNA at. least in the specific tissue or specific cells; and is fused at its 3' end to suitable transcription termination signals, including a polyadenylation sign.a_~. The marker DNA i~~ preferably in the same genetic locus as the male-sterility or male fertility-restorer DNA. This linkage between the male-sterility or male ferti:Lity-restorer DNA and the marker DNA
guarantees, with a high degree of certainty, the joint segregation of both the male-sterility or male fertilit~r-restorer DNA and the rr~.a:rker DNA into offspring of the plant regenerated from the transformed plant cell. However in some cases, such joint ;~egregatior~ i~~ not desirable, and in such cases, the marker DNA should be in a different genetic locus from the male-st.e:rility or nuale~ fertility-restorer DNA.
The male-sterility DNA of this invention can be any gene or gene fragrr~e:nt, whose expression product (RNA
and/or protein or polypeptide) disturbs significantly the metabolism, functionir:~g and/or development of stamen cells, preferably anther cells, and thus prevents the production of fertile pollen. Preferred male-sterility DNAs are described in publication of EP 344 029, for example those DNAs encoding: RNases such. as RNase T1 or barnase; DNases such as endonucleases (e. g., EcoRI); proteases such as papain;
enzymes which catalyse the synthesis of phytohormones (e. g., isopentenyl transferaae or the gene products of gene 1 and gene 2 of the T-DNA of: Agrobacterium; glucanases; lipases;

-3a-lipid peroxidases; plant cell wall inhibitors; or toxins (e.g. , the A-fragment ::o diphtheria toxin or botulin) .
Other preferred examplc~u=s of male-sterility DNAs f :"
~ i.r.~ cjJ 4 are antisense DNAs encoding RNAs complementary to genes, the products of which are essential for the normal development of fertile pollen. Further preferred examples of male-sterility DNAs encode ribozymes capable of cleaving specifically given target sequences of genes encoding products which are essential for the production of fertile pollen. Still other examples of male-sterility DNAs encode products which can render stamen cells, particularly anther cells - and not other parts of the plant - susceptible to specific diseases (e. g. fungi or virus infection) or stress conditions (e. g. herbicides).
The construction of a vector comprising a male-sterility DNA, such as a barnase-encoding DNA, under the control of a corn anther-specific promoter of this invention, is most conveniently effected in a bacterial host organism such as E. coli. However, depending on the nature of the male-sterility DNA and the specific configuration of the vector, problems can be encountered due to the expression of the male-sterility DNA in, and the concurrent decrease of viability of, the host organism.
Such problems can be solved in a number of ways. For instance, the host organism can be provided, on the same or a different plasmid from that containing the male-sterility DNA or even on its chromosomal DNA, with another DNA
sequence that prevents or inhibits significantly the effect of the expression of the male-sterility DNA in the host organism. Such an other DNA sequence can encode, for example: an antisense RNA so that the accumulation and translation of the male-sterility RNA is prevented: or a protein (e.g., barstar) which specifically inhibits the gene product of the male-sterility DNA (e. g., barnase;
Hartley (1988) J. Mol. Biol. 202, 913). Alternatively, the male-sterility DNA can contain elements, such as a plant intron, which will only result in an active gene product in SUBSTITUTE SHEET

_ 5 a plant cell envi:ronmeczt.. Examples of introns that can be used for this purpose a~~e intx-ons of the transcriptional units of: the adh-1 gene of maize (Luehrsen and Walbot (1991) Mol. Gen. Genet. 2.25, F31; Mascarenhas et al (1990) Plant Mol. Biol. 15, 91~:), the shrunken-1 gene of maize (Vasil et al (1989) Plaint: Phy:~iol. 97., 1575) , the cat-1 gene of castor bean (Tanaka E:t al (1990) Nucleic Acids Research ("NAR") 18, 6767), the act-1 gene of rice (McElroy et al.
(1990) The Plant Cell 2, 163; PCT ;publication WO 91/09948) and the TA36 gene (intron shown in SEQ ID no. 4).
The male fert:i.lity-restorer. DNA of this invention can be any gene or gene fragment, whose expression product (RNA and/or protein or ~x~lypeptide) inactivates, neutralizes, inhibits, blocksr offsets, overcomes or otherwise prevents the ~~pecific activity of the product of a male-sterility DNA in stamen cells, particularly in anther cells. Preferred male fertility-restorer DNAs are described in publication of EP 4:1a? 911, for example those DNAs encoding: barstar which is the inhibitor of barnase; EcoRI
methylase which prevent; the activity of_ EcoRI; or protease inhibitors (e. g. the inhibitors of papain). Other examples of male fertility-restorer DNAs are an.tisense DNAs encoding RNAs complementary to male-sterility DNAs. Further examples of male fertility-restorer DNAs encode ribozymes capable of cleaving specifically ~~:iven t<~rget sequences of male-sterility DNAs.
The marker DNA of this invention can be any gene or gene fragment encoding an RNA and/or protein or polypeptide that allows plants, expressing the marker DNA, 3C to be easily distinguished and separated from plants not expressing the marker DNA. Examples of the marker DNA are described in publication of E:P 344 029, such as marker DNAs which encode proteins or polypepti.des that: provide a distinguishable color Lei plant: cells, such as the A1 gene encoding dihydroquercei::i.n-4-reductase (,Meyer at al (1987) Nature 330, 677-678) anc~t the g:lucuronidase gene (Jefferson et al (1988) Proc. Natl. Acad. Sci. USA ("PNAS") 83, 8447);
provide a specific morpluological characteristic to a plant such as dwarf growth o:~~ a diff=erent shape of the leaves;
confer on a plant stres:> tolerance, such as is provided by the gene encoding supe:roxide dismutase as described in publication of EP 359 61.7; confer disease or pest resistance on a plant, such as is provided by a gene encoding a Bacillus thuringiensis e:ndotoxin conferring insect resistance on a plant, ~~s desc:ribed in publication of EP 193 259 or confer on a plant a bacterial resistance, such as is provided by the :b~:~cterial peptide described in publication of EP 299 828. Preferred marker DNAs encode proteins or polypeptide~~ inhibiting or neutralizing the activity of herbicides :such as: the sfr gene and the sfrv gene encoding enzymes conferring resistance to glutamine synthetase inhibitors such as Bialaphos and phosphinotricine as described in publication of EP 242 246.
In order for t:he protein or polypeptide encoded by the marker DNA to function as intended, it is often preferred to have it produced in the plant cell as a precursor, in which the mature protein is linked at its N-terminal end to another polypepti.de (a "targeting peptide") which will translocate the mature protein to a specific compartment such as the chloroplasts, the mitochondria, or the endoplasmic reticulum. Such targeting peptides and DNA
sequences coding for them (thc~ "targeting sequences") are well known. For example, if a marker DNA codes for a protein that confers tolerance or resistance to a herbicide or another selective agE_nt that acts on chloroplast metabolism, such as the sfr (or bar) gene or the sfrv gene _7-(European patent publication ("EP") 0,,242,236), it may be preferred that such gerze also comprise a chloroplast targeting sequence sucru as that coding for the transit peptide of the small s~.abunit of the enzyme 1,5-ribulose bisphosphate carboxyla~~e (Krebbers et al (1988) Plant Mol.
Biol. 11, 745; publication of EP 189 '707), although other targeting sequences coding for other transit peptides, such as those listed by Von Reijne et al (1991) Plant Mol. Biol.
Reporter 9 , 104 , can b~~s used .
Each of the stamen-specific, preferably anther-specific, promoters of t=.his invention, such as the CA55 promoter upstream from nucleotide 1180 in SEQ ID no. 3, which can be used to control t:he male-sterility DNA or the male fertility-restorer DNA, c:an be identified and isolated in a well known manner as described in publication of EP 344 029. In this regard, each of the SEQ ID no. 1 and no. 2 cDNAs of this imvE~ntion can be used as a probe to identify (i.e., to hybr_~dize too) the corresponding region of the corn genome (i.e., t:he region containing DNA coding for the stamen-specific mRNA, from which the cDNA was made).
Then, the portion of the. plant genome that is upstream (i.e., 5') from the DNA coding for such stamen-specific mRNA
and that contains the p:romot e:r of th:i s DNA. can be identified. For instance, the cDNA c~f SEQ ID no. 2 can be 25, used as a probe to identify and isolate a genomic clone from a genomic library of Zea mat's, such as a lambda EMBL3 or EMBL4 Zea m-ays genomic: library. In this way, a genomic DNA
clone can be isolated a:nd sequenced, such as the clone of SEQ ID no. 3. SEQ ID n~~. 3 contains a coding region which 3C> is homologous to the cD:NA of SEQ ID no. 2, and upstream of this coding region is a promoter sequence, with a TATA box, which directs the anther-specific transcription of the coding region.

_.g-The second p.rc>moter, which controls the marker DNA, can also be selected and isolated in a well known manner, for example as described in publication of EP 344 029, so that the marker DNA is expressed either selectively in one or more specifics tissues or cells or constitutively in the entire plant, as desired, depending on the nature of the RNA and/or protein or polypeptide encoded by the marker DNA.
In the foreign DNA sequence of this invention, 3' transcription termination signals or the "3' end" can be selected from among those which are capable of providing correct transcription ~~:rminat:ion and/or polyadenylation of mRNA in plant cells. The transcription termination signals can be the natural ones of the male-sterility or male fertility-restorer DNA, to be transcribed, or can be foreign or heterologous. Examples of heterologous 3' transcription termination signals are those of the octopine synthase gene (Gielen et al (1984) EMBO J. :3, 835-845) and of the T-DNA
gene 7 (Velten and Sch.e:L:L (1985) NAR 13, 6981-6998) . When the foreign DNA sequence= of this invention comprises more than one structural gene (e.g., a male-sterility DNA or a fertility-restorer DNP.<~nd a marker DNA), it is preferred that the 3' ends of the structural genes be different.
In plants, especially in monocotyledonous plants, 2c~ particularly cereals such as rice, cc>rn and wheat, the expression in accordance with this invention of a marker DNA, as well as a male-sterility DNA or a fertility-restorer DNA, can be enhanced f>y the presence at one or more, preferably one, appropriate positionfs) in the transcriptional unit c:~f each foreign DNA sequence of this invention, of a suitar>le plant intron (Luehrsen and Walbot (1991) Mol. Gen. Genet. 225, 81; Mascarenhas et al (1990) Plant Mol. Biol. 15, ~~13; Vasil et, a7. (1989) Plant Phys:iol.

WO 92/13957 PC'T/EP92/00275 2~.0~;~7;~

91, 1575f Tanaka et al (1990) NAR _1.8, 6767; McElroy et al (1990) The Plant Cell 2, 163; PCT publication WO 91/09948).
Preferably, each intron has a nucleotide sequence that: is recognizable by the cells of the plant species being transformed (for requirements of intron recognition by plants, see Goodall and Filipowicz (1989) Cell _58, 473;
Hanley and Schuler (1988) NAR 16, 7159), is longer than about 70-73 by (Goodall and Filipowicz (1990) Plant Mol.
Biol. 14, 727), and is positioned close to the 5' end of the encoded mRNA, particularly in any untranslated leader sequence.
Cells of a plant can be transformed with the foreign DNA sequence of this invention in a conventional manner.
Where the plant to be transformed is susceptible to Aqrobacterium infection, it is preferred to use a vector, containing the foreign DNA sequence, which is a disarmed Ti-plasmid. The transformation can be carried aut using procedures described, for example, in EP 0,116,718 and EP 0,270,822. Preferred Ti-plasmid vectors contain the foreign DNA sequence between the border sequences or at least located upstream of the right border sequence. Of course, other types of vectors can be used for transforming the plant cell, using procedures such as direct gene transfer (as described for example in EP 0,223,247), pollen mediated transformation (as described for example in EP
0,270,356, PCT publication WO/85/01856 and EP 0,275,069), in vitro protoplast transformation (as described for example in US patent 4,684,611), plant virus-mediated transformation (as described for example in EP 0,067,553 and US patent 4,407,956) and liposome-mediated transformation (as described for example in US patent 4,536,475). .
SUBSTITUTE SHEET

1,..:,...
s.::.:
i ~~y°,. C ~~ 10 N ~' Where the plant to be transformed is corn, recently developed transformation methods can be used such as the methods described for certain lines of corn by Fromm et al (1990) Bio/Technology 8, 833 and Cordon-Kamm et al (1990) The Plant Cell 2, 603.
Where the plant to be transformed is rice, recently developed transformation methods can be used such as the methods described for certain lines of rice by Shimamoto et al (1990) Nature 338, 274, Datta et al (1990) Bio/Technology 8, 736, Christou et al (1991) Bio/Technology 9, 957 and Lee et al (1991) PNAS 88, 6389.
Where the plant to be transformed is wheat, a method analogous to those described above for corn or rice can be used. Preferably for the transformation of a monocotyledonous plant, particularly a cereal such as rice, corn or wheat, a method of direct DNA transfer, such as a method of biolistic transformation or electroporation, is used. When using such a direct transfer method, it is preferred to minimize the DNA that is transferred so that essentially only the foreign DNA sequence of this invention, with its male-sterility DNA, fertility-restorer DNA and/or marker DNA, is integrated into the plant genome.
In this regard, when a foreign DNA sequence of this invention is constructed and multiplied on a plasmid in a bacterial host organism, it is preferred that, prior to transformation of a plant with the foreign DNA sequence, plasmid sequences that are required for propagation in the bacterial host organism, such as an origin of replication, an antibiotic resistance gene for selection of the host organism, etc., be separated from the parts of the plasmid that contain the foreign DNA sequence.
The Examples, which follow, describe: the isolation and the characterization of the two corn cDNA sequences SEQ

f.:" '° c. :.a c ID no. 1 and no. 2 of this invention; their use for isolating the two stamen-specific promoters of this invention from the corn genome, such as the CA55 promoter upstream from nucleotide 1180 in SEQ ID no. 3; the construction of promoter cassettes for the fusion of the promoters with male-fertility and male fertility-restorer DNAs: the construction of plant transformation vectors from the promoter cassettes: as well as the transformation of corn, rice and tobacco with the resulting plant transformation vectors.
Unless stated otherwise in the Examples, all procedures for making and manipulating recombinant DNA were carried out by the standard procedures described in Maniatis et al, Molecular Cloning - A Laboratory Manual, Cold Spring Harbor Laboratory Press, NY (1982) and Sambrook et al, Molecular Cloning - A Laboratory Manual Second Edition, Cold Spring Harbor Laboratory Press, NY (1989).
When making plasmid constructions, the orientation and integrity of cloned fragments were checked by means of restriction mapping and/or sequencing.
The sequence identification numbers referred to above and in the Examples are listed below.
Seguence Listing SEQ ID no. 1: cDNA sequence of the CA444 gene.
SEQ ID no. 2: cDNA sequence of the CA455 gene.
SEQ ID no. 3: genomic DNA clone obtained from a _Zea mars genomic library using the cDNA of SEQ ID no. 2 as a probe.
SEQ ID no. 4: intron from TA36 gene of Nicotiana tabacum linked to KpnI linkers SUBSTITUTE SHEET

WO 92/13957 PC1'/EP92/00275 ,.:..

,~ s SEQ ID no. 5: sequence of plasmid pVE149.
Example 1 Isolation and characterization of anther-specific cDNAs from corn For the cloning of cDNAs corresponding to genes which are expressed exclusively, or at least predominantly, in anthers of corn, a cDNA library was prepared from poly A' mRNA isolated from tassel spikelets from the publicly available corn line B73 bearing anthers in tetrad stage. By means of the Amersham cDNA Synthesis System Plus RPN 1256 Y/Z kit (Amersham International PLC, Buckinghamshire, England), cDNA was synthesized using reverse transariptase and an oligo dT primer according to the directions set forth in the kit for its use.
The cDNAs were cloned in lambda gti0 vector, using the Amersham cDNA Cloning System - lambda gtl0 - RPN1257 - kit, ' in accordance with the directions set forth in the kit for its use. From the cDNA library thus obtained (30,000 plaques), differential screening was performed with a labelled cDNA probe from corn _B73 seedlings and with a labeled cDNA probe from corn _B73 whole spikelets. 93 possible anther-specific cDNA clones were selected and again screened with labeled cDNA probes from corn _B73 anthers, seedlings and ears. With the 66 remaining clones from these additional selections, a Southern analysis was performed with differential cDNA probes from anthers at tetrad stage and from opened tassels, silk and ears from the corn line B73. This led to the selection of 2? anther-specific clones which were subcloned in pGEMl (Promega, Madison, Wisconsin, USA). Cross-hybridization between these subclones revealed the presence of at least 2 classes.
Probes of some of these subclones were prepared and checked again for their specificity in Northern blots with 5 to 10 SUBSTITUTE SHEET

r.

~g poly Aa mRNA isolated from different corn _B73 tissues (i.e., anthers, ears, silk, leaves, and spikelets at several stages). From this selectian, two anther--specific clones, called "pCA444" and "pCA455", were identified.
These clones were sequenced, and their sequences are shown in the sequence listing as SEQ ID no. 1 and SEQ ID no. 2, respectively. pCA455 was found to hybridize exclusively with mRNA from anthers in different stages of development.
pCA444 was found to hybridize with mRNA from anthers and to hybridize very weakly with a mRNA of similar size from embryos.
The cDNA sequence of pCA444 reveals the presence of two open reading frames ("ORF") over a total of 323 and 376 nucleotides. The cDNA sequence of pCA455 reveals the presence of two ORFs over a total of 387 and 300 nucleotides.
Example 2 Isolation of the anther-specific Qene corres ondin~ to the anther-specific cDNA clone, pCA444, of Example 1 To isolate the genomic DNA clones carrying the regulatory sequences of the gene, CA444, corresponding to pCA444, two approaches are taken.
The first approach uses inverse polymerase chain reactions ("PCR") (Ochman et al (1989) in "PCR: Application & Protocols", Innis, M., Gelfand, D., Sninsky, J., and White, T., eds. Academic Press, New York) for the geometric amplification of the DNA sequences which flank, upstream and downstream, a chosen core region of the CA444 gene sequence corresponding to the sequence of pCA444. DNA
digestions are carried out using conventional buffers and well known conditions. Fragments of a suitable size (less than 3 to 4 kb) for correct amplification and circularization are produced by using restriction enzymes SUBSTITUTE SHEET

.._.

which do not cleave the chosen core region of the CA444 gene sequence and which are preliminarily identified by Southern hybridization. Circularization is performed with T4 DNA ligase in a dilute DNA concentration favoring monomeric circles (Collins and Weissman (1984) PNAS _81, 6812-6815). Three polymerise chain reactions are performed in parallel with three different oligonucleotide pairs under conventional conditions (Saiki et al (1985) Science 230, 1250-1354) using the VentT" DNA polymerise (Catalog no. 254L - Biolabs New England, Beverly, MA 01915, U.S.A.) isolated from Thermococcus litoralis (Neuner et al (1990) Arch. Micobiol. 153, 205-207).
In one reaction, the flanking regions of the core region of CA444, from nucleotide 85 to nucleotide 358 of the corresponding cDNA sequence (SEQ ID no. 1), are amplified using the following pair of 22 and 20 oligonucleotides having the following respective ss~quences:
1) 5' CCG AGG ACC AGC AGG ACG AGG C 3' (nucleotide 64 to nucleotide 85 of pCA444 (SEQ ID no. 1)) and 2) 5' GGA TGG CAG GAG GGG AGA GG 3' (nucleotide 358 to nucleotide 377 of pCA444 (SEQ ID no. 1)).
In the second reaction, the flanking regions of the core region of CA444, from nucleotide 288 to nucleotide 392 of the corresponding cDNA sequence (SEQ ID no. 1), are amplified using the following pair of 20 and 23 oligonueleotides having the following respective sequences:
1) 5' GCA GGC TGT TGA TGA TGC CC 3' (nucleotide 269 to nucleotide 288 of pCA444 (SEQ ID no. 1)) and 2) 5' CCA TTT CAC AGT GAG AGC AGT CG 3' (nucleotide 392 to nucleotide 414 of pCA444 (SEQ ID no. 1)).
SUBSTITUTE SHEET

In the third reaction, the flanking regions of the core region of CA444, from nucleotide 43 to nucleotide 74 of the corresponding cDNA sequence (SEQ ID no. 1), are amplified using the following pair of 22 and 20 oligonucleotides having the following respective sequences:
1) 5' GGG GCG GTG GCfi GCT TCT AGC G 3' (nucleotide 22 to nucleotide 43 of pCA444 (SEQ ID no. 1)) and 2) 5' GCT GGT CCT CGG CGG CGG CA 3' (nucleotide 74 to nucleotide 93 of pCA444 (SEQ ID no. 1)).
The second approach uses a lambda EMBL3 or EMBL4 Zea mays genomic library that is screened with the whole cDNA
sequence of pCA444 as a probe. Corresponding genomic clones which hybridize to pCA444 are sequenced (Maxam and Gilbert (1977) PNAS 74, 560) and their orientation checked by Northern blot analysis with riboprobes of both senses.
Comparison of the sequences of pCA444 with the .genomic clone sequences leads to the identification of the homologous regions. At the 5' end of the region of each of these homologous genomic clones, the ATG codon and the consensus sequence TATA are determined. That the "TATA"-box is part of the promoter is confirmed by primer extension.
Example 3 Isolation of the gather-specific Qene oorrespondinQ to the anther-specific oDNA clone, pCA455, of Exam 1e 1 To isolate the genomic DNA clones carrying the regulatory sequences of the gene, CA455, corresponding to pCA455, the two approaches of Example 2 are used.
In the first approach, inverse PCR (Ochman et al, 1989) is used for the geometric amplification of the DNA
sequences which flank a chosen core region of the CA455 gene sequence corresponding to the sequence of pCA455. DNA
digestion and circularization are carried out as in Example SUBSTITUTE SHEET

cs , 16 _.
2. Two polymerase chain reactions are performed in parallel with two different oligonucleotide pairs under conventional conditions (Saiki et al, 1985) using the ventT" DNA
polymerase isolated from T. litoralis (Neuner et al, 1990).
In one reaction, the flanking regions of the core region of CA455, from nucleotide 54 to nucleotide 87 of the corresponding cDNA sequence (SEQ ID no. 2), are amplified using the following pair of 21 and 23 oligonucleotides having the following respective sequences:
1) 5' GCT CGA TGT ATG CAG TGC AGC 3' (nucleotide 34 to nucleotide 54 of pCA455 (SEQ ID no. 2)) and 2) 5' CGT CGC CGT GTC GGT GCT TCT CG 3' (nucleotide 87 to nucleotide 109 of pCA455 (SEQ ID no. 2)).
In the second reaction, the flanking regions of the core region of CA455, from nucleotide 54 to nucleotide 557 of the corresponding cDNA sequence (SEQ ID no. 2), are amplified using the following pair of 21 and 24 oligonucleotides having the following respective sequences:
1) 5' GCT CGA TGT ATG CAG TGC AGC 3' (nucleotide 34 to nucleotide 54 of pCA455 (SEQ ID no. 2)) and 2) 5' CCG TTG CGT TGC GTT GCG TAG ACG 3' (nucleotide 557 to nucleotide 580 of pCA455 (SEQ ID no. 2)).
The second approach uses a lambda EMBL3 or EMBL4 _Zea mat's genomic library that is screened with the whole cDNA
sequence of pCA455 as a probe. Corresponding genomic clones which hybridize to pCA455 are sequenced (Maxam and Gilbert, 1977), and their orientation is checked by Northern blot analysis with riboprobes of both senses. Comparison of pCA455 with the genomic clone sequences leads to the identification of the homologous regions. At the 5' end of the region of each of these homologous genomic clones, the SUBSTITUTE SHEET

ATG codon and the consensus sequence TATA are determined.
That the "TATA"-box is part of the promoter is confirmed by primer extension.
Using this second approach, an existing lambda EMBL4 Zea mans genomic library was screened with the whole cDNA
sequence of pCA455, as a probe. The library was obtained from Dr. H. Saedler of the Max Planck Institute in Koln, Germany, with the designation "GH#1417". The library comprised Zea mans genomic DNA which was partially digested with MboI and the resulting restriction fragments of which were cloned between the BamHI sites of the bacteriophage lambda EMBL4 replacement vectors (Frischauff et al (1983) J. Mol. Biol. 170, 827; Pouwels et al (1988) ClonincT
vectors - a Laboratory Manual (supplementary update), Elsevier Science Publishers, Amsterdam). The restriction fragments of the library could be excised from the vectors as EcoRI fragments.
One EcoRI fragment of about 6 kb in length from the library was found to hybridize with pCA455 and was called "VG55". VG55 was found to contain an unique BamHI site, and one of the EcoRI-BamHI fragments of VG55 still hybridized with pCA455 while the other did not. The _EcoRI-_BamHI
fragment that cross-hybridized with pCA455 was cloned between the EcoRI and BamHI sites of vector pGEMl (Promega), yielding a plasmid called "pVG55.3". pVG55.3 was sequenced (Maxam and Gilbert, 1977), and its orientation was checked by Northern blot analysis with riboprobes of both senses. The sequence of pVG55.3, apart from some nucleotides at its 5' end (which includes its _EcoRI site), is shown in SEQ ID no. 3 as having a high degree of homology with pCA455. The ATG codon of the presumed coding sequence of pVG55.3 is located at position 1180, the presumed coding sequence ends at position 1596, and the SUBSTITUTE SHEET
S

"TATA"-box is located at: posit:ion 1072. That the "TATA"-box is part of the promoter is confirmed by primer extension.
The unique BamHI site, mentioned above, is located at position 2770 in SEQ ID no. 3..
The sequence upstream from position 1180 in SEQ ID
no. 3 can be used as a promoter region for the anther-specific expression of d COd7.Ilg sequence of interest. This sequence is the CA55 promoter or PCA55. Preferably, the complete sequence from position 1 to position 1179 is used, but it appears that the minimum region which can serve as an anther-specific promoter extends about 300 to 500 by upstream from position :L:180 in SEQ ID no. 3. The use of the untranslated leader sequence :in the PCA55 promoter region, between the transcription initiation site (which can be determined by means of primer extension; and the ATG start of translation, appears to be preferred but not essential for anther-specific expression of a h.eterologous structural gene under the control of the PCA55 promoter, and the leader sequence apparently ca.n be replaced by the untranslated 2C leader sequence of other genes, such as plant genes.
Example 4 Construction of promoter cassettes derived from the anther-specific genes of Examples 2 and 3 The 5' regulatory sequences, including the promoter, of each of t:he anther-s~>eci_fic genes of Examples 2 and 3 are subcloned into the polylinker of pMa5-8 and pMc5-8 (publication of EP 315 353). This produces vectors which can be used to isolate single stranded DNA for use in site-directed mutagenesis. Using site-~diz-ected mutagenesis (publication of EP 319 353) sequences surrounding the ATG
translation initiation codon of tree >' regulatory sequences of each of the anther--specific genes are modified to create _1g_.
a unique recognition s.it:e for a restriction enzyme for which there is a corresponding recognition site at the 5' end of each of the male-steri.li.ty and male fertility-restorer DNAs (that are to be fused to the 5' regulatory sequences in Example 5, below). The resulting plasmids each contain the newly created restriction site. The precise nucleotide sequence spanning each newly created restriction site is determined in order to c:onfirrn that ~t only differs from the 5' regulatory sequences of the corresponding corn anther-specific gene by the su~~stitution, creating the new restriction site.
In using this procedure for constructing promoter cassettes, a NcoI site :Ls introduced at the ATG translation initiation codon of pVGp5.3 of Example 3 as follows. A 1280 by EcoRI-AvaI fragment of pVG55.3 (th.e AvaI site is located at position 1276 of SEQ ID no. 3; the EcoRI site is derived from pGEMl (promega) and is located at the 5' end of SEQ ID
no. 3 [not shown]) is c:Loned between the EcoRI and AvaI
sites of the vectors pMa5-8 and pMc5-8 (Stanssens et al (1989) NAR 17, 4441; publication of E~P 319 353) yielding plasmids called "pMa5-VC355.3" and "pNIcS-VG55.3"
respectively. These plasmids are used for site-directed mutagenesis by a gapped duplex DNA method using alternating selectable markers as described by St.anssens et al (1989) supra. The gapped duplex DNA is constructed from the single stranded pMc5-VG55.3 and the large: EcoRI-AvaI fragment of pMaS-VG55.3. For mutagenesis, use i~o made of the oligonucleotide with t:he following sequence:
CAG GAci CGA GCC ATG GCT GCA G.
This mutager~esis introduces the NcoI site at the ATG codon of the codirag sequence of ~3EQ ID no. 3. The resulting cassette comprises the promoter and leader sequence of SEQ ID no. _ in a EcoRI-NcoI fragment that is to be fused to the coding ~Eequenc:es of male-sterility and male fertility-restorer DNAs as described in Example 5, below.
Alternatively, the NcoI site is introduced at the ATG translation initiation colon of pVG55.3 during amplification by PCR o:f a DNA fragment, containing the C'.A55 promoter region, using t:he following two oligonucleotides as primers:
5'-GAT TCG AAT TCT GGT ATG CAT CAA TAG AGC CG-3' 5' -CAG GAG C(J~~ GCC ATG GCT GCA G-3' The amplified DNA fragment is used directly as a promoter cassette for constructixig plant transformation vectors as described in Example 5.
Example 5 Construction of plant transformation vectors from the promoter cassettes of Example 4 Using the procedures described in EP 344 029 and 412 911, the promoter cassettes of Example 4 are used to construct plant transformation vectors comprising foreign chimaeric DNA sequences of this invention, each of which contains the 5' regulatory sequences, including the anther-specific promoter, of oTle Of the anther-specific genes isolated in Example 2 or 3. Each of these 5' regulatory sequences is upstream o:~, is :in the same transcriptional unit as, and controls e=ither a male-sterility DNA (from EP
344 029) encoding barrAase from Bacillus a~loliquefaciens (Hartley and Rogerson (:1972) Preparative Biochemistry 2 (3), 243-250) or a male fertility-restorer DNA (from EP 412 911) encoding barstar (Hartlf=y and Rogersc~n (1972) supra; Hartley and Sweaton (1973) J. B:iol. Chem. 248 (16), 5624-5626).

Downstream of each male-sterility or male fertility-restorer DNA is the 3' end of tine nopaline synt~hase gene (An et al (1985) EMBO J. 4 (2), 277). Each chimaeric DNA sequence also comprises the 35 :3'3 promoter (Hull and Howell (1987) Virology 86, 482-493) ::used in frame with the neo gene encoding kanamycin resistance (EP 131 623) and the 3' end of the octopine synthase gene (Dhaese et al (1983) EMBO
J. 2, 419) .
Alternatively, the plant transformation vector pVE149, pVE139 and pVEl36 are constru~~ted as follows.
In a first step, the 1083 by EcoRI-HindIII DNA
fragment of pMT416, cont,aininc~ the barnase and barstar coding sequences (Hart:i.F>y (1988) J. Mol. B.iol. 203, 913), is ligated to the large EcoRI-HindIII fragment of plasmid pMa5-8, yielding plasmid pMa~~tpbsl., By means of site-directed mutagenesis (PCT publication WO 89/03887), a NcoI site is then introduced at the ATG t.ranslatian initiation codon of the barnase coding sequ~:nce. For this purpose, a gapped duplex DNA is constructed from the single stranded pMa5tpbsl, the large EcoRI-HindIII fragment of pMc5-8, and the following oligonuc:Leotide:
5'-GAT AAC CGG '1.'AC CAT GGT TGT CAC AGG GG-3'.
The resulting plasmid is designated as "pVE145A".
In a subsequent mutagenesis round, a Nsil site is introduced 14 by downstream of the ATG translation initiation codon of the barna~se coding sequence using a gapped duplex DNA consisting of single stranded DNA from pVE145A, the large EcoR:I-HindIII fragment of pMa5-8, and the following oligonucleot.ide 5'-CCC CGT C:AA ATG CAT TGA TAA CCG G-3' -21a-The resulting plasmid :~s designated as "pVE145"
Plasmids pMa.':>--8 and pMc5-8 have been deposited on May 3, 1988 at the Deutsche Sammlung fur Mikroorganismen and Zellkulturen (DSM) , Ma~rherodE~rweg 1B, D-330 Braunschweig, Germany under accession numbers DSM 4567 and DSM 4566 respectively.
pVE145 is cut with NsiI, filled in with Klenow, and ligated to the 111 by DNA fragment shown in SEQ ID no. 4 which contains the TA36 intron and which has been cleaved with KpnI and made blunt-ended with Klenow, yielding plasmid pVE146. The fragment of SEQ ID no. 4 is obtained by amplification, from genomic DNA of Nicotiana tabacum cv.
Samsun, by means of PCR using the following two oligonucleotides as primers 5'-CGA CGG TAC CAC GTA ATT AG-3' 5'-CAT AGG GTA CCT GTA TGT AAT ~.AA AAC-3'.
Plasmid pVE147 is constructed by ligation of the 2820 by EcoRI-HindIII fragment of pGEMl (Promega), the 979 by HindIII-NcoI fragment of pVE146 (carrying the barnase gene with intron), and the 1184 by PCR fragment of Example 4 carrying the CA55 anther-specific promoter from corn.
Finally, pVE149 is obtained by ligation of the following four DNA fragments:
- the 1715 by EcoRI (filled-in with Klenow) - XbaI
fragment of pVE147, carrying the barnase gene under control of the CA55 promoter, - a 296 by EcoRI-XbaI fragment of pTTM6 (deposited on March 7, 1988 at the DSM under DSM accession number 4468), carrying the 3' untranslated end of the nopaline synthase gene of Aqrobacterium T-DNA, - a 1728 by BglII (filled-in with Klenow)-HindIII
fragment, carrying the bar gene (EP 0,242,236) under the control of the 3553 promoter (EP 0,359,617) and with a 3' untranslated end of the nopaline synthase gene of Aqrobacterium T-DNA (this fragment corresponds to the sequence in SEQ ID no. 5 between positions 2409 and 4137), and SUBSTITUTE SHEET

1 f~ J
r the large EcoRI-HindIII fragment of pUCl9 (New England Biolabs Inc. Beverly, MA, U.S.A.).
The complete sequence of pVE149 is shown in SEQ ID no. 5.
Plasmid pVE136 is identical to pVE149 except it lacks the TA36 intron in the barnase gene. pVE136 is constructed by replacing, in pVE149, the 534 by NcoI-BamHI fragment, carrying the barnase gene, with the 449 by NcoI-BamHI
fragment of pVE145A.
pVE136 is constructed and maintained in E. coli WK6 containing the plasmid pMcS-BS. pMcS-BS contains the barstar gene under the control of the tac promoter (De Boer et al (1983) PNAS 80, 21) and is constructed by cloning the EcoRI-HindIII fragment of pMT416 (Hartley (1988) J.Mol.Biol. 202, 913) into pMcS-8. Then, the sequence, starting with the PhoA signal sequence and ending with the last nucleotide before the translation initiation codon of the barstar coding region, is deleted by loaping-out mutagenesis according to the general procedures described by Sollazi et al (1985) Gene 37, 199. The availability of an ampicillin resistance gene on the pUClB-derived plasmids carrying the chimaeric barnase gene and the chloramphenicol resistance gene on pMcS-BS permits the strain to be kept stable on plates provided with two antibiotics or to select for any one plasmid. While normally repressed, gene expression from this promoter can be induced by addition of a commonly used inducer of the lac operon, IPTG
(isopropyl-,B-d- thiogalactopyranoside).
The 5843 by NcoI-BamHI fragment of partially digested -pVE149, carrying all of the plasmid except the barnase coding sequence, is filled in with Klenow and ligated to a DraI-HindIII fragment (filled- in with Klenow) of pVE151, carrying the barstar coding sequence. The resulting plasmid is designated as "pVE139" . pVE151 is obtained by means of SUBSTITUTE SHEET

..;r-.
,Z r~~~~ ~ 24 site-directed mutagenesis of pMcS-BS, so that a DraI site is introduced at the ATG translation initiation codon of the barstar coding sequence. For this purpose, a gapped duplex DNA is constructed from the single stranded pMcS-BS, the large EcoRI-HindIII fragment of pMa5-8, and the following oligonucleotide:
5'-GCT TTT TTA AAT TTA TTT TCT CC-3'.
T-DNA vectors for Agrobacterium-mediated plant transformations are prepared by cloning the appropriate EcoRI (filled-in with Klenow)-HindIII fragments of pVE149 or pVE136 (containing the 3553-bar and corn anther-specific promoter-barnase chimaeric genes) or pVE139 (containing the 3553-bar and corn anther-specific promoter-barstar chimaeric genes) between the HindIII and XbaI (filled-in with Klenow) sites of the known T-DNA vectors pGSC1700 or pGSC1701A. pGSC1700 has been deposited on March 21, 1988 at the DSM under DSM accession number 4469, and pGSC1701A has been deposited on October 22, 1987 at the DSM under DSM
accession number 4286. The T-DNA vectors containing pVE149, pVE139 and pVE136 are used for transformation of tobacco as described in Example 7.
Example 6 Transformation of corn withthe plant transformation vectors from Egamule 5.
Using the procedures described by Fromm et al (1990) supra, embryogenic suspension cultures of a _B73 X A188 corn line are transformed with the plant transformation vectors described in Example 5, including pVE149, pVE136 and pVE139 -- either directly or after suitable linearization (e.g., ' after digestion winch EcoRI and/or HindIII). Transformed plants regenerated from the embryogenic suspension cultures, each containing an anther-specific promoter of Example 2 or 3 controlling either a male-sterility DNA or a SUBSTITUTE SHEET

_25-male fertility-restorer DNA, are normal except for their flowers. In this rega,~ci, each plant containing a male-sterility DNA under tht~ control of one of the anther-specific promoters expresses such DNA at least predominantly in its anthers and pro~:~uc;es no normal. :pollen, and each plant containing a male fertility-restorer DNA under the control of one of the anther-spE~c;ific promoters expresses such DNA
at least predominantly :i.n its anthers but produces normal pollen.
Example 7 Transformation of tobacco with the plant transformation vectors from Example 5 Using the pros;edures described in publication of EP 344 029 and 412 911, tobacco plants are transformed by Agrobacterium-mediated transfer with the plant transformation vectors containing the foreign chimaeric DNA
sequences from Example >. The transformed tobacco plants, each containing an ant:lzer-specific promoter of Example 2 or 3 controlling either a nnale-sterility DNA or a male fertility-restorer DNA, are normal except for their flowers.
In this regard, each p:L~~nt containing a male-sterility DNA
under the control of 011E' of the anther-specific promoters expresses such DNA at :lE:ast predominantly in its anthers and produces no normal pollen, and each plant containing a male fertility-restorer DNA under the control of one of the anther-specific promoteos expresses such DNA at least predominantly in its anthers but produces normal pollen.

-25a-Example $
Transformation of rice with the plant transformation vectors from Example 5 Using the procedures described by Datta et al (1990) supra, protopla;at.s of t;he rice line, Oryza sativa var.

PCf/EP92/00275 r~'~

Chinsurah Boro II, are transformed with the plant transformation vectors described in Example 5, including pVE149, pVE136 and pVE139 -- either directly or after suitable linearization (e. g., after digestion with EcoRI
and/or HindIII). Transformed plants regenerated from the protoplasts, each containing an anther-specific promoter of Example 2 or 3 controlling either a male-sterility DNA or a male fertility-restorer DNA, are normal except for their flowers. In this regard, each plant containing a male-sterility DNA under the control of the anther-specific promoters expresses such DNA at least predominantly in its anthers and produces no normal pollen, and each plant containing a male fertility-restorer DNA under the control of the anther-specific promoter expresses such DNA at least predominantly in its anthers but produces normal pollen.
Alternatively, immature embryos from rice varieties Gulfmont, Lemont, IR26, IR36, IR54, or IR72 are bombarded with gold particles, carrying appropriate plasmid DNA of Examples 5, and plants are regenerated according to the procedures described by Christou et al (1991) Bio/Technology 9, 957.
Needless to say, the use of the anther-specific corn promoters of this invention is not limited to the transformation of any specific plant(s), Such corn promoters can be useful in any crop where they are capable of controlling gene expression, and preferably where such expression occurs at least predominantly, preferably specifically, in stamen cells of the crop. Also, the use of such promoters is not limited to the control of male-sterility DNAs or male fertility-restorer DNAs but can be used to control the expression of any gene selectively in stamen cells.
SUBSTITUTE SHEET

~'O 92/13957 c T ~~ c'3 .'.',.;.. N 1 ~~~~<: ; ~~
i .. 27 Furthermore, this invention is not limited to the specific stamen-specific, preferably anther-specific, particularly tapetum-specific, promoters described in the foregoing Examples. Rather, this invention encompasses promoters equivalent to those of Examples 2 and 3 which can be used to control the expression of a structural gene, such as a male-sterility DNA or a male fertility-restorer i DNA, selectively in stamen cells, preferably anther cells, particularly tapetum cells, of a plant. Indeed it is believed that the DNA sequences of the promoters of Examples 2 and 3 can be modified by replacing some of their nucleotides with other nucleotides, provided that such modifications do not alter substantially the ability of polymerase complexes, including transcription activators, of stamen cells, particularly anther cells, to recognize the promoters, as modified.
SUBSTITUTE SHEET

~t:.._:.s.
._ 2g :~ ~N
SEQUENCE LISTING
1. General Information i) APPLICANT : PLANT GENETIC SYSTEMS N.V.
ii) TITLE OF INVENTION : stamen-specific promoters from corn iii) NUMBER OF SEQUENCES : 5 SEQ. ID. NO 1 : cDNA CA444 SEQ. ID. NO 2 : cDNA CA455 SEQ. ID. NO 3 : genomic sequence from corn comprising the CA55 promoter SEQ. ID. No 4 : TA36 intron SEQ. ID. NO 5 : plasmid pVE149.
iv) CORRESPONDENCE ADDRESS
A. ADDRESSEE : Plant Genetic Systems N.V.
B. STREET : Plateaustraat 22, C. POSTAL CODE AND CITY : 9000 Ghent, D. COUNTRY : Belgium v) COMPUTER READABLE FORM
A. MEDIUM TYPE 5.25 inch, double sided, high density 1.2 Mb floppy disk ,; B. COMPUTER : IBM PC/AT
C. OPERATING SYSTEM : DOS version 3.3 D. SOFTWARE : WordPerfect 5.1 vi) CURRENT APPLICATION DATA : Not Available (vii) PRIOR APPLICATION DATA
EPA 91400300.9, filed February 7, 1991 EPA 91401787.6, filed June 28, 1991 :~, ~ ~ ..f ~"

2. Sequence Description : SEQ ID NO. 1 SEQUENCE TYPE: nucleotide SEQUENCE LENGTH: 533 STRANDEDNESS: double°stranded TOPOLOGY: linear MOLECULAR TYPE: cDNA to mRNA
ORIGINAL SOURCE
ORGANISM: Corn ORGAN: anther FEATURES: from nucleotide 2 TO nucleotide 376 : Open Reading Frame 1 from nucleotide 3 TO nucleotide 326 : open Reading Frame 2 PROPERTIES: anther specific cDNA
MISCELLANEOUS : cDNA designated as CA444 SUBSTITUTE SHEET

3. Sequence Description : SEQ ID NO. 2 SEQUENCE TYPE: nucleotide SEQUENCE LENGTH: 796 STRANDEDNESS: double-stranded TOPOLOGY: linear MOLECULAR TYPE: cDNA to mRNA
ORIGINAL SOURCE
ORGANISM: corn ORGAN: anther FEATURES: from nucleotide 2 TO nucleotide 388 : Open Reading Frame 1 from nucleotide 70 TO nucleotide 369 : Open Reading Frame 2 PROPERTIES: anther specific cDNA
MISCELLANEOUS : cDNA designated as CA455 AATGCTAGTTCAATAAAAAAAAAAAAAAAA A,~~~AAAAAAA750 f~~A A,~u~AAAAAAACCATGGTACCCGGATC 796 SUBSTITUTE SHEET

2~~a;~'7fo ..
s::~.;

4. Sequence Description : SEQ ID NO. 3 SEQUENCE TYPE: nucleic acid SEQUENCE LENGTH: 2784 basepairs STRANDEDNESS: double stranded TOPOLOGY: linear MOLECULAR TYPE: genomic DNA
ORIGINAL SOURCE: corn ORGANISM: anther FEATURES: - nucleotide 1 to nucleotide 1179 : region comprising the promoter and the leader sequence - nucleotide 1072 : TATA Box - nucleotide 1180 to nucleotide 1596 : presumed coding sequence - nucleotide 1120 to nucleotide 11839 : region corresponding to cDNA of SEQ ID. 2 PROPERTIES: genomic DNA obtained by probing a Zea mat's genomic library using the cDNA of SEQ ID 2 as a probe. The genomic DNA is designated as pVG55.3.

GTh~AAAAAAACTGTTGTCACATTTGCCTTCGCTGTGACTTGGATGTATCA 150 AGTGTTTCA'fTGGACGAAGGTCCAAGTCCTTCAGATCATCTCAATTTTCT 800 ATG TCT
CGC

MET Ser Ser Cys Arg SUBSTITUTE SHEET

';.:

GTG GTG

Cys Val Ala Ser LeuLeu AlaValAla Ala Thr Val Val ACC CCG

Ala Ser Ala Ala AlaTrp LeuHisGlu Glu Gln Thr Pro GAG ATG

His Leu Glu Ala AlaThr GlyProLeu Val Ala Glu MET

AGG GCG

Glu Gly Ala Val ProSer AlaSerThr Trp Ala Arg Ala GCG CCG

Ala Asp Lys Ser AlaArg ProSerGly Gly MET
Ala Pro GGC GAC

Ala Thr Gln Asp GlnSer SerSerGly Gly Ser Gly Asp GGT CAC

Gly Ser Ser Glu GlyLys AlaGluGly Glu Lys Gly His 85 9p 95 AGC CTC
~

Gln Gly Lys Cys ThrLys GluGluCys His Lys Ser Leu ATC GGC

Lys Lys MET Cys LysGly CysThrLeu Ser Ala Ile Gly TGC GCC

His Ser Lys Ala LysCys ThrLysSer Cys Val Cys Ala TAGG GGCCG TGCCGG GAGACCTCGA

Pro Thr Cys GCTTCACTTC
ACTTCTTTGT

GTTGCGTAGA
CGAAGG

AAGCGT

GATGGATAGA
TCGCGC

GTGCATCATG A
TT

AG

AATGCACAAT
TTATCT

AGACTCATCC

CCTCATATCT
AT

2~ ~J:~'~l A

C

SUBSTITUTE SHEET

WO 92!13957 PCT/EP92/00275 ..

L, 5. Sequence Description : SEQ ID NO. 4 SEQUENCE TYPE: nucleotide SEQUENCE LENGTH: 111 by STRANDEDNESS: double-stranded TOPOLOGY: linear MOLECULAR TYPE: genomic DNA amplified by means of PCR

ORIGINAL SOURCE

ORGANISM: Nicotiana tabacum cv. Samsun FEATURES: from nt 1 TO nt 7 : spacer sequence from nt 8 TO nt 13 : KpnI site from nt 16 TO nt 100 : intron sequence from nt 101 TO nt 106 : KpnI site from nt 107 TO nt 111 : spacer sequence PROPERTIES: intron of TA36 gene with KpnI linkers and 3' end at 5' SUBSTITUTE SHEET

. . ,, 6. Sequence Description : SEQ ID NO. 5 SEQUENCE TYPE: nucleotide SEQUENCE LENGTH: by STRANDEDNESS: double-stranded TOPOLOGY: circular MOLECULAR TYPE: d DNA
plasmi FEATURES:

- nt 1 to nt 396 : pUCl8 ived DNA
der - nt 397 to nt 80 2 : 3' om the untranslated end derived fr nopaline synthase gene of robacteriumT-DNA
Ag - nt 803 to nt 12 23 : barnase TA36 coding sequence with intron - nt 1119 to nt 203 : TA36intron - nt 1224 to nt 408 : CA55 2 promoter from corn - nt 2409 to nt 272 . 35S3 3 promoter - nt 3273 to nt 824 . coding 3 sequence of _bar gene - nt 3825 to nt 137 . 3' from the 4 untranslated end derived nopaline synthase gene of robacteriumT-DNA
Ag - nt 4138 to nt 376 : pUClBderived PROPERTIES: plasmid as pVE149 DNA replicable in E. coli designated SUBSI°iTUTE SHEET

., w ~

TTCTTCTCTCGCATTGGTTTCATCCAGCCAGGAGACCCGAATCGAA''!'TGA1500 ' GTATGTATTCCCTCCATTCCATATTCTAGGAGGTTTTGGCTTTTCATACC 1950 i TCAAGAACACAGAGAAAGACATATTTCTCAAGATCAGAAGTACTAT'rCCA2600 SUBSTITUTE SHEET

2 .~. 0 3 ~ '7 ,'~

GGCGTT'TTTCCATAGGCTCCGCCCCCCTGACGAGCATCACAAAAATCGAC 4600 TTGGTAGCTCTTGATCCGGCAAACAAACCACCGCTGGTAGCGGTGGT'rTT5100 TTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAA.AGGATCTCAAGp..AGA5150 C

SUBSTITUTE SHEET .

Claims (9)

CLAIMS:

1. A recombinant anther-specific promoter region, wherein said promoter region has the sequence of SEQ ID
No. 5 between nucleotide position 1224 and nucleotide position 2408.

2. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising the promoter region of claim 1 operably linked to a heterologous structural gene, for the production of plants in which said heterologous structural gene is to be expressed predominantly in anther cells.

3. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising a DNA encoding a ribonuclease operably linked to the promoter region of claim 1, for the production of male-sterile plants.

4. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising a DNA encoding a ribonuclease operably linked to the promoter region of claim 1, for the production of hybrid monocotyledonous plants.

5. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising a DNA encoding a barnase operably linked to the promoter region of claim 1, for the production of male-sterile plants.

6. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising a DNA encoding a barnase operably linked to the promoter region of claim 1, for the production of hybrid monocotyledonous plants.

7. Use of a plant or seed containing, stably integrated in the genome of its cells, a foreign DNA
comprising a DNA encoding a barstar operably linked to the promoter region of claim 1.

8. A method for isolating a stamen-specific promoter from corn which comprises performing inverse PCR on genomic corn DNA using a pair of oligonucleotides consisting of a first oligonucleotide complementary to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 2 or complementary to a sequence of 21 to 24 consecutive.
nucleotides of the DNA of SEQ ID No. 3 between nucleotide positions 1120 and 1839; and a second oligonucleotide corresponding to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 2 or corresponding to a sequence of 21 to 24 consecutive nucleotides of the DNA of SEQ ID No. 3 between nucleotide positions 1120 and 1839.

9. The method of claim 8, wherein said first oligonucleotide has the following sequence:
5' GCT CGA TGT ATG CAG TGC AGC 3' and wherein the second oligonucleotide has a sequence selected from the group of:
5' CGT CGC CGT GTC GGT GCT TCT CG 3' and 5' CCG TTG CGT TGC GTT GCG TAG ACG 3'.