Gene Exhibiting Resistance to Acetolactate Synthase Inhibitor Herbicides

This invention relates to a novel gene and enzyme isolated from cocklebur (Xanthium sp.) which confers resistance to several structurally unrelated classes of herbicides in plants, plant tissues and seeds. In particular, the class of herbicides consists of herbicides wherein acetolactate synthase (ALS) is the site of action and includes sulfonylureas, imidazolinones, triazolopyrimidines, pyrimidyloxybenzoates and phthalides.

The control of undesirable plants by the use of herbicides is an extensively used practice and the market for herbicidal compounds continues to expand. However, some weedy plant species are resistant to some of these herbicidal compounds. As a result, either greater amounts of the herbicidal compounds must be applied to control these weeds, or herbicides with greater potency have to be used. The result in either case can frequently be a sensitivity of desirable crop plants to the herbicidal compounds. An alternative to the use of increased amounts of herbicides or the application or identification and development of new herbicides for use with particular crop plants is the modification of susceptible or sensitive crop species so that they are resistant or tolerant to specific herbicides. This strategy of modifying susceptible or sensitive crop species should reduce chemical herbicide input while at the same time maximizing weed control. A number of methods exist for achieving this goal, and one such method is through the genetic transformation of crop plants for herbicide resistance.

Resistance to specific herbicides has been shown to be the result of changes in enzymes which are involved in particular biosynthetic pathways. For example, the non-selective postemergence herbicide glyphosate acts by inhibiting the enzyme 5-enolpyruvyl-3-phosphoshikimate synthase (EPSP). Glyphosate tolerant plants have been produced by inserting into the genome of a plant the capacity to produce high levels of EPSP synthase and reference is made to U.S. Pat. No. 5,312,910. Also glyphosate tolerant plants have been produced by desensitizing EPSP synthase to glyphosate.

Acetolactate synthase (ALS) which catalyzes the first reaction in the biosynthetic pathway to the branched amino acids has been shown to be the site of action of several structurally unrelated classes of herbicides, including: sulfonylureas (LaRossa et al., J. Biol. Chem. (1984) 259:8753-8757), imidazolinones (Shaner et al., Plant Physiol. (1984) 76:545-546), triazolopyrimidines (Subramanian et al., ACS Sym. Series 389 (1989) pp 277-288) and pyrimidyloxybenzoates (EPA 223 406, EPA 249 707 and EPA 249 708). Other classes of herbicides with ALS as the target include pyrimidylsalicylates, carbamoylpyrazolines, sulfonylimino-triazinyl heteroazoles, N-protected valylanilides, sulfonylamide azines, pyrimidyl madelie acids, benzenesulfonyl carboxamide compounds, substituted sulfonyldiamides, and ubiquinone-o. Transgenic plants with decreased sensitivity to inhibition by sulfonylurea and imidazolinone herbicides have been disclosed. Particular mention is made of EPA 0 525 384; U.S. Pat. Nos. 4,761,373; 5,198,599 and 5,331,107 and references cited therein.

Since many of the ALS inhibitor herbicides are known for their low mammalian toxicity, high herbicidal potency at low use rates and broad range crop selectivity, crop hybrids or varieties with resistance to these herbicides would provide an attractive solution to allowing herbicidal use without risk of damage to sensitive crops.

The present invention provides a novel functional acetolactate synthase enzyme (ALS) which exhibits resistance to herbicides which target ALS. The subject peptide isolated from a resistant cocklebur biotype was cloned and is described hereinafter. The invention also contemplates transgenic plants having enhanced resistance to herbicides which target ALS wherein a DNA sequence encoding the novel functional ALS resistant to herbicidal inhibition is introduced into a plant of interest. The plants can then be grown to produce seed having the resistant genotype.

It is to be understood that the following detailed description presents a single embodiment of the invention. This embodiment relates to a particular polypeptide and gene encoding the polypeptide which renders certain cocklebur plants tolerant or resistant to ALS inhibitor herbicides. However, it is understood that this gene may be made in whole or part by chemical or enzymatic synthetic methods.

FIG. 1 shows the nucleic acid sequence of the resistant cocklebur biotype (SEQ ID No: 1). The amino acid translation product starts at nucleotide 111 corresponding to the beginning of the chloroplast targeting sequence. The mature protein starts at nucleotide 342.

FIG. 2 shows the amino acid sequence of the resistant cocklebur biotype (SEQ ID No: 2). Amino acid residues 1-77 depict the chloroplast targeting sequence and amino acid residues 78-648 depict the mature protein sequence.

FIG. 3 shows the nucleic acid sequence of wild type, sensitive cocklebur (SEQ ID No: 3).

FIG. 4 shows the amino acid sequence of wild type, sensitive cocklebur (SEQ ID No: 4).

FIG. 5 depicts the comparison of ALS inhibitor herbicides against susceptible and resistant cocklebur.

FIG. 6 depicts the map of pSCI 697 and the relative position of subclones.

FIG. 7 depicts the sequencing strategy for susceptible cocklebur clones.

FIG. 8 depicts the relative position of the subclones for resistant cocklebur.

The resistant cocklebur biotype (hereinafter R-XANST) was discovered in a soybean field characterized by a lack of crop rotation and the use of imidazolinone type herbicides.

The complete sequence of the R-XANST cDNA is described in FIG. 1 (SEQ ID No: 1) and the 2156 bp cDNA codes for a 648 residue protein described in FIG. 2 (SEQ ID No: 2). Also described is the 77 residue chloroplast targeting sequence. The cDNA sequence from wild type sensitive cocklebur (hereinafter S-XANST) is described in FIG. 3 (SEQ ID No: 3) and the corresponding protein is described in FIG. 4 (SEQ ID No: 4). The S-XANST is a 648 residue protein with a 77 residue chloroplast targeting sequence. The mature form (571 residues) is 89%, 86%, 78%, and 46% identical with respect to the amino acid residues to the corresponding enzyme from tobacco, arabidopsis, maize and yeast, respectively. The S-XANST and R-XANST sequences are 99.4% identical at the DNA level. Most of the mismatches occur on the third base of the codons and do not modify the translated product. It is well known in the art that the Proline and Serine/Alanine domains are important in conferring resistance to the ALS inhibitor. (See EP 0 525 384 and Lee, K. Y., et al., EMBO J. (1988) 7:1241-1248). In Lee et al. supra two distinct ALS genes were identified in tobacco. One resistant mutant was found to have a single Pro to Gln replacement at amino acid residue 196 and this gene was classified as a class I gene. In a second resistant mutant two amino acid changes were identified in the second ALS gene. This gene was classified as class II gene and included a replacement at amino acid residue 196, Proline to Alanine. A change at the amino acid residue 621 (Ala) is disclosed in EP 0 525 384. However, between the R-XANST and S-XANST of this invention there are no differences found in the Proline or Serine/Alanine domains.

With respect to the differences between S-XANST and R-XANST at the amino acid level there are five differences. These include the change of Lys (63) to Glu (63); Phe (258) to Leu (258); Gln (269) to His (269); Asn (522) to Ser (522) and Trp (552) to Leu (552). The change at residue 522 is the change of a basic residue to an uncharged residue. The changes at positions 522 and 552 are thought to be particularly important in conferring resistance or tolerance. Therefore this invention relates not only to a functional ALS enzyme having the amino acid sequence of SEQ ID No.: 2 but also to an enzyme having modifications to the amino acid sequence of SEQ ID No.:2 wherein said modifications include an amino acid sequence having the same function and about 90% or greater similarity to the sequence of SEQ ID No.:2 and more preferably having 95% or greater similarity. In this context similarity means having at least 90% identical or conservatively replaced amino acid residues in a like position. Additionally, modified functional ALS enzymes may have specific changes at amino acid residue 552 wherein said residue is other than Leu and is preferably but not limited to Ser, Arg, Gly and Cys.

ALS inhibitor compounds comprise a large class of structurally unrelated herbicidal compounds including sulfonylureas, imidazolinones, triazolopyrimidines, pyrimidyloxybenzoates and phthalides.

Representative examples of herbicidal sulfonylureas include:

chlorsulfuson, [2-chloro-N-[[CH-methoxy-6-methyl-1,3,5-triazin-2-yl-amino]carbonyl]benzen esulfonamide];

metsulfuron, [2-[(L-4-methoxy-6-methyl-1,3,5-triazin-2-yl)amino]carbonyl];

thifensulfuron, {3-[[[[(4-methoxy-6-methyl-1,3,5,triazin-2-yl)amino]carbonyl]amino]sulfony l]-2-thiophenecarboxylic acid};

bensulfuron, {2-[[[[[[(4-6-dimethoxy-2-pyrimidinyl)amino]carbonyl]amino]sulfonyl]methyl ]benzoic acid }; and

chlorimuron, {2-[[[[(4-chloro-6-methoxy-2-pyrimidinyl)amino]carbonyl]amino]sulfonyl]ben zoic acid}.

Representative examples of herbicidal imidazolinones include:

imazapyr, {(+)-2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazole-2-yl]-3- pyridinecarboxlyic acid};

imazaquin, {2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazole-2-yl]-3-quin olinecarboxylic acid}; and

imazethapyr, {(+)-2-[4,5-dihydro-4-methyl-4-(1-methylethyl)-5-oxo-1H-imidazole-2-yl]-5- ethyl-3-pyridinecarboxylic acid}.

Certain phthalides exhibit general and selective herbicidal activity against plants. Members of this family include those compounds disclosed in EPA 461 079 and PCT WO/9110653 hereby incorporated by reference.

For the purpose of this specification the structurally different classes of herbicides referred to above which inhibit ALS as their primary site of action are collectively referred to as ALS inhibitor compounds or herbicides. The term herbicidally effective amount refers to the amount of herbicide which achieves inhibitive control or modification of undesired plant growth when applied to plants or to the area in which these plants are growing.

The term "herbicide resistance" as used herein concerning the R-XANST biotype means the naturally-occurring inheritable ability of a plant biotype within a plant population to survive a herbicide treatment that would, under normal conditions effectively control that population. Resistant is not due to the herbicidal dosage the resistant plant is able to withstand, but rather the difference in response between the response of the resistant biotype and the susceptible population. The term, when used in conjunction with genetic transformation or transgenic plants also means wherein resistance or tolerance is conferred by a foreign gene encoding an enzyme which is resistant to deactivation by an ALS herbicide or tolerant to a ALS herbicide at a concentration which would normally inhibit the activity of an unaltered enzyme. Resistance in this context includes resistance of a plant to multiple herbicides having the same target site due to the presence of a predominantly single resistance mechanism.

Recombinant DNA techniques may be used to obtain resistant lines of plants. These techniques include the use of nucleic acid sequences encoding the ALS derived from resistant plants. The ALS gene may be derived from cDNA or synthesized in whole or part. The advantage to synthesizing a gene is the desirability of modifying a portion of the codons to enhance expression by employing host preferred codons. Additionally DNA constructs of the invention may comprise not only the claimed nucleic acid sequences but also may include promoters, terminating sequences, polylinkers and other regulatory regions well known to those skilled in the art.

Promoters refer to nucleotide sequences at the 5'-end of a structural gene which direct the initiation of transcription and include all the regulatory regions required for transcription including the region coding for the leader sequence of mRNA. A number of promoters which are active in microbial and plant cells have been described in the literature. Suitable plant promoters include nopaline synthase (NOS), octopine synthase (OCS), cauliflower mosaic virus (CaMV) 19S and 35S, ribulose bisphosphate carboxylase (RUBISCO), and heat shock Brassica promoter (HSP 80). Suitable promoters used in the transformation of E. coli include, but are not limited to P.sub.Tac, lambda .sub.pr and T.sub.7 which are available commercially. The promoters used in the present invention may be modified to affect control characteristics and further may be a composite of segments derived from more than one source, naturally occurring or synthetic. Termination sequences refer to a nucleotide sequence at the end of a transcriptional unit that signals termination of transcription. Terminators are 3'-non-translated DNA sequences that contain a polyadenylated signal. Examples of terminators are known and described in the literature and include but are not limited to nos (nopaline synthase terminator), the 35S terminator of CaMV and the zein terminator.

Vectors comprising the nucleic acid sequences described above represent another embodiment of the invention. Expression vectors are typically plasmids. Plasmids in general are circular double stranded DNA loops including a promoter operably linked to the DNA sequence or sequences encoding the protein of interest, transcription termination sequences and the remaining vector with 3' and 5' elements. One skilled in the art is aware of many types of vectors including virus vectors, baculovirus; phage vectors; Agrobacterium-Ti plasmid; binary vectors and other vectors suitable for plant transformation. The insertion of the DNA sequences into plant host cells according to the invention occurs according to techniques known in the art.

Vectors of the invention may also include other DNA sequences known in the art, including but not limited to: a) stability sequences; b) one or more marker sequences, for example, antibiotic and other herbicide resistance markers including cat (chloramphenicol acetyl transferase), npt II (neomycin phosphotransferase II), PAT (phosphinothricin acetyltransferase the expression of which confers resistance to the herbicide Basta); EPSP (5-enolpyruvyl-shikimate 3-phosphate synthase, the enzyme inhibited by glyphosate, the active ingredient in the herbicide Roundup) and bxn (bromoxynil-specific nitrilase); c) signal sequences or leader peptides which are specific N-terminal sequences known to efficiently direct the mature peptide to the endoplasmic reticulum, vacuole or extracellular space via translocation through the endoplasmic reticulum membrane and which is excised during translocation; d) intron sequences; and e) enhancer or other elements necessary to increase or decrease levels of expression obtained in particular parts of plants under certain conditions. These examples are stated by way of example only and are not intended to limit the invention in any manner.

A wide variety of techniques are available and known in the art for carrying out plant cell or tissue transformation. These include but are not limited to, direct transfer of DNA into whole cells, tissues or protoplasts, optionally assisted by chemical or physical agents to increase cell permeability to DNA, for example treatment with polyethylene glycol, and dextran sulfate; electroporation, heat shock and ballistic implantation of DNA coated particles. Transformation is also mediated by Agrobactedum strains, notably A. tumefaciens, and also by various genetically engineered transformation plasmids which include portions of the T-DNA of the tumor inducing plasmids of Agrobactria. Other means for effecting entry of DNA into cells include the use of viral vectors, agroinfection, binary vectors and cotransformation. (See Jensen et al., (1993), Techniques for Gene Transfer, pp 125-146 in Transgenie Plants, vol. 1, eds. King and Wu, Academic Press). Transformed or transfected bacterial cells are included in the present invention, for example E. coli and Bacillus thuringenis (B.t.). Considerable experience in biotechnology has already been achieved and a wide variety of suitable operatively functional plasmids and transfer expression vectors systems are known and available for these organisms.

As used herein the term "genetic transformation" means the stable integration of a foreign gene into the genome of a plant regenerated from nucleic acid treated plant protoplasts, cells or tissue. "Transgenic plants" as used herein refers to plants carrying the stably integrated foreign gene. A "foreign gene" is a term used in the art to denote a gene or group of genes which has been transferred to a host cell or host plant from a source other than the host cell or host plant.

Virtually all plants of agronomic or horticultural value are known to be both transformable and regenerable, and all crop plants which may be sensitive to any herbicide of the ALS inhibitory herbicide class would be considered suitable host material. The techniques vary in individual detail from species to species as is well recognized by one skilled in the art. Means of regenerating plants are well documented in the literature. For a review on plant transformation and regeneration, see Ritchie and Hodges, pps. 147-178, in Kung and Wu, Transgenic Plants, vol. 1, 1993, Academic Press. Plant tissue includes differentiated and undifferentiated tissue and cells of plants including but not limited to roots, shoots, leaves, pollen, embryos, seed and various forms of aggregations of plant cells including callus.

Crop plants can be evaluated for ALS herbicide resistance to one or more herbicides of interest in the ALS inhibitor herbicide class, particularly a sulfonylurea, imidazolinone or phthalide by the ability of the plant to grow in the presence of the herbicide as compared to nontransgenic plants. Crop plants of particular interest include soybeans; cereal crops, such as maize; tomatoes; and sunflower. The invention herein provides not only for the transgenic plants, but also the seeds and progeny thereof which comprise and express the resistant R-XANST gene. In the field, genotypes expressing the R-XANST may be used with ALS inhibitory herbicides to effectively combat weed pests.

Plant Material:

Cross-resistant seeds were obtained from Pemiscoll County Missouri (herein after R-XANST). The field from which the R-XANST seeds were obtained had been treated with imidazolinone herbicides during the 1990, 1991 and 1992 growing seasons. Susceptible cocklebur seeds were obtained from Azlin Seed Company (hereinafter S-XANST).

Herbicidal Evaluations:

Seeds of S-XANST and R-XANST are planted one per pot in soil mix. Pots are placed on a heating mat and the soil temperature increased from about 65.degree. F. to about 90.degree. F. At the 2-3 leaf stage, plants are treated with test herbicides using a linear track laboratory sprayer calibrated to deliver 42.5 GPA. Plants are watered by subirrigation after herbicide application. Observations of percent control are made 21 days after herbicide application. The test ALS inhibitor herbicides are listed below including their chemical class:

Broadstrike, common name: flumetsulam also known as DE-498, (triazolopyrimidine sulfonanilide) applied at 0.02 lb ai/A;

Glean, common name: chlorsulfuron, (sulfonyl-urea) applied at 0.01 lb ai/A;

Staple, common name: pyrthiobac-sodium, (benzoate phthalide) applied at 0.03 lb ai/A;

3-[(4,6-Dimethoxy-2-pyrimidinyl)carbonyl]-N,N-dimethyl-2-pyridinecarboxamid e, (phthalide) hereinafter SAN 1, applied at 0.05 lb ai/A;

3-6-Dichloro-2-[(4,6-dimethoxy-2-pyrimidinyl)carbonyl]benzoic acid, isopropyl ammonium salt, (phthalide), hereinafter SAN 2, applied at 0.1 lb ai/A and

Pursuit, common name: imazethapyr (imidazolinone) applied at 0.03 lb ai/A.

The effect of the various ALS inhibitors against R-XANST and S-XANST is depicted in FIG. 5. The results are reported as % control wherein 100% injury is equal to complete control as determined by growth. All treatments are replicated four times in a randomized complete block design. This data confirms the presence of resistance of XANST to different classes of herbicides, which had previously been unknown, particularly resistance was unknown to the phthalide herbicidal compounds.

Additionally tests are preformed at much higher herbicide rates to determine the extent of resistance. SAN 2 performed greater than 95% injury to S-XANST at 0.1 lb ai/A. However, even at 30 lb ai/A the injury to R-XANST did not exceed 50% injury. At 0.03 lb ai/A Staple provides greater than 90% injury to S-XANST, but injury to R-XANST did not exceed 55% at the highest test rate, 9.0 lb ai/A. Broadstrike at 0.01 lb ai/A causes greater than 95% injury to S-XANST, but application of 6 lb ai/A produced only 25 % injury on R-XANST. Glean tested at 1.56 lb ai/A, a rate of 100 times field use, provided 88% injury to R-XANST. Pursuit provides 90% injury to R-XANST when used at 100 times the normal field rate, (6.25 lb ai/A).

Enzymatic evaluations:

Cocklebur is grown in a greenhouse. Five grams of fresh green leaf tissue are ground to a powder in a mortar at the temperature of liquid nitrogen. The powder is extracted with 100 ml of buffer containing 50 mM N-[2-hydroxyethyl]piperazine-N'-[3-propanesulfonic acid], (EPPS); pH 7.2; 5 mM MgCl.sub.2 ; 2 mM EDTA; 1 mM valine; 1 mM leucine; 10% glycerol; 10 mM pyruvate; 5 mM Dithiothreitol; 1% polyvinyl polypyrrolidone (PVPP) and 10 .mu.dM flavin adenine dinucleotide and filtered through cheesecloth. The filtrate was centrifuged at 15,000.times.g for 15 min.

The procedures used for the isolation of ALS from cocklebur are similar to those known in the art. The supernatant resulting from centrifugation of the crude extract is brought to 40% ammonium sulfate. The ammonium sulfate pellet is not frozen and is resuspended in standard buffer used for I.sub.50 determinations and then gel-filtered through the same buffer. This preparation is used for determinations of Km for pyruvate and I.sub.50. The Km is equal to the substrate concentration at which the initial reaction velocity is half maximal and also referred to as the Michaelis-Menten constant. The Km is determined using Lineweaver-Burke double reciporcial plots. I.sub.50 is the concentration in which 50% inhibition is observed and is determined from linear regression analysis of the linear portion of the dose/response curve.

Substrate saturation with pyruvate is hyperbolic for R-XANST and S-XANST The Km value for R-XANST is 6 mM and is very close to the Km value of the S-XANST enzyme which is 3 mM. This result suggests that the resistant enzyme is unimpaired with respect to pyruvate.

The I.sub.50 values by various ALS inhibitors are determined on both sensitive enzymes isolated from S-XANST and resistant enzymes isolated from R-XANST and are reported in Table 1. The results indicate that the resistant enzyme is highly resistant to the ALS inhibitors tested.

cDNA sequencing:

1. Preparation of polyA+mRNA from S-XANST and R-XANST.

Mature green leaves are obtained from S-XANST plants at the flowering stage. The guanidium isothiocyanate method is used for preparation of total RNA from leaves. (See Chirgwin, J. J. et al., (1979) Biochemistry 18:5294). The total RNA is further purified by standard CsCl gradient preparation and ethanol precipitation. (See Ausubel, F. M. et al., (1993) Current Protocols in Molecular Biology, Wiley & Sons Publishers, New York). RNA quality is checked by the presence on a formaldehyde gel of the two undegraded rRNA bands according to procedures standard in the art. The mRNA fraction is isolated from the total RNA by affinity chromatography on oligo (dT) cellulose using the procedure described in Aviv, H. and Leder, P., (1972), Purification of biologically active globin messenger RNA by chromatography on oligothymidylic acid-cellulose., Proc. Natl. Acad. Sci. 69:1408-1412. The mRNA fraction represents about 1% of the total RNA.

2. Preparation of double stranded cDNA.

Double stranded cDNA is synthesized using the Pharmacia-LKB (Piskataway, N.J.) cDNA synthesis kit using 10 .mu.g of poly A+mRNA and oligo d(T) primers (Pharmacia-LKB). Using the same kit, NolI/EcoRI linkers/adaptors are added and the constructs purified by gel chromatography on Sephacryl-S-400 (Pharmacia-LKB).

3. Construction of a cDNA library from S-XANST.

The cDNA from S-XANST is cloned by ligation into EcoRI digested and dephosphorylated .lambda.gt10 arms (Promega Corp., Madison Md.). The ligation mixture is packaged into phage extracts (Promega) and used to infect c600 Hfl E. Coli. The phage library obtained has a complexity of 10.sup.5 different phages, large enough to obtain several copies of the ALS cDNA.

4. Design of degenerated PCR primers for the amplification of a partial ALS gene from any organism, to be used as a screening probe.

ALS sequences from different organisms are obtained from GeneBank. They are used to define two evolutionary conserved regions for PCR primer design. From alignment of the known sequences of maize, tobacco, Brassica, arabidopsis and yeast, degenerated primers designated ALS-1 and ALS-2 are obtained from Keystone Laboratories, Inc. (Menlo Park, Calif.).

The ALS-1 primer corresponds to the amino acid sequence Met Leu Gly Met His Gly, (SEQ ID No: 5) and the nucleotide sequence ATG[CT]T[ACTG]GG[ACTG]ATGCA[CT]GG (SEQ. ID No: 6).

The ALS-2 complement primer corresponds to the amino acid sequence Val Gly Gln His Gln Met Trp/Phe (SEQ ID No: 7) and the nucleotide sequence GT[ACGT]GG[CAGT]CA[AG]CA[CT]CA[AG]ATGT (SEQ ID No: 8), and the real primer corresponds to the sequence ACAT[CT]TG[AG]TG[CT]TG[ACTG]CC[ACGT]AC (SEQ ID No: 9).

5. PCR amplification, cloning and sequencing of a fragment of S-XANST.

Using 10 ng of S-XANST cDNA obtained in step 2, and the primers designed in step 4, a 400 bp cDNA fragment is amplified, and cloned into the sequencing vector pBluescript (Stratagene, LA Jolla Calif.) to yield the plasmid pSCI696. The plasmid is completely sequenced using the standard dedeoxy termination method and [.sup.33 P]dATP labeling. The sequencing products are separated on saturated urea polyacrylamide gel electrophoresis, fixed and exposed for autoradiography. These methods are standard methods known to those skilled in the art and described in Ausubel, F. M. et al., (1993) Current Protocols in Molecular Biology, Wiley & Sons, New York. The 100 amino acid residue sequence shows 75 % amino acid residue identity with Arabidopsis ALS.

6. Screening of the library for S-XANST.

The library obtained above in step 3 is screened with a .sup.32 P-labelled probe obtained by random labelling of the PCR fragment obtained in step 4. Three rounds of screening provided successive enrichment of positive plaques from 8 in 10.sup.5 to 6 in 1000 to 4 in 4. DNA from these four positives is prepared using the Promega kit (Magic DNA lambda prep). The size of the cDNA insert for each is determined by PCR using the primers gt10L and gt10R that anneal on the lambda vector around the cloning site.

The gt10L primer has a nucleotide sequence corresponding to

GTTCAGCCTGGTTAAGTCCAAGC (SEQ ID No: 10).

The gt10R primer has a nucleotide sequence corresponding to GAGTATTCTITCCAGGGTAAAAAGC (SEQ ID No: 11).

The sizes correspond to 2.5 kb, 2.4 kb, 1.6 kb and 1.1 kb. Using the ALS specific PCR primer CALSC; GGCGAAGCTATTCCTCCG (SEQ ID No: 12) and gt10R, it is determined that the 5' untranslated region (between the stop codon and the polyadenylation site ) is about 460 bp for all cDNAs. The largest one therefore has a coding region about 2 kb allowing the encoding of 660 amino acids.

7. Sequencing the S-XANST clones.

The cDNA from the largest clone from the four positives is used, and is amplified using gt10L and gt10R. The resulting fragment is cloned into pBluescript (Stratagene), generating the plasmid pSCI697. FIG. 6 gives the map of pSCI697, and the relative position of the subclones generated by restriction digest. HindlII digest of pSCI697 and subsequent cloning of the HindlII pieces into new pBluescript give the plasmids pSCI671, pSCI672 and pSCI673. Another subclone is obtained by PCR amplification of the lambda DNA using gt10L and 673C (pSCI707). The plasmids are sequenced using the primers T3, T7 (Stratagene), specific for the polylinker of pBluescript, and the primers designated, CALSNC, CALSC, 671A, 672A, 672B, 672C, 673A, 673B, 673C, CK.sub.-- ALS.sub.-- 1, CK.sub.-- ALS.sub.-- 2, and CK.sub.-- ALS.sub.-- 3, specific for the ALS sequence. FIG. 7 depicts the extend of the different sequences obtained. The sequence is disclosed for the primers hereinbelow.

FIG. 3 (SEQ ID No: 3) gives the complete nucleotide sequence of pSCI697 with the translation of the coding region. S-XANST is a 648 residue protein as described in FIG. 4 (SEQ ID No: 4) with a 77 residue chloroplast targeting sequence.

8. Sequencing of R-XANST.

Double stranded cDNA is prepared as above from mature leaves of the R-XANST. 10 ng of the double stranded DNA is used in the PCR amplification of the ALS. This is achieved in two fragments. The first fragment is amplified with the primer 673A and 672B and cloned into pBluescript as above to generate pSCI704. The second fragment is amplified with CK.sub.-- ALS.sub.-- 1 and CK.sub.-- ALS.sub.-- 3 and cloned into pSCI709. Partial clones are also obtained and cloned using CK.sub.-- ALS.sub.-- 1 and 673C (pSCI708), 671A and 672B (pSCI706), CALSC and 672B (pSCI705), 672A and 672C (pSCI702), 672A and 672B (pSCI701), 673A and 673B (pSCI703). The relative position of these clones with respect to pSCI697 is given in FIG. 8. The sequence of these clones is established using the primers described hereinabove. FIG. 1 (SEQ ID No: 1) gives the complete nucleotide sequence of the cDNA of R-XANST with the translation of the coding region. The S-XANST and R-XANST are 99.4% identical at the DNA level. R-XANST is a 648 residue protein as described in FIG. 2 (SEQ ID No: 2) with a 77 residue chloroplast targeting sequence. The use of these primers is also one aspect of the instant invention.