| PATENT ASSIGNEE'S COUNTRY | USA |
| UPDATE | 09.99 |
| PATENT NUMBER | This data is not available for free |
| PATENT GRANT DATE | 14.09.99 |
| PATENT TITLE |
Fatty acid desaturase genes from plants |
| PATENT ABSTRACT |
The preparation and use of nucleic acid fragments encoding fatty acid desaturase enzymes are described. The invention permits alteration of plant lipid composition. Chimeric genes incorporating such nucleic acid fragments with suitable regulatory sequences may be used to create transgenic plants with altered levels of unsaturated fatty acids. |
| PATENT INVENTORS | This data is not available for free |
| PATENT ASSIGNEE | This data is not available for free |
| PATENT FILE DATE | 26.08.94 |
| PATENT CT FILE DATE | 03.12.92 |
| PATENT CT NUMBER | This data is not available for free |
| PATENT CT PUB NUMBER | This data is not available for free |
| PATENT CT PUB DATE | 10.06.93 |
| PATENT REFERENCES CITED |
Walter, M.H. et al, Mol. Gen. Genet, 222, 353-360 (1990). Browse, J. et al, Plant Physiol., 81, 859-864 (1986). Wada, H. et al, Nature, 347, 200-203 (1990). Lemieux, B. et al, Theor. Appl. Genet, 80, 234-240 (1990). Brockman, J. et al, Biological Abstracts, 89(12), Abstract No. 131194 (Jun. 1990). Thompson, G.A. et al, Proc. Natl. Acad. Sci. USA, 88, 2578-2582 (1991). Browse, J., Annu. Rev. Plant Physiol. Plant Mol. Biol., 42, 467-506 (1991). Mattson, F.H. et al, Journal of Lipid Research, 26, 194-202 (1985). Gailliard, T. In: Stumpf, P.K. Ed., The Biochemistry of Plants, 4, 85-116, Academic Press, NY (1980). Mensink, R.P. et al, New England J. Med., N323, 439-445 (1990). Ohlrogge, J.G. et al, Biochim. Biophys. Acta., 1082, 1-26 (1991). Knowles, P.F. In, Applewhite, T.H. Ed., World Conf. on Biotechnology for the Fats and Oils Industry Proceedings, American Oil Chemists' Society, pp. 35-38 (1980). Wang, X.M. et al, Plant Physiol. Biochem., 26, 777-792 (1988). Gunstone et al, Eds. The Lipds Handbook, Chapman and Hall, Ltd., Cambridge (1986) pp. 55-112. Shanklin, J. et al, Proc. natl. Acad. Sci, USA, 88, 2510-2514 (1991). Stukey, J.E. et al, J. Biol. Chem., 265, 20144-20149 (1990). Thiede, et al, J. Biol. Chem., 261, 13230-13235 (1986). Kaestner, K.H. et al, J. Biol. Chem., 264, 14755-14756 (1989). Browse, J. et al, UCLA Symp. Mol. Cell. Biol.; New Ser., Plant Gene Transfer, 129, 301-309 (1990). Somerville, C. et al, Science, 252, 80-87 (1991). Arondel, V. et al, Science, 258, 1353-1355 (1992). |
| PATENT PARENT CASE TEXT | This data is not available for free |
| PATENT CLAIMS |
We claim: 1. An isolated nucleic acid fragment encoding a plant plastid or microsomal delta-15 fatty acid desaturase enzyme, which catalyzes a reaction at C15-C16 of a fatty acyl chain and further wherein said isolated nucleic acid fragment hybridizes to one of the nucleotide sequences set forth in SEQ ID NOS:1, 3, 5, 7, 9, 11, and 15 under one of the following sets of conditions: (a) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, repeat with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min, then repeat twice with 0.2X SSC, 0.5% SDS at 50.degree. C. for 30 min each; (b) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, repeat with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min, then repeat twice with 0.2X SSC, 0.5% SDS at 60.degree. C. for 30 min each; (c) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for eighteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, then washing for 5 min at 50.degree. C. in 0.2X SSPE, 1% SDS; (d) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for sixteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, then wash with fresh solution for 10 min, then wash for 5 min at 50.degree. C. in 0.5X SSPE, 1% SDS; (e) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, then wash twice with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min each and then wash twice with 0.2X SSC, 0.5% SDS at 60.degree. C. for 30 min each; or (f) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for eighteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, followed by washing for 5 min at 50.degree. C. in 0.2X SSPE, 1% SDS. 2. An isolated nucleic acid fragment comprising a nucleic acid sequence encoding a plant plastid or microsomal enzyme which catalyzes the formation of a double bond between carbon positions 3 and 4 numbered from the methyl end of a fatty acyl chain, and further wherein the amino acid sequence comprising said enzyme contains at least one of the following amino acid sequences selected from the group consisting of FVLGHDCGHGSF, GHDCGH, HDIGTHVIHHLFP, HDIGTH, or HVIHHL. 3. An isolated nucleic acid fragment encoding a plant plastid or microsomal enzyme which catalyzes the formation of a double bond between carbon positions 3 and 4 numbered from the methyl end of a fatty acyl chain, wherein said isolated nucleic acid fragment encodes a protein comprising any one of the amino acid sequences set forth in SEQ ID NOS:2, 5, 7, 9, 11, 12, 15 or 17. 4. An isolated nucleic acid fragment encoding an enzyme which catalyzes a reaction at C15-C16 of a fatty acyl chain wherein said isolated nucleic acid fragment hybridizes to the isolated nucleic acid fragment of claim 3 under one of the following sets of conditions: (a) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, repeat with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min, then repeat twice with 0.2X SSC, 0.5% SDS at 50.degree. C. for 30 min each; (b) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, repeat with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min, then repeat twice with 0.2X SSC, 0.5% SDS at 60.degree. C. for 30 min each; (c) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for eighteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, then washing for 5 min at 50.degree. C. in 0.2X SSPE, 1% SDS; (d) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for sixteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, then wash with fresh solution for 10 min, then wash for 5 min at 50.degree. C. in 0.5X SSPE, 1% SDS; (e) hybridization in 50 mM Tris, pH 7.6, 6X SSC, 5X Denhardt's, 0.5% sodium dodecyl sulfate (SDS), 100 .mu.g denatured calf thymus DNA at 50.degree. C. overnight and wash with 6X SSC, 0.5% SDS at room temperature for 15 min, then wash twice with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min each and then wash twice with 0.2X SSC, 0.5% SDS at 60.degree. C. for 30 min each; or (f) hybridization in 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% sodium dodecyl sulfate (SDS), 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA at 50.degree. C. for eighteen hours and wash twice at room temperature with 2X SSPE, 1% SDS for 5 min, followed by washing for 5 min at 50.degree. C. in 0.2X SSPE, 1% SDS. 5. An isolated nucleic acid fragment of claim 1, claim 2, claim claim 3 or claim 4 wherein said fragment is isolated from a plant selected from the group consisting of soybean, oilseed Brassica species, Arabidopsis thaliana and corn. 6. A chimeric gene capable of causing altered levels of linolenic acid in a transformed plant cell, the gene comprising a nucleic acid fragment of any of claims 1, claim 2, claim 3 or claim 4 the fragment operably linked to suitable regulatory sequences. 7. Plants containing the chimeric gene of claim 6. 8. A method of producing seed oil containing altered levels of linolenic (18:3) acid comprising: (a) transforming a plant cell of an oil-producing species with a chimeric gene of claim 5; (b) growing fertile plants from the transformed plant cells of step (a); (c) screening progeny seeds from the fertile plants of step (b) for the desired levels of linolenic (18:3) acid; and (d) processing the progeny seed of step (c) to obtain seed oil containing altered levels of linolenic (18:3) acid. 9. A method of claim 8 wherein said plant cell of an oil-producing species is selected from the group consisting of Arabidopsis thaliana, soybean, oilseed Brassica napus, sunflower, cotton, peanut, and corn. 10. The isolated genomic DNA of Arabidopsis thaliana comprising the microsomal delta-15 desaturase identified by accession number ATCC 75167. 11. A plasmid designated pXF1 and bearing accession number ATCC 68874 comprising the DNA sequence of SEQ ID NO:10 which encodes a soybean delta-15 desaturase enzyme. 12. A plasmid designated pBNSF3 and bearing accession number ATCC 68854 comprising the DNA sequence of SEQ ID NO:6 which encodes an oilseed Brassica species delta-15 desaturase enzyme. 13. An isolated Polymerase Chain Reaction Product designated PCR20 comprising the DNA sequence of SEQ ID NO:14 which encodes a portion of the Zea mays delta-15 desaturase enzyme. 14. The method of claim 8 wherein said plant cell of an oil-producing species is selected from the group consisting of Arabidopsis thaliana, soybean, oilseed Brassica napus, sunflower, cotton and corn. -------------------------------------------------------------------------------- |
| PATENT DESCRIPTION |
FIELD OF THE INVENTION The invention relates to the preparation and use of nucleic acid fragments encoding fatty acid desaturase enzymes to modify plant lipid composition. BACKGROUND OF THE INVENTION Plant lipids have a variety of industrial and nutritional uses and are central to plant membrane function and climatic adaptation. These lipids represent a vast array of chemical structures, and these structures determine the physiological and industrial properties of the lipid. Many of these structures result either directly or indirectly from metabolic processes that alter the degree of unsaturation of the lipid. Different metabolic regimes in different plants produce these altered lipids, and either domestication of exotic plant species or modification of agronomically adapted species is usually required to economically produce large amounts of the desired lipid. Plant lipids find their major use as edible oils in the form of triacylglycerols. The specific performance and health attributes of edible oils are determined largely by their fatty acid composition. Most vegetable oils derived from commercial plant varieties are composed primarily of palmitic (16:0), stearic (18:0), oleic (18:1), linoleic (18:2) and linolenic (18:3) acids. Palmitic and stearic acids are, respectively, 16- and 18-carbon-long, saturated fatty acids. Oleic, linoleic, and linolenic acids are 18-carbon-long, unsaturated fatty acids containing one, two, and three double bonds, respectively. Oleic acid is referred to as a mono-unsaturated fatty acid, while linoleic and linolenic acids are referred to as poly-unsaturated fatty acids. The relative amounts of saturated and unsaturated fatty acids in commonly used, edible vegetable oils are summarized below (Table 1): TABLE 1 ______________________________________ Percentages of Saturated and Unsaturated Fatty Acids in the Oils of Selected Oil Crops Mono- Poly- Saturated unsaturated unsaturated ______________________________________ Canola 6% 58% 36% Soybean 15% 24% 61% Corn 13% 25% 62% Peanut 18% 48% 34% Safflower 9% 13% 78% Sunflower 9% 41% 51% Cotton 30% 19% 51% ______________________________________ Many recent research efforts have examined the role that saturated and unsaturated fatty acids play in reducing the risk of coronary heart disease. In the past, it was believed that mono-unsaturates, in contrast to saturates and poly-unsaturates, had no effect on serum cholesterol and coronary heart disease risk. Several recent human clinical studies suggest that diets high in mono-unsaturated fat and low in saturated fat may reduce the "bad" (low-density lipoprotein) cholesterol while maintaining the "good" (high-density lipoprotein) cholesterol (Mattson et al., Journal of Lipid Research (1985) 26:194-202). A vegetable oil low in total saturates and high in mono-unsaturates would provide significant health benefits to consumers as well as economic benefits to oil processors. As an example, canola oil is considered a very healthy oil. However, in use, the high level of poly-unsaturated fatty acids in canola oil renders the oil unstable, easily oxidized, and susceptible to development of disagreeable odors and flavors (Gailliard, 1980, Vol.4, pp. 85-116 In: Stumpf, P. K., Ed., The Biochemistry of Plants, Academic Press, New York). The levels of poly-unsaturates may be reduced by hydrogenation, but the expense of this process and the concomitant production of nutritionally questionable trans isomers of the remaining unsaturated fatty acids reduces the overall desirability of the hydrogenated oil (Mensink et al., New England J. Medicine (1990) N323: 439-445). Similar problems exist with soybean and corn oils. For specialized uses, high levels of poly-unsaturates can be desirable. Linoleate and linolenate are essential fatty acids in human diets, and an edible oil high in these fatty acids can be used for nutritional supplements, for example in baby foods. Linseed oil, derived from the Flax plant (Linum usitatissimum), contains over 50% linolenic acid and has widespread use in domestic and industrial coatings since the double bonds of the fatty acids react rapidly with oxygen to polymerize into a soft and flexible film. Although the oil content of flax is comparable to canola (around 40% dry weight of seed), high yields are only obtained in warm temperatures or subtropical climates. In the USA flax is highly susceptible to rust infection. It will be commercially useful if a crop such as soybean or canola could be genetically transformed by the appropriate desaturase gene(s) to synthesize oils with a high linolenic acid content. Mutation-breeding programs have met with some success in altering the levels of poly-unsaturated fatty acid levels found in the edible oils of agronomic species. Examples of commercially grown varieties are high (85%) oleic sunflower and low (2%) linolenic flax (Knowles, (1980) pp. 35-38 In: Applewhite, T. E., Ed., World Conference on Biotechnology for the Fats and Oils Industry Proceedings, American Oil Chemists' Society). Similar commercial progress with the other plants shown in Table 1 has been largely elusive due to the difficult nature of the procedure and the pleiotropic effects of the mutational regime on plant hardiness and yield potential. The biosynthesis of the major plant lipids has been the focus of much research (Browse et al., Ann. Rev. Plant Physiol. Mol. Biol. (1991) 42:467-506). These studies show that, with the notable exception of the soluble stearoyl-acyl carrier protein desaturase, the controlling steps in the production of unsaturated fatty acids are largely catalyzed by membrane-associated fatty acid desaturases. Desaturation reactions occur in plastids and in the endoplasmic reticulum using a variety of substrates including galactolipids, sulfolipids, and phospholipids. Genetic and physiological analyses of Arabidopsis thaliana nuclear mutants defective in various fatty acid desaturation reactions indicates that most of these reactions are catalyzed by enzymes encoded at single genetic loci in the plant. The analyses show further that the different defects in fatty acid desaturation can have profound and different effects on the ultra-structural morphology, cold sensitivity, and photosynthetic capacity of the plants (Ohlrogge, et al., Biochim. Biophys. Acta (1991) 1082:1-26). However, biochemical characterization of the desaturase reactions has been meager. The instability of the enzymes and the intractability of their proper assay has largely limited researchers to investigations of enzyme activities in crude membrane preparations. These investigations have, however, demonstrated the role of delta-12 desaturase and delta-15 desaturase activities in the production of linoleate and linolenate from 2-oleoyl-phosphatidylcholine and 2-linoleoyl-phosphatidylcholine, respectively (Wang et al., Plant Physiol. Biochem. (1988) 26:777-792). Thus, modification of the activities of these enzymes represents an attractive target for altering the levels of lipid unsaturation by genetic engineering. Genes from plants for stearoyl-acyl carrier protein desaturase, the only soluble fatty acid desaturase known, have been described (Thompson, et al., Proc. Natl. Acad. Sci. U.S.A. (1991) 88:2578-2582; Shanklin et al., Proc. Natl. Acad. Sci. USA (1991) 88:2510-2514). Stearoyl-coenzyme-A desaturase genes from yeast, rat, and mice have also been described (Stukey, et al., J. Biol. Chem.(1990) 265:20144-20149; Thiede, et al., J. Biol. Chem. (1986) 261:13230-13235; Kaestner, et al., J. Biol. Chem. (1989) 264:14755-1476). No evidence exists in the public art that describes the isolation of fatty acid desaturases other than stearoyl-ACP desaturases from higher plants or their corresponding genes. A fatty acid desaturase gene from the cyanobacterium, Synechocystis PCC 6803, has also been described (Wada, et al., Nature (1990) 347:200-203). This gene encodes a fatty acid desaturase, designated des A, that catalyzes the conversion of oleic acid at the 1 position of galactolipids to linoleic acid. However, these genes have not proven useful for isolating plant fatty acid desaturases other than stearoyl-ACP desaturase via sequence-dependent protocols, and the present art does not indicate how to obtain plant fatty acid desaturases other than stearoyl-ACP desaturases or how to obtain fatty acid desaturase-related enzymes. Thus, the present art does not teach how to obtain glycerolipid desaturases from plants. Furthermore, there is no evidence that a method to control the nature and levels of unsaturated fatty acids in plants using nucleic acids encoding fatty acid desaturases other than stearoyl-ACP desaturase is known in the art. The biosynthesis of the minor plant lipids has been less well studied. While hundreds of different fatty acids have been found, many from the plant kingdom, only a tiny fraction of all plants have been surveyed for their lipid content (Gunstone, et al., Eds., (1986) The Lipids Handbook, Chapman and Hall Ltd., Cambridge). Accordingly, little is known about the biosynthesis of these unusual fatty acids and fatty acid derivatives. Interesting chemical features found in such fatty acids include, for example, allenic and conjugated double bonds, acetylenic bonds, trans double bonds, multiple double bonds, and single double bonds in a wide number of positions and configurations along the fatty acid chain. Similarly, many of the structural modifications found in unusual lipids (e.g., hydroxylation, epoxidation, cyclization, etc.) are probably produced via further metabolism following chemical activation of the fatty acid by desaturation or they involve a chemical reaction that is mechanistically similar to desaturation. For example, evidence for the mechanism of hydroxylation of fatty acids being part of a general mechanism of enzyme-catalyzed desaturation in eukaryotes has been obtained by substituting a sulfur atom in the place of carbon at the delta-9 position of stearic acid. When incubated with yeast cell extracts the thiostearate was converted to a 9-sulfoxide (Buist et al. (1987) Tetrahedron Letters 28:857-860). This sulfoxidation was specific for sulfur at the delta-9 position and did not occur in a yeast delta-9-desaturase deficient mutant (Buist & Marecak (1991) Tetrahedron Letters 32:891-894). The 9-sulfoxide is the sulfur analogue of 9-hydroxyoctadecastearate, the proposed intermediate of stearate desaturation. Thus fatty-acid desaturase cDNAs may serve as useful probes for cDNAs encoding fatty-acid hydroxylases and other cDNAs which encode enzymes with reaction mechanisms similar to fatty-acid desaturation. Many of these fatty acids and derivatives having such features within their structure could prove commercially useful if an agronomically viable species could be induced to synthesize them by introduction of a gene encoding the appropriate desaturase. SUMMARY OF THE INVENTION Applicants have discovered a means to control the nature and levels of unsaturated fatty acids in plants. Nucleic acid fragments from glycerolipid desaturase cDNAs or genes are used to create chimeric genes. The chimeric genes may be used to transform various plants to modify the fatty acid composition of the plant or the oil produced by the plant. More specifically, one embodiment of the invention is an isolated nucleic acid fragment comprising a nucleotide sequence encoding a plant delta-15 fatty acid desaturase or a fatty acid desaturase-related enzyme with an amino acid identity of 50%, 65%, 90% or greater to the polypeptide encoded by SEQ ID NOS:1, 4, 6, 8, 10, 12, 14, or 16. The isolated fragment in these embodiments is isolated from a plant selected from the group consisting of soybean, oilseed Brassica species, Arabidopsis thaliana and corn. Another embodiment of this invention involves the use of these nucleic acid fragments in sequence-dependent protocols. Examples include use of the fragments as hybridization probes to isolate other glycerolipid desaturase cDNAs or genes. A related embodiment involves using the disclosed sequences for amplification of DNA fragments encoding other glycerolipid desaturases. Another aspect of this invention involves chimeric genes capable of causing altered levels of the linolenic acid in a transformed plant cell, the gene comprising nucleic acid fragments encoding encoding a plant delta-15 fatty acid desaturase or a fatty acid desaturase-related enzyme with an amino acid identity of 50%, 65%, 90% or greater to the polypeptide encoded by SEQ ID NOS:1, 4, 6, 8, 10, 12, 14, or 16 operably linked in suitable orientation to suitable regulatory sequences. Preferred are those chimeric genes which incorporate nucleic acid fragments encoding delta-15 fatty acid desaturase cDNAs or genes. Plants and oil from seeds of plants containing the chimeric genes described are also claimed. Yet another embodiment of the invention involves a method of producing seed oil containing altered levels of linolenic (18:3) acid comprising: (a) transforming a plant cell with a chimeric gene described above; (b) growing fertile plants from the transformed plant cells of step (a); (c) screening progeny seeds from the fertile plants of step (b) for the desired levels of linolenic (18:3) acid, and (d) processing the progeny seed of step (c) to obtain seed oil containing altered levels of the unsaturated fatty acids. Preferred plant cells and oils are derived from soybean, rapeseed, sunflower, cotton, cocoa, peanut, safflower, coconut, flax, oil palm, and corn. Preferred methods of transforming such plant cells would include the use of Ti and Ri plasmids of Agrobacterium, electroporation, and high-velocity ballistic bombardment. The invention also is embodied in a method of breeding plant species to obtain altered levels of poly-unsaturated fatty acids, specifically linolenic (18:3) acid in seed oil of oil-producing plants. This method involves (a) making a cross between two varieties of an oilseed plant differing in the linolenic acid trait; (b) making a Southern blot of restriction enzyme digested genomic DNA isolated from several progeny plants resulting from the cross of step (a); and (c) hybridizing the Southern blot with the radiolabeled nucleic acid fragments encoding the claimed glycerolipid desaturases. The invention is also embodied in a method of RFLP mapping that uses the isolated Arabidopsis thaliana delta-15 desaturase sequences described herein. The invention is also embodied in plants capable of producing altered levels of glycerolipid desaturase by virtue of containing the chimeric genes described herein. Further, the invention is embodied by seed oil obtained from such plants. The invention is also embodied in a method of RFLP mapping ina genomic RFLP marker comprising (a) making a cross between two varieties of plants; (b) making a Southern blot of restriction enzyme digested genomic DNA isolated from several progeny plants resulting from the cross of step (a); and (c) hybridizing the Southern blot with a radiolabelled nucleic acid fragments of the claimed fragments. The invention is also embodied in a method to isolate nucleic acid fragments encoding fatty acid desaturases and fatty acid desaturase-related enzymes, comprising (a)comparing SEQ ID NOS:2, 5, 7, 9, 11, 13, 15 and 17 with other fatty acid desaturase polypeptide sequences; (b) identifying the conserved sequence(s) of 4 or more amino acids obtained in step a; (c) making region-specific nucleotide probe(s) or oligomer(s) based on the conserved sequences identified in step b; and d) using the nucleotide probe(s) or oligomers(s) of step c to isolate sequences encoding fatty acid desaturases and fatty-acid desaturase-related enzymes by sequence-dependent protocols. The product of the method of isolation method described is also part of the invention. BRIEF DESCRIPTION OF THE SEQUENCE DESCRIPTIONS The invention can be more fully understood from the following detailed description and the Sequence Descriptions which form a part of this application. The Sequence Descriptions contain the one letter code for nucleotide sequence characters and the three letter code for amino acids in conformity with the IUPAC-IUB standard described in Nucleic Acids Research 13:3021-3030 (19085) and 37 C.F.R. 1.822 which are incorporated herein by reference. SEQ ID NO:1 shows the complete 5' to 3' nucleotide sequence of 1350 base pairs of the Arabidopsis cDNA which encodes delta-15 desaturase in plasmid pCF3. Nucleotides 46 to 48 are the putative initiation codon of the open reading frame (nucleotides 46 to 1206). Nucleotides 1204 to 1206 are the termination codon. Nucleotides 1 to 45 and 1207 to 1350 are the 5' and 3' untranslated nucleotides, respectively. The 386 amino acid protein sequence in SEQ ID NO:1 is that deduced from the open reading frame. SEQ ID NO:2 is the deduced peptide of the open-reading frame of SEQ ID NO:1. SEQ ID NO:3 is a partial nucleotide sequence of the Arabidopsis genomic DNA insert in plasmid pF1 which shows the genomic sequence in the region of the Arabidopsis genome that encodes delta-15 desaturase. Nucleotides 68-255 are identical to nucleotides 1-188 of SEQ ID NO:1. Nucleotides 47 to 49 and 56 to 58 are termination codons in the same reading frame as the open reading frame in SEQ ID NO:1. SEQ ID NO:4 shows the 5' to 3' nucleotide sequence of the insert in plasmid pACF2-2 of 1525 base pairs of the Arabidopsis thaliana cDNA that encodes a plastid delta-15 fatty acid desaturase. Nucleotides 10-12 and nucleotides 1348 to 1350 are, respectively, the putative initiation codon and the termination codon of the open reading frame (nucleotides 10 to 1350). Nucleotides 1 to 9 and 1351 to 1525 are, respectively, the 5' and 3' untranslated nucleotides. SEQ ID NO:5 is the deduced peptide of the open reading frame of SEQ ID NO:4. SEQ ID NO:6 shows the complete 5' to 3' nucleotide sequence of 1336 base pairs of the Brassica napus seed cDNA, found in plasmid pBNSF3-2, which encodes a microsomal delta-15 glycerolipid desaturase. Nucleotides 79 to 82 are the putative initiation codon of the open reading frame (nucleotides 79 to 1212). Nucleotides 1210 to 1212 are the termination codon. Nucleotides 1 to 78 and 1213 to 1336 are the 5' and 3' unstranslated nucleotides respectively. SEQ ID NO:7 is the deduced peptide of the open reading frame of SEQ ID NO:6. SEQ ID NO:8 is the complete 5' to 3' nucleotide sequence of 1416 base pairs of the Brassica napus seed cDNA found in plasmid pBNSFd-2 which encodes a plastid delta-15 glycerolipid desaturase. Nucleotides 1 to 1215 correspond to a continuous open reading frame of 404 amino acids. Nucleotides 1213 to 1215 are the termination codon. Nucleotides 1215 to 1416 are the 3' untranslated nucleotides. SEQ ID NO:9 is the deduced peptide of the open reading frame of SEQ ID NO:8. SEQ ID NO:10 is the complete nucleotide sequence of the soybean (glycine max) microsomal delta-15 desaturase cDNA, found in plasmid pXF1, which the 2184 nucleotides of this sequence contain-both the coding sequence and the 5' and 3' non-translated regions of the cDNA. Nucleotides 855 to 857 are the putative initiation codon of the open reading frame (nucleotides 855 to 2000). Nucleotides 1995 to 1997 are the termination codon. Nucleotides 1 to 854 and 1998 to 2184 are the 5' and 3' unstranslated nucleotides respectively. The 380 amino acid protein sequence in SEQ ID NO:7 is that deduced from the open reading frame. SEQ ID NO:11 is the deduced peptide of the open reading frame in SEQ ID NO:10. SEQ ID NO:12 is the complete 5' to 3' nucleotide sequence of 1676 base pairs of the soybean (Glycine max) seed cDNA found in plasmid pSFD-118bwp which encodes a soybean plastid delta-15 desaturase. Nucleotides 169 to 1530 correspond to a continuous open reading frame of 453 amino acids. Nucleotides 169 to 171 are the putative initiation codon of the open reading frame. Nucleotides 1528 to 1530 are the termination codon. Nucleotides 1531 to 1676 are the 3' untranslated nucleotides. Nucleotides 169 to 382 encode the putative plastid transit peptide, based on comparison of the deduced peptide with the soybean microsomal delta-15 peptide. SEQ ID NO:13 is the deduced peptide of the open reading frame in SEQ ID NO:12. SEQ ID NO:14 is the complete nucleotide sequence of a 396 bp polymerase chain reaction product derived from corn seed mRNA that is found in the insert of plasmid pPCR20. Nucleotides 1 to 31 and 364 to 396 correspond to the amplification primers described in SEQ ID NO:18 and SEQ ID NO:19, respectively. Nucleotides 31 to 363 encode an internal region of a corn seed delta-15 desaturase that is 61.9% identical to the region between amino acids 137 and 249 of the Brassica napus delta-15 desaturase peptide sequence shown in SEQ ID NO:7. SEQ ID NO:15 is the deduced amino acid sequence of SEQ ID NO:14. SEQ ID NO:16 shows the partial composite 5' to 3' nucleotide sequence of 472 bp derived from the inserts in plasmids pFadx-2 and pYacp7 for Arabidopsis thaliana cDNA that encodes a plastid delta-15 fatty acid desaturase. Nucleotides 2-4 and nucleotides 468 to 470 are, respectively, the first and the last codons in the open reading frame. SEQ ID NO:17 is deduced partial peptide sequence of the open reading frame in SEQ ID NO:16. SEQ ID NO:18 One hundred and twenty eight fold degenerate sense 31-mer PCR primer. Nucleotides 1 to 8 correspond to the Bam H1 restriction enzyme recognition sequence. Nucleotides 9 to 137 correspond to amino acid residues 130 to 137 of SEQ ID NO:6 with a deoxyinosine base at nucleotide 11. SEQ ID NO:19 Two thousand and forty eight-fold degenerate antisense 35-mer PCR primer. Nucleotides 1 to 8 correspond to the Bam H1 restriction enzyme recognition sequence. Nucleotides 9 to 35 correspond to amino acid residues 249 to 256 of SEQ ID NO:6 with a deoxyinosine base at nucleotide 15. SEQ ID NO:20 Sixteen-fold degenerate sense 36-mers made to amino acid residues 97-108 in SEQ ID NO:2. SEQ ID NO:21 Sixteen-fold degenerate sense 36-mers made to amino acid residues 97-108 in SEQ ID NO:2. SEQ ID NO:22 Seventy two-fold degenerate sense 18-mers made to amino acid residues 100-105 in SEQ ID NO:2. SEQ ID NO:23 Seventy two-fold degenerate sense 18-mers made to amino acid residues 100-105 in SEQ ID NO:2. SEQ ID NO:24 Seventy two-fold degenerate antisense 18-mers made to amino acid residues 299-304 in SEQ ID NO:2. SEQ ID NO:25 Seventy two-fold degenerate antisense 18-mers made to amino acid residues 299-304 in SEQ ID NO:2. SEQ ID NO:26 Seventy two-fold degenerate antisense 18-mers made to amino acid residues 304-309 in SEQ ID NO:2. SEQ ID NO:27 Seventy two-fold degenerate antisense 18-mers made to amino acid residues 304-309 in SEQ ID NO:2. SEQ ID NO:28 Sixteen-fold degenerate sense 36-mers made to amino acid residues 97-108 in SEQ ID NO:2. SEQ ID NO:29 Sixteen-fold degenerate sense 36-mers made to amino acid residues 97-108 in SEQ ID NO:2. SEQ ID NO:30 Sixty four-fold degenerate antisense 38-mers made to amino acid residues 299-311 in SEQ ID NO:2. SEQ ID NO:31 Sixty four-fold degenerate antisense 38-mers made to amino acid residues 299-311 in SEQ ID NO:2. SEQ ID NO:32 A 135-mer made as an antisense strand to amino acid residues 97-141 in SEQ ID NO:2. DETAILED DESCRIPTION OF THE INVENTION Applicants have isolated nucleic acid fragments that encode plant fatty acid desaturases and that are useful in modifying fatty acid composition in oil-producing species by transformation. Thus, transfer of the nucleic acid fragments of the invention or a part thereof that encodes a functional enzyme, along with suitable regulatory sequences that direct the transciption of their mRNA, into a living cell will result in the production or over-production of plant fatty acid desaturases and will result in increased levels of unsaturated fatty acids in cellular lipids, including triacylglycerols. Transfer of the nucleic acid fragments of the invention or a part thereof, along with suitable regulatory sequences that direct the transciption of their antisense RNA, into plants will result in the inhibition of expression of the endogenous fatty acid desaturase that is substantially homologous with the transferred nucleic acid fragment and will result in decreased levels of unsaturated fatty acids in cellular lipids, including triacylglycerols. Transfer of the nucleic acid fragments of the invention or a part thereof, along with suitable regulatory sequences that direct the transciption of their mRNA, into plants may result in inhibition by cosuppression of the expression of the endogenous fatty acid desaturase gene that is substantially homologous with the transferred nucleic acid fragment and may result in decreased levels of unsaturated fatty acids in cellular lipids, including triacylglycerols. The nucleic acid fragments of the invention can also be used as restriction fragment length polymorphism (RFLP) markers in Arabidopsis genetic mapping and plant breeding programs. The nucleic acid fragments of the invention or oligomers derived therefrom can also be used to isolate other related glycerolipid desaturase genes using DNA, RNA, or a library of cloned nucleotide sequences from the same or different species by well known sequence-dependent protocols, including, for example, methods of nucleic acid hybridization and amplification by the polymerase chain reaction. Definitions In the context of this disclosure, a number of terms shall be used. The term "fatty acid desaturase" used herein refers to an enzyme which catalyzes the breakage of a carbon-hydrogen bond and the introduction of a carbon-carbon double bond into a fatty acid molecule. The fatty acid may be free or esterified to another molecule including, but not limited to, acyl-carrier protein, coenzyme A, sterols and the glycerol moiety of glycerolipids. The term "glycerolipid desaturases" used herein refers to a subset of the fatty acid desaturases that act on fatty acyl moieties esterified to a glycerol backbone. "Delta-12 desaturase" refers to a fatty acid desaturase that catalyzes the formation of a double bond between carbon positions 6 and 7 (numbered from the methyl end), (i.e., those that correspond to carbon positions 12 and 13 (numbered from the carbonyl carbon) of an 18 carbon-long fatty acyl chain or carbon positions 10 and 11 (numbered from the carbonyl carbon) of a 16 carbon-long fatty acyl chain). "Delta-15 desaturase" refers to a fatty acid desaturase that catalyzes the formation of a double bond between carbon positions 3 and 4 (numbered from the methyl end), (i.e., those that correspond to carbon positions 15 and 16 (numbered from the carbonyl carbon) of an 18 carbon-long fatty acyl chain and carbon positions 13 and 14 (numbered from the carbonyl carbon) of a 16 carbon-long fatty acyl chain). Examples of fatty acid desaturases include, but are not limited to, the microsomal delta-12 and delta-15 desaturases that act on phosphatidylcholine lipid substrates; the chloroplastic delta-12 and delta-15 desaturases that act on phosphatidyl glycerol and galactolipids; and other desaturases that act on such fatty acid substrates such as phospholipids, galactolipids, and sulfolipids. "Microsomal desaturase" refers to the cytoplasmic location of the enzyme, while "chloroplast desaturase" and "plastid desaturase" refer to the plastid location of the enzyme. These fatty acid desaturases may be found in a variety of organisms including, but not limited to, higher plants, diatoms, and various eukaryotic and prokaryotic microorganisms such as fungi and photosynthetic bacteria and algae. The term "homologous fatty acid desaturases" refers to fatty acid desaturases that catalyze the same desaturation on the same lipid substrate. Thus, microsomal delta-15 desaturases, even from different plant species, are homologous fatty acid desaturases. The term "heterologous fatty acid desaturases" refers to fatty acid desaturases that catalyze desaturations at different positions and/or on different lipid substrates. Thus, for example, microsomal delta-12 and delta-15 desaturases, which act on phosphatidylcholine lipids, are heterologous fatty acid desaturases, even when from the same plant. Similarly, microsomal delta-15 desaturase, which acts on phosphatidylcholine lipids, and chloroplast delta-15 desaturase, which acts on galactolipids, are heterologous fatty acid desaturases, even when from the same plant. It should be noted that these fatty acid desaturases have never been isolated and characterized as proteins. Accordingly the terms such as "delta-12 desaturase" and "delta-15 desaturase" are used as a convenience to describe the proteins encoded by nucleic acid fragments that have been isolated based on the phenotypic effects caused by their disruption. The term "fatty acid desaturase-related enzyme" refers to enzymes whose catalytic product may not be a carbon-carbon double bond but whose mechanism of action is similar to that of a fatty acid desaturase (that is, catalysis of the displacement of a carbon-hydrogen bond of a fatty acid chain to form a fatty-hydroxyacyl intermediate or end-product). This term is different from "related fatty acid desaturases", which refers to structural similarities between fatty acid desaturases. The term "nucleic acid" refers to a large molecule which can be single-stranded or double-stranded, composed of monomers (nucleotides) containing a sugar, a phosphate and either a purine or pyrimidine. A "nucleic acid fragment" is a fraction of a given nucleic acid molecule. In higher plants, deoxyribonucleic acid (DNA) is the genetic material while ribonucleic acid (RNA) is involved in the transfer of the information in DNA into proteins. A "genome" is the entire body of genetic material contained in each cell of an organism. The term "nucleotide sequence" refers to the sequence of DNA or RNA polymers, which can be single- or double-stranded, optionally containing synthetic, non-natural or altered nucleotide bases capable of incorporation into DNA or RNA polymers. The term "oligomer" refers to short nucleotide sequences, usually up to 150 bases long. "Region-specific nucleotide probes" refers to isolated nucleic acid fragments derived from a cDNA or gene using a knowledge of the amino acid regions conserved between different fatty-acid desaturases which may be used to isolate cDNAS or genes for other fatty-acid desaturases or fatty acid desaturase-related enzymes using sequence dependent protocols. As used herein, the term "homologous to" refers to the relatedness between the nucleotide sequence of two nucleic acid molecules or between the amino acid sequences of two protein molecules. Estimates of such homology are provided by either DNA-DNA or DNA-RNA hybridization under conditions of stringency as is well understood by those skilled in the art (Hames and Higgins, Eds. (1985) Nucleic Acid Hybridisation, IRL Press, Oxford, U.K.); or by the comparison of sequence similarity between two nucleic acids or proteins, such as by the method of Needleman et al. (J. Mol. Biol. (1970) 48:443-453). As used herein, "substantially homologous" refers to nucleotide sequences that have more than 90% overall identity at the nucleotide level with the coding region of the claimed sequence, such as genes and pseudo-genes corresponding to the coding regions. The nucleic acid fragments described herein include molecules which comprise possible variations, both man-made and natural, such as but not limited to (a) those that involve base changes that do not cause a change in an encoded amino acid, or (b) which involve base changes that alter an amino acid but do not affect the functional properties of the protein encoded by the DNA sequence, (c) those derived from deletions, rearrangements, amplifications, random or controlled mutagenesis of the nucleic acid fragment, and (d) even occasional nucleotide sequencing errors. "Gene" refers to a nucleic acid fragment that expresses a specific protein, including regulatory sequences preceding (5' non-coding) and following (3' non-coding) the coding region. "Fatty acid desaturase gene" refers to a nucleic acid fragment that expresses a protein with fatty acid desaturase activity. "Native" gene refers to an isolated gene with its own regulatory sequences as found in nature. "Chimeric gene" refers to a gene that comprises heterogeneous regulatory and coding sequences not found in nature. "Endogenous" gene refers to the native gene normally found in its natural location in the genome and is not isolated. A "foreign" gene refers to a gene not normally found in the host organism but that is instead introduced by gene transfer. "Pseudo-gene" refers to a genomic nucleotide sequence that does not encode a functional enzyme. "Coding sequence" refers to a DNA sequence that codes for a specific protein and excludes the non-coding sequences. It may constitute an "uninterrupted coding sequence", i.e., lacking an intron or it may include one or more introns bounded by appropriate splice junctions. An "intron" is a nucleotide sequence that is transcribed in the primary transcript but that is removed through cleavage and re-ligation of the RNA within the cell to create the mature mRNA that can be translated into a protein. "Initiation codon" and "termination codon" refer to a unit of three adjacent nucleotides in a coding sequence that specifies initiation and chain termination respectively, of protein synthesis (mRNA translation). "Open reading frame" refers to the coding sequence uninterrupted by introns between initiation and termination codons that encodes an amino acid sequence. "RNA transcript" refers to the product resulting from RNA polymerase-catalyzed transcription of a DNA sequence. When the RNA transcript is a perfect complementary copy of the DNA sequence, it is referred to as the primary transcript or it may be a RNA sequence derived from posttranscriptional processing of the primary transcript and is referred to as the mature RNA. "Messenger RNA (mRNA)" refers to the RNA that is without introns and that can be translated into protein by the cell. "cDNA" refers to a double-stranded DNA that is complementary to and derived from mRNA. "Sense" RNA refers to RNA transcript that includes the mRNA. "Antisense RNA" refers to a RNA transcript that is complementary to all or part of a target primary transcript or mRNA and that blocks the expression of a target gene by interfering with the processing, transport and/or translation of its primary transcript or mRNA. The complementarity of an antisense RNA may be with any part of the specific gene transcript, i.e., at the 5' non-coding sequence, 3' non-coding sequence, introns, or the coding sequence. In addition, as used herein, antisense RNA may contain regions of ribozyme sequences that increase the efficacy of antisense RNA to block gene expression. "Ribozyme" refers to a catalytic RNA and includes sequence-specific endoribonucleases. As used herein, "suitable regulatory sequences" refer to nucleotide sequences in native or chimeric genes that are located upstream (5'), within, and/or downstream (3') to the nucleic acid fragments of the invention, which control the expression of the nucleic acid fragments of the invention. The term "expression", as used herein, refers to the transcription and stable accumulation of the sense (mRNA) or the antisense RNA derived from the nucleic acid fragment(s) of the invention that, in conjunction with the protein apparatus of the cell, results in altered levels of the fatty acid desaturase(s). Expression or overexpression of the gene involves transcription of the gene and translation of the mRNA into precursor or mature fatty acid desaturase proteins. "Antisense inhibition" refers to the production of antisense RNA transcripts capable of preventing the expression of the target protein. "Overexpression" refers to the production of a gene product in transgenic organisms that exceeds levels of production in normal or non-transformed organisms. "Cosuppression" refers to the expression of a foreign gene which has substantial homology to an endogenous gene resulting in the suppression of expression of both the foreign and the endogenous gene. "Altered levels" refers to the production of gene product(s) in transgenic organisms in amounts or proportions that differ from that of normal or non-transformed organisms. "Promoter" refers to a DNA sequence in a gene, usually upstream (5') to its coding sequence, which controls the expression of the coding sequence by providing the recognition for RNA polymerase and other factors required for proper transcription. In artificial DNA constructs promoters can also be used to transcribe antisense RNA. Promoters may also contain DNA sequences that are involved in the binding of protein factors which control the effectiveness of transcription initiation in response to physiological or developmental conditions. It may also contain enhancer elements. An "enhancer" is a DNA sequence which can stimulate promoter activity. It may be an innate element of the promoter or a heterologous element inserted to enhance the level and/or tissue-specificity of a promoter. "Constitutive promoters" refers to those that direct gene expression in all tissues and at all times. "Tissue-specific" or "development-specific" promoters as referred to herein are those that direct gene expression almost exclusively in specific tissues, such as leaves or seeds, or at specific development stages in a tissue, such as in early or late embryo-genesis, respectively. The "3' non-coding sequences" refers to the DNA sequence portion of a gene that contains a poly-adenylation signal and any other regulatory signal capable of affecting mRNA processing or gene expression. The polyadenylation signal is usually characterized by affecting the addition of polyadenylic acid tracts to the 3' end of the mRNA precursor. The term "Transit Peptide" refers to the N-terminal extension of a protein that serves as a signal for uptake and transport of that protein into an organelle such as a plastid or mitochondrion. "Transformation" herein refers to the transfer of a foreign gene into the genome of a host organism and its genetically stable inheritance. "Restriction fragment length polymorphism" refers to different sized restriction fragment lengths due to altered nucleotide sequences in or around variant forms of genes. "Fertile" refers to plants that are able to propagate sexually. "Oil-producing species" herein refers to plant species which produce and store triacylglycerol in specific organs, primarily in seeds. Such species include soybean (Glycine max), rapeseed and canola (including Brassica napus, B. campestris), sunflower (Helianthus annus), cotton (Gossypium hirsutum), corn (Zea mays), cocoa (Theobroma cacao), safflower (Carthamus tinctorius), oil palm (Elaeis guineensis), coconut palm (Cocos nucifera), flax (Linum usitatissimum), castor (Ricinus communis) and peanut (Arachis hypogaea). The group also includes non-agronomic species which are useful in developing appropriate expression vectors such as tobacco, rapid cycling Brassica species, and Arabidopsis thaliana, and wild species which may be a source of unique fatty acids. "Sequence-dependent protocols" refer to techniques that rely on a nucleotide sequence for their utility. Examples of sequence-dependent protocols include, but are not limited to, the methods of nucleic acid and oligomer hybridization and methods of DNA and RNA amplification such as are exemplified in various uses of the polymerase chain reaction. "PCR product" refers to the DNA product obtained through polymerase chain reaction. Various solutions used in the experimental manipulations are referred to by their common names such as "SSC", "SSPE", "Denhardt's solution", etc. The composition of these solutions may be found by reference to Appendix B of Sambrook, et al. (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989), Cold Spring Harbor Laboratory Press). T-DNA Mutagenesis and Identification of an Arabidopsis Mutant Defective in Delta-15 Desaturation In T-DNA mutagenesis (Feldmann, et al., Science (1989) 243:1351-1354), the integration of T-DNA in the genome can interrupt normal expression of the gene at or near the site of the integration. If the resultant mutant phenotype can be detected and shown genetically to be tightly linked to the T-DNA insertion, then the "tagged" locus and its wild type counterpart can be readily isolated by molecular cloning by one skilled in the art. Armbidopsis thaliana seeds were transformed by Agrobacterium tumefaciens C58C1rif strain harboring the avirulent Ti-plasmid pGV3850::pAK1003 that has the T-DNA region between the left and right T-DNA borders replaced by the origin of replication region and ampicillin resistance gene of plasmid pBR322, a bacterial kanamycin resistance gene, and a plant kanamycin resistance gene (Feldmann, et al., Mol. Gen. Genetics (1987) 208:1-9). Plants from the treated seeds were self-fertilized and the resultant progeny seeds, germinated in the presence of kanamycin, were self-fertilized to give rise to a population, designated T3, that was segregating for T-DNA insertions. T3 seeds from approximately 6000 T2 plants were analyzed for fatty acid composition. One line, designated 3707, showed a reduced level of linolenic acid (18:3). One more round of self-fertilization of mutant line 3707 produced T4 progeny seeds. The ratio of 18:2/18:3 in seeds of the homogyzous mutant in T4 population was ca. 14; this ratio is ca 1.8 and ca. 23, respectively, in wild-type Arabidopsis and Arabidopsis fad 3 mutant ›Lemieux et al. (1990) Theor. App. Gen. 80:234-240! obtained via chemical mutagenesis. These seeds were planted and 263 individual plants were analyzed for the presence of nopaline in leaf extracts. T5 seeds from these plants were further analyzed for fatty acid composition and the ability to germinate in the presence of kanamycin. The mutant fatty acid phenotype was found to segregate in a 1:2:1 ratio, as was germinability on kanamycin. Nopaline was found in all plants with an altered fatty acid phenotype, but not in wild type segregants. These results provided evidence that the locus controlling delta-15 desaturation was interrupted by T-DNA in mutant line 3707. Isolation of Arabidopsis Genomic DNA Containing the Gene Controlling Delta-15 Desaturation In order to isolate the gene controlling delta-15 desaturation from wild-type Arabidopsis, a T-DNA-plant DNA "junction" fragment containing a T-DNA border integrated into the host plant DNA was isolated from Arabidopsis mutant 3707. For this, genomic DNA from the mutant plant was isolated and completely digested by either Bam HI or Sal I restriction enzymes. In each case, one of the resultant fragments was expected to contain the origin of replication and ampicillin-resistance gene of pBR322 as well as the left T-DNA-plant DNA junction fragment. Such fragments were rescued as plasmids by ligating the digested genomic DNA fragments at a dilute concentration to facilitate self-ligation and then using the ligated fragments to transform E. coli cells. Ampicillin-resistant E. coli transformants were isolated and screened by colony hybridization to fragments containing either the left or the right T-DNA border. Of the 192 colonies obtained from the plasmid rescue of Sal I digested genomic DNA, 31 hybridized with the left T-DNA border fragment, 4 hybridized to the right T-DNA border fragment, and none hybridized to both. Of the 85 colonies obtained from the plasmid rescue of Bam HI digested genomic DNA, 63 hybridized to the left border and none to the right border. Restriction analysis of seven rescued plasmids that were obtained from the Bam HI digestion and that hybridized to the left T-DNA border showed that they were indistinguishable and contained 1.4 kb of putative, flanking plant DNA. Restriction analysis of another rescued plasmid, pS1, that was obtained from the Sal I digestion and hybridized only to the left T-DNA border, showed that it contained 2.9 kb of putative, flanking plant DNA. This flanking DNA had a Bam HI site and a Hind III site 1.4 kb and 2.2 kb, respectively, away from the left T-DNA border, suggesting that the 1.4 kb putative plant DNA in Bam HI rescued plasmids was contained within the 2.9 kb putative plant DNA in the Sal I rescued plasmids. Southern blot analysis of wild type and mutant 3707 Arabidopsis genomic DNA using the radiolabeled 1.4 kb DNA fragment as the hybridization probe confirmed that this fragment contained plant DNA and that the T-DNA integration site was in a 2.8 kb Bam HI, a 5.2 kb Hind III, a 3.5 kb Sal I, a 5.5 kb Eco RI, and an approximately 9 kb Cla I fragment of wild type Arabidopsis DNA. Nucleotide sequencing of plasmid pS1 with a primer made to a left T-DNA border sequence revealed that pS1 was colinear with the sequence of the left T-DNA border (Yadav et al., Proc. Natl. Acad. Sci. USA (1982) 79:6322-6326) up to nucleotide position 65, which is in the T-DNA border repeats. Approximately 800 bp of additional sequence in pS1 beyond the T-DNA-plant DNA junction, that is, in the plant DNA adjoining the left T-DNA border, showed no significant homology to the T-DNA of pGV3850::pAK1003 and no significant open reading frame. The nucleic acid fragment from wild-type Arabidopsis corresponding to the plant DNA flanking T-DNA in the line 3707 was isolated by screening a lambda phage Arabidopsis thaliana genomic library with the 1.4 kb plant DNA isolated from the rescued plasmids as a hybridization probe. Seven positively-hybridizing genomic clones were isolated that fell in one of five classes based on partial restriction mapping. While their average insert size was approximately 15 kb, taken together they spanned a total of approximately 40 kb of genomic DNA. A combination of restriction and Southern analyses revealed that the five clones overlapped the site of integration of the left border of the T-DNA and that there was no detectable rearrangement of plant DNA in the rescued plasmids as compared to that in the wild type genomic plant DNA. One of these lambda phage clones, designated 1111, was representative of the recovered clones and contained an approximately 20 kb genomic DNA insert which was more or less symmetrically arranged around the site of insertion of the left border of the T-DNA. This clone was deposited on November 27, 1991 with the American Type Culture Collection of Rockville, Maryland under the provisions of the Budapest Treaty and bears accession number ATCC 75167. Isolation of Arabidopsis Delta-15 Desaturase cDNA A 5.2 kb Hind III fragment containing wild-type genomic DNA, which hybridized to the 1.4 kb flanking plant DNA recovered from line 3707 and which was interrupted near its middle by the T-DNA insertion in line 3707, was isolated from lambda phage clone 41A1 and cloned into the Hind III site of the pBluescript SK vector (Stratagene) by standard cloning procedures described in Sambrook et al., Molecular Cloning, A Laboratory Manual, 2nd ed. (1989), Cold Spring Harbor Laboratory Press). The resultant plasmid was designated pF1. The isolated 5.2 kb Hind III fragment was also used as a radiolabeled hybridization probe to screen a cDNA library made to poly A.sup.+ mRNA from 3-day-old etiolated Arabidopsis thaliana (ecotype Columbia) seedling hypocotyls in a lambda ZAP II vector (Stratagene). Of the several positively-hybridizing plaques, four strongly-hybridizing ones were subjected to plaque purification. Sequences of the pBluescript (Stratagene) vector, including the cDNA inserts, from each of the purified phage stocks were excised in the presence of a helper phage. The resultant phagemids were used to infect E. coli cells which yielded double-stranded plasmids, pCF1, pCF2, pCF3, and pCF4. All four were shown to contain at least one approximately 1.3 to 1.4 kb Not I insert fragment (Not I/Eco RI adaptors were used in the preparation of the cDNA library) which hybridized to the same region of wild-type plant genomic DNA present in the isolated phage clones. This region, which was near the site of integration of the left T-DNA border in line 3707, was on the side of the T-DNA insertion opposite to that of the plant DNA flanking the left T-DNA border isolated previously via plasmid rescue. Partial sequence determination of the different cDNAs revealed common identity. Since multiple versions of only one type of cDNA were obtained from a cDNA library made from etiolated tissue which is expected to express delta-15 desaturation, and since these cDNAs hybridized to the genomic DNA that corresponds to the site of T-DNA integration in line 3707 which had a high linoleic acid/low linolenic acid phenotype, Applicants were lead to conclude that the T-DNA in line 3707 interrupted the normal expression of the gene encoding delta-15 desaturase. The complete nucleotide sequence of one cDNA, designated pCF3, was determined and is shown as SEQ ID NO:1. It reveals an open reading frame that encodes a 386 amino acid polypeptide. One of the sequencing primers made to the pCF3 insert was also used to obtain 255 bp of sequence from pF1 that is shown as SEQ ID NO:3. Nucleotides 68 to 255 of the genomic DNA in pF1 (SEQ ID NO:3) are identical to nucleotides 1 to 188 of the cDNA (SEQ ID NO:1), which shows that they are colinear and that the cDNA is encoded for by the gene in the isolated genomic DNA. Nucleotides 113 to 115 in SEQ ID NO:3 are the initiation codon of the largest open reading frame corresponding to nucleotides 46-48 in SEQ ID NO:1. This is evident from the presence of in-frame termination codons at nucleotides 47 to 49 and nucleotides 56 to SB and the absence of observable intron splice junctions in SEQ ID NO:3. The identification of the 386 amino acid polypeptide as a desaturase was confirmed by comparing its amino acid sequence with all the protein sequences found in Release 19.0 of the SWISSPROTEIN database using the FASTA algorithm of Pearson and Lipman (Proc. Natl. Acad. Sci. USA (1988) 85:2444-2448) and the BLAST program (Altschul et al., J. Mol. Biol. (1990) 215:403-410). The most homologous protein found in both searches was the desA fatty acid desaturase from the cyanobacterium Synechocystis PCC6803 (Wada, et al., Nature (1990) 347:200-203; Genbank ID:CSDESA; GenBank Accession No:X53508). The 386 amino acid peptide in SEQ ID NO:1 was also compared to the 351 amino acid sequence of desA by the method of Needleman et al. (J. Mol. Biol. (1970) 48:443-453). Over their entire length, these proteins were 26% identical, the comparison imposing four major gaps in the desA protein sequence. While this overall homology is poor, homology in shorter stretches was better. For instance, in a stretch of 78 amino acids the Arabidopsis delta-15 desaturase (amino acids 78 to 155 in SEQ ID NO:1) and the desA protein (amino acids 67 to 144) showed 40% identity and 66% similarity. Homology in yet shorter stretches was even greater as shown in Table 2. TABLE 2 ______________________________________ Peptide AA positions AA positions Percent Length in SEQ ID NO: 1 in desA Identity ______________________________________ 12 97-108 86-97 83 7 115-121 104-110 71 9 133-141 22-130 56 11 299-309 282-292 64 ______________________________________ These high percent identities in short stretches of amino acids between the cyanobacterial desaturase polypeptide and SEQ ID NO:2 suggests significant relatedness between the two. To analyse the developmental expression of the gene encoding mRNA coresponding to SEQ ID NO:1, the cDNA insert in plasmid pCF3 was used as a radiolabeled hybridization probe on mRNA samples from leaf, root, germinating seedling, and developing siliques from both wild type amd mutant 3707 Arabidopsis plants, essentially as described in Maniatis et al., Molecular Cloning, A Laboratory Manual (1982) Cold Spring Harbor Laboratory Press. The results indicated that while the mRNA corresponding to SEQ ID NO:1 is detected in all tissues from the mutant plant, its levels are lower than in wild-type tissues. This is consistent with the observation that the fatty acid mutation in line 3707 is leaky relative to the known Arabidosis fad 3 mutant obtained via chemical mutagenesis. These results confirmed that the T-DNA in line 3707 had interrupted the normal expression of a fatty acid desaturase gene. Based on the fatty acid phenotype of homozygous mutant line 3707, Applicants concluded that the cDNA insert in pCF3 encoded the delta-15 desaturase. Further, Applicants concluded that it was the microsomal delta-15 desaturase, and not the chloroplastic delta-15 desaturase, since: a) the mutant phenotype was expressed strongly in the seed but expressed poorly, if at all, in the leaf of line 3707, and b) the delta-15 desaturase polypeptide, by comparison to the desA polypeptide, did not have an N-terminal extension of a transit peptide expected for a nuclear-encoded chloroplast desaturase. The identity of SEQ ID NO:2 as the Arabidopsis microsomal delta-15 desaturase was confirmed by its biological overexpression in plant tissues. For this, the 1.4 kB Not I fragment of plasmid pCF3 containing the delta-15 desaturase cDNA was placed in the sense orientation behind either the CaMV 35S promotor, to provide constituitive expression, or behind the promotor for the gene encoding soybean a' subunit of the .beta.-conglycinin (75) seed storage protein, to provide embryo-specific expression. The chimeric genes 35S promoter/sense SEQ ID NO:1/3' nopaline synthase and .beta.-conglycinin/sense SEQ ID NO:1/3' phaseolin were then transformed into plant cells by Agrobacterium tumefaciens's binary Ti plasmid vector system ›Hoekema et al. (1983) Nature 303:179-180; Bevan (1984) Nucl. Acids Res. 12:8711-8720). To confirm the identity of SEQ ID NO:1 and to test the biological effect of its overexpression in a heterologous plant species, the chimeric genes 35S promoter/sense SEQ ID NO:1/3' nopaline synthase was transformed into a binary vector, which was then transferred into Agrobacterium tumefaciens strain R1000, carrying the Ri plasmid pRiA4b from Agrobacterium rhizogenes ›Moore et al. (1979) Plasmid 2:617-626!. Carrot (Daucus carota L.) cells were transformed by co-cultivation of carrot root disks with strain R1000 carrying the chimeric gene by the method of Petit et al. (1986) ›Mol. Gen. Genet. 202:388-393!. Fatty acid analyses of transgenic carrot "hairy" roots show that overexpression of Arabidopsis microsomal delta-15 desaturase can result in over 10-fold increase in 18:3 at the expense of 18:2. To complement the delta-15 desaturation mutation in the T-DNA mutant line 3707 and to test the biological effect of overexpression of SEQ ID NO:1 in seed, the embryo-specific promoter/SEQ ID NO:1/3' phaseolin chimeric gene was transformed into a binary vector, which was then transformed into the avirulent Agrobacterium strain LBA4404/pAL4404 ›Hoekema et al. (1983) Nature 303:179-180!. Roots of line 3707 were transformed by the engineered Agrobacterium, transformed plants were selected and grown to give rise to seeds. Fatty acid analysis of the seeds from two plants showed that the one out of six seeds in each plant showed the mutant fatty acid phenotype, while the remaining seeds show more than 10-fold increase in 18:3 to ca. 55%. While the sample size is small, this segregation suggests Mendelian inheritance of the fatty acid phenotype. While most of the increase occurs at the expense of 18:2, some of it also occurs at the expense of 18:1. Thus, overexpression of this gene in oils crops, especially canola, which is a close relative of Arabidopsis, is also expected to result in the high levels of 18:3 that are found in specialty oil of linseed. Comparisons of the sequence of the 386 amino acid polypeptide by the method of Needleman et al. (J. Mol. Biol. (1970) 48:443-453) with those for the microsomal stearoyl-CoA (delta-9) desaturases from rat, mouse and yeast revealed 21%, 19%, and 17% identities, respectively. While the membrane-associated Arabidopsis delta-15 desaturase protein showed significant but limited homology to the desA protein, it showed no significant homology to the soluble stearoyl-ACP (delta-9) desaturases from higher plants, including one from Arabidopsis. Comparison of partial nucleotide sequences of plasmids pF1 and pS1 showed that the left T-DNA border:plant DNA junction is ca. 700 bp from the initiaton codon in SEQ ID NO:1. To determine the position of the other T-DNA:plant DNA junction with respect to the pF1 sequence, the T-DNA:plant DNA junction fragment was isolated. Genomic DNA from mutant line 3707, isolated as described previously, was partially digested by restriction enzyme Mbo I to give an average fragment size of ca. 15 kB. The fragment ends were partially-filled with dGTP and gATP by Klenow and cloned into Xho I half-sites of LambdaGEM.RTM.-11 (Promega Corporation) following the manufacturer's protocol. The phage library was titered and used essentially as described in Ausubel et al. (Current Protocols in Molecular Biology (1989) John Wiley & Sons!. The genomic phage library was screened with radiolabeled PCR product, ca. 0.6 kB, derived from 5' end of the gene in pF1. This product spans from 3 bp to the right of where the left-T-DNA border inserted to 15 bp to the left of nucleotide position 1 in SEQ ID NO:1. Southern blot analysis of DNA from one of the purified, positively-hybridizing phages following Eco RI restriction digestion and electrophoresis showed that a 4 kB Eco RI fragment hybridized to the 0.6 kB PCR product. The Eco RI fragment was subcloned and subject to sequence analyses. Comparison of the sequences derived from this fragment, pF1 and pS1 showed that the insertion of T-DNA resulted in a 56 bp deletion at the site of insertion and that the T-DNA interrupted the Arabidopsis gene 711 bp 5' to the initiaton codon in SEQ ID NO:1. Thus, the T-DNA inserts 5' to the open reading frame, consistent with the leaky expresssion of the gene encoding SEQ ID NO:1 and the leaky fatty acid phenotype in mutant 3707. While the left T-DNA:plant DNA junction is precise, that is without any sequence rearrangement in either the left T-DNA border or the flanking plant DNA, the other T-DNA:plant DNA junction is complex and not fully characterized. Plasmid pCF3 was deposited on Dec. 3, 1991 with the American Type Culture Collection of Rockville, Md. under the provisions of the Budapest Treaty and bears accession number ATCC 68875. Using Arabidopsis Delta-15 Desaturase cDNA as a Hybridization Probe to Isolate cDNAs Encoding Related Desaturases from Arabidopsis The 1.4 kb Not I insert fragment isolated from plasmid pCF3 was purified, radiolabeled, and used to screen approximately 80,000 clones from the cDNA library made to poly A.sup.+ mRNA from 3-day-old etiolated Arabidopsis thaliana as described above, except that lower stringency hybridizations (1 M NaCl, 50 mM Tris-HCl, pH 7.5, 1% SDS, 5% dextran sulfate, 0.1 mg/mL denatured salmon sperm DNA and 50.degree. C.) and washes (sequentially with 2X SSPE, 0.1% SDS at room temperature for 5 min and then again with fresh solution for 10 min, and finally with 0.5X SSPE, 0.1% SDS at 50.degree. C. for 5 min.) were used. Approximately 17 strongly-hybridizing and 17 weakly-hybridizing plaques were identified in the primary screen. Four of the weakly-hybridizing plaques were picked and subjected to one or two further rounds of screening with the radiolabeled probe as above until they were pure. To ensure that these were not delta-15 desaturase clones, they were further analyzed to determine whether they hybridized to an 18 bp oligomer specific to the 3' non-coding region of delta-15 desaturase cDNA (pCF3). After autoradiography of the filters, one of the clones was found not to hybridize to this probe. This clone was picked, and a plasmid clone containing the cDNA insert was obtained as described above. Restriction analysis of this plasmid, designated pCM2, showed that it had an approximately 1.3 kb cDNA insert which lacked a 0.7 kb Nco I--Bgl II fragment characteristic of the Arabidopsis delta-15 desaturase cDNA of pCF3. (This fragment corresponds to the DNA located between the Nco I site at nucleotides 474 to 479 and the Bgl II site at nucleotides 1164 to 1169 in SEQ ID NO:1). Partial nucleotide sequences of single strands from the 5' region and 3' region of pCM2 revealed that the cDNA insert was incomplete and that it encoded a polypeptide that is similar to, but distinct from, that encoded by the cDNA in pCF3. In order to isolate a full-length version of the cDNA in plasmid pCM2, the 1.3 kB Not I fragment from plasmid pCM2 containing the cDNA insert was isolated and used as a radiolabeled hybridization probe to rescreen the same Arabidopsis cDNA library as above. Three strongly hybridizing plaques were purified and the plasmids excised as described previously. The three resultant plasmids were digested by Not I restriction enzyme and shown to contain cDNA inserts ranging in size between 1 kB and 1.5 kB. Complete nucleotide sequence determination of the cDNA insert in one of these plasmids, designated pACF2-2, is shown in SEQ ID NO:4. SEQ ID NO:4 shows the 5' to 3' nucleotide sequence of base pairs of the Arabidopsis thaliana cDNA which encodes a fatty acid desaturase. Nucleotides 10-12 and nucleotides 1358 to 1350 are, respectively, the putative initiation codon and the termination codon of the open reading frame (nucleotides 10 to 1350). The open reading frame was confirmed by comparison of its deduced amino acid sequences with that of the related delta-15 fatty acid desaturase from soybean in this application. Nucleotides 1 to 9 and 1351 to 1525 are, respectively, the 5' and 3' untranslated nucleotides. The 446 amino acid protein sequence in SEQ ID NO:5 is that deduced from the open reading frame in SEQ ID NO:4 and has an estimated molecular weight of 51 kD. Alignment of SEQ ID NOS:2 and 5 shows an overall homology of approximately 80% and that the former has an approximately 55 amino acid long N-terminal extension, which is deduced to be a transit peptide found in nuclear-encoded plastid proteins. To analyse the developmental expression of the gene corresponding to SEQ ID NO:4, this sequence was used as a radiolabeled hybridization probe on mRNA samples from leaf, root, germinating seedling, and developing siliques from both wild type and mutant line 3707 Arabidopsis plants, essentially as described in Maniatis et al. ›Molecular Cloning, A Laboratory Manual (1982) Cold Spring Harbor Laboratory Press!. The results indicated that, in contrast to the constitutive expression of the gene encoding SEQ ID NO:1, the mRNA corresponding to SEQ ID NO:4 is abundant in green tissues, rare in roots and leaves, and is about three-fold more abundant in leaf than that of SEQ ID NO:1. The cDNA in plasmid pCM2 was also shown to hybridize polymorphically to genomic DNA from Arabidopsis thaliana (ecotype Wassileskija and marker line W100 ecotype Landesberg background) digested with Eco RI. It was used as a RFLP marker to map the genetic locus for the gene encoding this fatty acid desaturase in Arabidopsis. A single genetic locus was positioned corresponding to this desaturase cDNA. Its location was thus determined to be on chromosome 3 between the lambda AT228 and cosmid c3838 RFLP markers, "north" of the glabrous locus (Chang et al., Proc. Natl. Acad. Sci. USA (1988) 85:6856-6860; Nam et al., Plant Cell (1989) 1:699-705). This approximates the region to which Arabidopsis fatty acid desaturase fad 2, fad D, and fad B mutations map ›Somerville et al., (1992) in press!. Unsuccessful efforts to clone the microsomal delta-12 fatty acid desaturase using cDNA inserts of pCF3 and pACF2-2 alongwith the above data led Applicants to conclude that the cDNA in pACF2-2 encodes a plastid delta-15 fatty acid desaturase that corresponds to the fad D locus. This conclusion will be confirmed by biological expression of the cDNA in pACF2-2. Plasmid pCM2 was deposited on Nov. 27, 1991 with the American Type Culture Collection of Rockville, Md. under the provisions of the Budapest Treaty and bears accession number ATCC 68852. The 1.4 kb, 1.3 kB, and 1.5 kB Not I cDNA insert fragments isolated from plasmids pCF3, pCM2 and pACF2-2 were purified, radiolabeled, and used several times to screen at low stringency as described above two different cDNA libraries: one was made to poly A.sup.+ mRNA from 3-day-old etiolated Arabidopsis thaliana ("etiolated" library) as described above and one made to polyA.sup.+ mRNA from the above-ground parts of Arabidopsis thaliana plants, which varied in size from those that had just opened their primary leaves to plants which had bolted and were flowering ›Elledge et al. (1991) Proc. Natl. Acad Sci. USA 88:1731-1735!. The cDNA inserts in the library were made into an Xho I site flanked by Eco RI sites in lambda Yes vector ›Elledge et al. (1991) Proc. Natl. Acad Sci. USA 88:1731-1735! ("leaf" library). Several plaques from both libraries that hybridized weakly and in duplicate lifts to both SEQ ID NOS:1 and 4 were subjected to plaque purification. Phagemids were excised from the pure phages from "etiolated" library as described above. Plasmids were excised from the purified phages of the "leaf" library by site-specific recombination using the cre-lox recombination system in E. coli strain BNN132 ›Elledge et al. (1991) Proc. Natl. Acad Sci. USA 88:1731-1735!. In all cases, nucleotide sequencing of the cloned DNA revealed clones either identical to SEQ ID NOS:1 or 4 or unrecognizable sequences. In another set of experiments ca. 400,000 phages in the "leaf" library was screened with SEQ ID NOS:1 and 4 at low stringency (26 C, 1 M Na.sup.+, 50% formamide) and high stringency (42 C, 1 M Na.sup.+, 50% formamide). Of the several positive signals on the primary plaque lifts, 11 showed high stringency hybridization to SEQ ID NO:1, 35 showed high stringency hybridization to SEQ ID NO:4, and 39 hybridized to both at low stringency only. Twenty seven plaques of the low stringency signals came through a secondary low-stringency screen, 17 of which were used to make DNA from excised plasmids. Of the 7 plasmid DNA were sequenced, 8 were unrecognizable sequences, 5 were identical to SEQ ID NO:1, 2 were identical to SEQ ID NO:2, and 2 were identical to one another and related but distinct to SEQ ID NOS:1 and 4. The novel desaturase sequence, designated pFad-x2, was also isolated from the "leaf" library independently by using as a hybridization probe a 0.6 kB PCR product derived by polymerase chain reaction on poly A.sup.+ RNA made from both canola seed as well as Arabidopsis leaves, as described elsewhere in this application, using degenerate oligomers made to conserved sequences between plant delta-15 desaturases and the cyanobacterial des A desaturase. The PCR-derived plasmid, designated pYacp7, was sequenced partially from both ends. Comparison of the sequences of pFad-x2 and pYacp7 revealed that the two independently cloned cDNAs contained an identical sequence that was related to the other delta-15 desaturases and that both were incomplete cDNAs. A partial composite sequence derived from both plasmids, pFadx-2 and pYacp7, is shown in SEQ ID NO:16 as a 5' to 3' nucleotide sequence of 472 bp. Nucleotides 2-4 and nucleotides 468 to 470 are, respectively, the first and the last codons in the open reading frame. This open reading frame is shown in SEQ ID NO:17. Comparison of SEQ ID NO:17 to the other delta-15 desaturase polypeptides disclosed in this application by the method of Needleman et al. ›J. Mol. Biol. (1970) 48:443-453)! using gap weight and gap length weight values of 3.0 and 0.1, respectively. The overall identities are between 65% and 68% between SEQ ID NO:17 and the microsomal delta-15 desaturases from Arabidopsis, canola and soybean and the overall identities are between 77% and 87% between SEQ ID NO:17 and the plastid delta-15 desaturases from Arabidopsis, canola and soybean. In addition SEQ ID NO:17 has an N-terminal peptide extension compared to the microsomal delta-15 desaturases that shows homology of the transit peptide sequence in Arabidopsis plastid delta-15 desaturase. On the basis of these comparisons it is deduced that SEQ ID NO:16 encodes a plastid delta-15 desaturase. There is genetic data in Arabidopsis suggesting the presence of two loci for plastid delta-15 desaturase. The full-length version of SEQ ID NO:16 can be readily isolated by one skilled in the art. The biological effect of introducing SEQ ID NO:16 or its full-length version into plants will be used to confirm its identity. Plasmid pYacp7 was deposited on Nov. 20, 1992 with the American Type Culture Collection of Rockville, Maryland under the provisions of the Budapest Treaty and bears accession number ATCC 69129. Using Arabidopsis Delta-15 Desaturase cDNAs as Hybridization Probes to Isolate Delta-15 Desaturase cDNAs from Other Plant Species For the purpose of cloning the Brassica napus seed cDNAs encoding delta-15 fatty acid desaturases, the cDNA inserts from pCF3 and pCM2 were isolated by polymerase chain reaction from the respective plasmids, radiolabeled, and used as hybridization probes to screen a lambda phage cDNA library made with poly A.sup.+ mRNA from developing Brassica napus seeds 20-21 days after pollination. This cDNA library was screened several times at low stringency, using the Arabidopsis cDNA probes mentioned above. One of the Brassica napus cDNAs obtained in the initial screens was used as probe in a subsequent high stringency screen. Arabidopsis pCM2 insert was radiolabeled and used as probe to screen approximately 300,000 plaques under low stringency hybridization conditions. The filter hybridizations were performed in 50 mM Tris pH 7.6, 6X SSC, 5X Denhardt's, 0.5% SDS, 100 ug denatured calf thymus DNA at 50.degree. C. overnight, and the posthybridization washes were carried out in 6X SSC, 0.5% SDS at room temperature for 15 min, then repeated with 2X SSC, 0.5% SDS at 45.degree. C. for 30 min, and then repeated twice with 0.2X SSC, 0.5% SDS at 50.degree. C. for 30 min. Five strongly-hybridizing phages were obtained. These were plaque purified and used to excise the phagemids as described in the manual of the pBluescriptII Phagemid Kit from Stratagene (Stratagene 1991 catalogue, item 212205). One of these, designated pBNSF3-2, contained a 1.3 kb insert. pBNSF3-f2 was sequenced completely on both strands and the nucleotide sequence is shown in SEQ ID NO:6. Plasmid pBNSF3-2 was deposited on Nov. 27, 1991 with the American Type Culture Collection of Rockville Md., USA under the provisions of the Budapest Treaty and bears the accession number 68854. An additional low stringency screen using pCM2 probe provided eight strongly hybridizing phages. One of these, designated pBNSFd 8, contained a 0.4kb insert. pBNSFd-8 was sequenced completely on one strand, this nucleotide sequence showed significant divergence from the sequence SEQ ID NO:6 in the homologous region, which suggested that it corresponded to a novel Brassica napus seed desaturase different from that shown in SEQ ID NO:6. pBNSFd-8 insert was radiolabelled and used as hybridization probe in a high stringency screen of the Brassica napus seed cDNA library. The hybridization conditions were identical to those of the low stringency screen described above except for the temperature of the final two 30 min posthybridization washes in 0.2.times.SSC, 0.5% SDS was increased to 60.degree. C. This screen resulted in three strongly hybridizing phages that were purified and excised. One of the excised plasmids pBNSFd-3 contained a 1.4 kb insert that was sequenced completely on both strands. SEQ ID NO:8 shows the complete nucleotide sequence of pBNSFd-2. Using Arabidopsis Delta-15 Desaturase cDNA as a Hybridization Probe to Isolate a Glycerolipid Desaturase cDNA from Soybean A cDNA library was made to poly A.sup.+ mRNA isolated from developing soybean seeds, and screened essentially as described above, except that filters were prehybridized in 25 mL of hybridization buffer consisting of 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 1% SDS, 5% dextran sulfate and 0.1 mg/mL denatured salmon sperm DNA (Sigma Chemical Co.) at 50.degree. C. for 2 h. Radiolabeled probe prepared from pCF3 as described above was added, and allowed to hybridize for 18 h at 50.degree. C. The probes were washed twice at room temperature with 2X SSPE, 1% SDS for five min followed by washing for 5 min at 50.degree. C. in 0.2X SSPE, 1% SDS. Autoradiography of the filters indicated that there was one strongly hybridizing plaque, and approximately five weakly hybridizing plaques. The more strongly hybridizing plaque was subjected to a second round of screening as before, except that the final wash was for 5 min at 60.degree. C. in 0.2X SSPE, 1% SDS. Numerous, strongly hybridizing plaques were observed, and one, well-isolated from other phage, was picked for further analysis. Sequences of the pbluescript vector from the purified phage, including the cDNA insert, were excised in the presence of a helper phage and the resultant phagemid was used to infect E. coli XL-1 Blue cells. DNA from the plasmid, designated pXF1, was made by the alkaline lysis miniprep procedure described in Sambrook et al. (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989) Cold Spring Harbor Laboratory Press). The alkali-denatured double-stranded DNA from pXF1 was completely sequenced on both strands. The insert of pXF1 contained a stretch of 1783 nucleotides which contained an unknown open-reading frame and also contained a poly-A stretch of 16 nucleotides 3' to the open reading frame, from nucleotides 1767 to 1783, followed by an Eco RI restriction site. The 2184 bases that followed this Eco RI site contained a 1145 bp open reading frame which encoded a polypeptide of about 68% identity to, and colinear with, the Arabidopsis delta-15 desaturase polypeptide listed in SEQ ID No:2. The putative start methionine of the 1145 bp open-reading frame corresponded to the start methionine of the Arabidopsis microsomal delta-15 peptide and there were no amino acids corresponding to a plastid transit peptide 5' to this methionine. When the insert in pXF1 was digested with Eco RI four fragments were observed, fragments of approximately 370 bp and 1400 bp fragments, derived from the first 1783 bp of the insert in pxF1, and fragments of approximately 600 bp and 1600 bp derived from the the other 2184 nucleotides of the insert in pXF1. Only the 600 bp and 1600 bp fragments hybridized with probe derived from pCF3 on Southern blots. It was deduced that pXF1 contained two different cDNA inserts separated by an Eco RI site and the second of these inserts was a 2184 bp cDNA encoding a soybean microsomal delta-15 desaturase. The complete nucleotide sequence of the 2184 bp soybean microsomal delta-15 cDNA contained in plasmid pXF1 is listed in SEQ ID No:10. Plasmid pXF1 was deposited on Dec. 3, 1991 with the American Type Culture Collection of Rockville, Md. under the provisions of the Budapest Treaty and bears accession number ATCC 68874. Using Soybean Microsomal Delta-15 Desaturase cDNA as a Hybridization Probe to Isolate cDNAs Encoding Related Desaturases from Soybean A 1.0 kb fragment of DNA corresponding to part of the coding region of the soybean microsomal delta-15 desaturase cDNA contained in plasmid pXF1, was excised with the restriction enzyme Hha I and gel purified. The fragment was labeled with .sup.32 p as described above and used to probe a soybean cDNA library as described above. Autoradiography of the filters indicated that there were eight hybridizing plaques and these were subjected to a second round of screening. Sequences of the pbluescript vector from all eight of the purified phages, including the cDNA inserts, were excised in the presence of a helper phage and the resultant phagemids were used to infect E. coli XL-1 Blue cells. DNA from the plasmids was made by the alkaline lysis miniprep procedure described in Sambrook et al. (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989) Cold Spring Harbor Laboratory Press). Restriction analysis showed they contained inserts ranging from 1.0 kb to 3.0 kb in size. One of these inserts, designated pSFD-118bwp, contained an insert of about 1700 bp. The alkali-denatured double-stranded DNA from pSFD-118bwp was completely sequenced on both strands, shown in SEQ ID NO:12. The insert of pSFD-118bwp contained a stretch of 1675 nucleotides which contained an open-reading frame encoding a polypeptide, shown in SEQ ID NO:13, of about 80% identity with, and colinear with, the Arabidopsis plastid delta-15 desaturase polypeptide listed in SEQ ID No:5. The open-reading frame also encoded amino acids corresponding to a plastid transit peptide at the 5' end of the open-reading frame. The transit peptide was colinear with, and shared some homology to, the transit peptide described for the Arabidopsis plastid delta-15 glycerolipid desaturase. The complete nucleotide sequence of the 1675 bp soybean plastid delta-15 glycerolipid desaturase cDNA is listed in SEQ ID No:12. Comparison of the different delta-15 desaturase sequences disclosed in the application by the method of Needleman et al. (J. Mol. Biol. (1970) 48:443-453) using gap weight and gap length weight values of 3.0 and 0.1, respectively, reveals the relatedness between them as shown in Table 3. TABLE 3 ______________________________________ Percent Identities Between Different Delta-15 Fatty Acid Desaturases at the Amino Acid Level aD c3 cD s3 sD ______________________________________ a3 66 93 66 68 67 aD -- 67 90 67 69 c3 -- -- 68 68 68 cD -- -- -- 68 74 ______________________________________ a3, ad, c3, cD, s3 and sD refer, respectively, to SEQ ID NO:2 (Arabidopsis microsomal delta-15 desaturase), SEQ ID NO:5 (Arabidopsis plastid delta-15 desaturase), SEQ ID NO:7 (canola microsomal delta-15 desaturase), SEQ ID NO:9 (canola plastid delta-15 desaturase), SEQ ID NO:11 (soybean microsomal delta-15 desaturase), and SEQ ID NO:13 (soybean plastid delta-15 desaturase). Based on these comparisons, the delta-15 desaturases, of both microsomal and plastid types, have overall identities of 65% or more at the amino acid levels, even when from different plant species. Isolation of Nucleotide Sequences Encoding Homologous and Heterologous Glycerolipid Desaturases Fragments of the instant invention may be used to isolate cDNAs and genes of homologous and heterologous glycerolipid desaturases from the same species as the fragment of the invention or from different species. Isolation of homologous genes using sequence-dependent protocols is well-known in the art. Southern blot analysis revealed that the Arabidopsis microsomal delta-15 desaturase cDNA (SEQ ID NO:1) hybridized to genomic DNA fragments of corn and soybean. In addition, Applicants have demonstrated that it can be used to isolate cDNAs encoding seed microsomal delta-15 desaturases from Brassica napus (SEQ ID NO:6) and soybean (SEQ ID NO:10). Thus, one can isolate cDNAs and genes for homologous glycerolipid desaturases from the same or different higher plant species, especially from the oil-producing species. More importantly, one can use the fragments of the invention to isolate cDNAs and genes for heterologous glycerolipid desaturases, including those found in plastids. Thus, Arabidopsis microsomal delta-15 desaturase cDNA (SEQ ID NO:1) was successfully used as a hybridization probe to isolate cDNAs encoding the related plastid delta-15 desaturases from Arabidopsis (SEQ ID NO:4) and Brassica napus (SEQ ID NO: 8), and the soybean microsomal delta-15 soybean (SEQ ID NO:10) was successfully used to isolate soybean cDNA encoding plastid delta-15 desaturase (SEQ ID NO:12). In a particular embodiment of the present invention, regions of the nucleic acid fragments of the invention that are conserved between different desaturases may be used by one skilled in the art to design a mixture of degenerate oligomers for use in sequence-dependent protocols aimed at isolating nucleic acid fragments encoding other homologous or heterologous glycerolipid desaturase cDNA's or genes. For example, by comparing all desaturase polypeptides one can identify stretches of amino acids that are conserved between them, and then use the conserved amino acid sequence to design oligomers, both short degenerate or long ones, or "guessmers" as known by one skilled in the art (see Sambrook et al., (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989), Cold Spring Harbor Laboratory Press). Such oligomers and "quessmers" may be used as hybridization-probes as known to one skilled in the art. For example, comparison of cyanobacterial desA and plant delta-15 desaturases revealed a particularly well conserved stretch of amino acids (amino acids 97-108 in SEQ ID NO:1). SEQ ID NOS:20 and 21 represent two sets of 36-mers each 16-fold degenerate made to this region. End-labeled oligomers represented in SEQ ID NOS:20 and 21 were mixed and used as hybridzation probes to screen Arabidopsis cDNA libraries. Most of the positively-hybridizing plaques also hybridized to cDNAs encoding Arabidopsis microsomal and plastid delta-15 desaturases (SEQ ID NOS:1 and 4). However, the use of SEQ ID NOS:20 and 21 did not give consistent and reproducible results. A 135 base-long oligomer (SEQ ID NO:32) was also made as an antisense strand to a longer stretch of the same conserved region, amino acids 97 to 141 in SEQ ID NO:1 (FVLGHDCGHGSFSDIPLLNSVVGHILHSFILVPYHGWRISHRTHH). At positions of ambiguity, the design used either deoxyinosines or most frequently used codons based on the codon usage in Arabidopsis genes. When used as a hybridization probe, the 135-mer hybridized to all plaques that also hybridized to cDNAs encoding Arabidopsis microsomal and plastid delta-15 desaturases (SEQ ID NOS:1 and 4). In addition, it also hybridized to plaques that did not hybridize to SEQ ID NOS:1 and 4). The latter were purified and excised as described previously. Nucleotide sequencing of the cDNA inserts in the resultant plasmids revealed DNA sequences that did not show any relatedness to any desaturase. For another example, in the polymerase chain reaction (Innis, et al., Eds, (1990) PCR Protocols: A Guide to Methods and Applications, Academic Press, San Diego), two short pieces of the present fragment of the invention can be used to amplify a longer glycerolipid desaturase DNA fragment from DNA or RNA. The polymerase chain reaction may also be performed on a library of cloned nucleotide sequences with one primer based on the fragment of the invention and the other on either the poly A.sup.+ tail or a vector sequence. These oligomers may be unique sequences or degenerate sequences derived from the nucleic acid fragments of the invention. The longer piece of homologous glycerolipid desaturase DNA generated by this method could then be used as a probe for isolating related glycerolipid desaturase genes or cDNAs from Arabidopsis or other species. The design of oligomers, including long oligomers using deoxyinosine, and "guessmers" for hybridization or for the polymerase chain reaction are known to one skilled in the art and discussed in Sambrook et al., (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989), Cold Spring Harbor Laboratory Press). Stretches of conserved amino acids between delta-15 desaturase and other desaturases, especially desA, allow for the design of such oligomers. For example, conserved stretches of amino acids between desA and delta-15 desaturase, discussed above, are useful in designing long oligomers for hybridization as well as shorter ones for use as primers in the polymerase chain reaction. In this regard, the conserved amino acid stretch of amino acids 97 to 108 of SEQ ID NO:2 is particularly useful. Other conserved regions in SEQ ID NO:2 useful for this purpose are amino acids 299 to 309, amino acids 115 to 121, and amino acids 133 to 141. Amino acid stretch 133 to 141 in SEQ ID NO:2 shows especially good homology to several desaturases. For example, in this stretch, amino acids 133, 137, 138, 140 and 141 are conserved in plant delta-15 desaturases, cyanobacterial desA, yeast and mammalian microsomal stearoyl-CoA desaturases. Comparison of cyanobacterial des A and plant delta-15 desaturases revealed two-particularly well conserved stretch of amino acids (amino acids 97-108 and amino acids 299-311 in SEQ ID NO:1) that can be used for PCR. The following sets of PCR primers were made to these regions: ______________________________________ AA positions SEQ Fold in ID NO Length Degeneracy SEQ ID NO: 2 AA Sequence ______________________________________ 20 36 16 97-108 (S) FVLGHDCGHGSF 21 36 16 97-108 (S) FVLGHDCGHGSF 28 36 16 97-108 (S) FVLGHDCGHGSF 29 36 16 97-108 (S) FVLGHDCGHGSF 22 18 72 100-105 (S) GHDCGH 23 18 72 100-105 (S) GHDCGH 24 18 72 299-304 (AS) HDIGTH 25 18 72 299-304 (AS) HDIGTH 26 23 416 304-309 (AS) HVIHHL 27 23 416 304-309 (AS) HVIHHL 30 38 64 299-311 (AS) HDIGTHVIHHLFP 31 38 64 299-311 (AS) HDIGTHVIHHLFP ______________________________________ In one experiment, PCRs were performed using SEQ ID NOS:22 and 23 as sense primers and either SEQ ID NOS:24 and 25 or SEQ ID NOS:26 and 27 as antisense primers on poly A.sup.+ RNA purified from both Arabidopsis leaf and canola developing seeds. All PCRs resulted in PCR products of the correct size (ca. 630 bp). The PCR products from Arabidopsis and canola were purified and used as radiolabeled hybridization probes to screen the Lambda Yes Arabidopsis cDNA library, as described above. This led to the isolation of a pure phage, which was excised to give plasmid pYacp7. The cDNA insert in pYacp7 was partially sequenced. It's sequence showed that it encoded an incomplete desaturase polypeptide that was identical to another cDNA (in plasmid pFadx-2) isolated by low-stringency hybridization as described previously. The composite sequence derived from the partial sequences from the cDNA inserts in pFadx-2 and pYacp7 is shown in SEQ ID NO:16 and the polypeptide encoded by it in SEQ ID NO:17. As discussed previously, SEQ ID NO:17 is a putative plastid delta-15 desaturase. This is further supported by Southern blot analysis using radiolabeled cDNA inserts from either pCF3, pACF2-2, or pYacp7 on Arabidopsis genomic DNA digested with one of several enzymes. It shows that the different inserts hybridize to different restriction fragments and that only the inserts from pACF2-2 and pYacp7 show some cross-hybridization. In another PCR experiment, PCR was performed using ca. 80 pmoles each of SEQ ID NOS:28 and 29 as sense primers and ca. 94 pmoles each of SEQ ID NOS:30 and 31 as antisense primers on poly A.sup.+ RNA purified from Arabidopsis mutant line 3707. This was performed using GeneAmp.RTM. RNA PCR Kit (Perkin Elmer Cetus) following manufacturer's protocol and using the following program: a) 1 cycle of 2 min at 95.degree. C., b) 35 cycles of 1 min at 95.degree. C. (denaturation), 1 min at 50.degree. C. (annealing) and 1 min at 65.degree. C. (extension), and c) 1 cycle of 7 min at 65.degree. C. The resulting PCR product, of the correct size (ca. 630 bp), was purified, radiolabeled, and used as a hybridization probe on a Southern blot of Arabidopsis genomic DNA as described above. While it hybridized to restriction fragments that also hybridized to SEQ ID NOS:1 (Arabidopsis microsomal delta-15 desaturase), 4 (Arabidopsis plastid delta-15 desaturase), and 16 (Arabidopsis plastid delta-15 desaturase), it also hybridized to novel fragments that did not hybridze to previously cloned desaturase cDNAs. However, even after several attempts, the radiolabeled PCR product did not hybridize to any novel cDNA clone when used as a probe on different Arabidopsis cDNA libraries: in all cases it hybridzed only to plaques that also hybridized to the known desaturase cDNAs. Furthermore, the PCR product was subcloned into a plasmid vector and after screening about a 100 of these, none gave rise to a clone with a novel desaturase sequence. The isolation of other glycerolipid desaturases will become easier as more examples of glycerolipid desaturases are isolated using the fragments of the invention. Knowing the conserved amino acid sequences from diverse desaturases will also allow one to identify more and better consensus sequences. Such sequences can be used to make hybridization probes or amplification primers which will further aid in the isolation of different glycerolipid desaturases, including those from non-plant sources such as fungi, algae, and even cyanobacteria, as well as other membrane-associated desaturases from other organisms. The function of the diverse nucleotide fragments encoding glycerolipid desaturases that can be isolated using the present invention can be identified by transforming plants with the isolated desaturase sequences, linked in sense or antisense orientation to suitable regulatory sequences required for plant expression, and observing the fatty acid phenotype of the resulting transgenic plants. Preferred target plants for the transformation are the same as the source of the isolated nucleotide fragments when the goal is to obtain inhibition of the corresponding endogenous gene by antisense inhibition or cosuppression. Preferred target plants for use in expression or overexpression of the isolated nucleic acid fragments are plants with known mutations in desaturation reactions, such as the Arabidopsis desaturase mutants, mutant flax deficient in delta-15 desaturation, or mutant sunflower deficient in delta-12 desaturation. Alternatively, the function of the isolated nucleic acid fragments can be determined similarly via transformation of other organisms, such as yeast or cyanobacteria, with chimeric genes containing the nucleic acid fragment and suitable regulatory sequences followed by analysis of fatty acid composition and/or enzyme activity. Overexpression of the Glycerolipid Desaturase Enzymes in Transgenic Species The nucleic acid fragment(s) of the instant invention encoding functional glycerolipid desaturase(s), with suitable regulatory sequences, can be used to overexpress the enzyme(s) in transgenic organisms. Such recombinant DNA constructs may include either the native glycerolipid desaturase gene or a chimeric glycerolipid desaturase gene isolated from the same or a different species as the host organism. For overexpression of glycerolipid desaturase(s), it is preferable that the introduced gene be from a different species to reduce the likelihood of cosuppression. For example, overexpression of delta-15 desaturase in soybean, rapeseed, or other oil-producing species to produce altered levels of polyunsaturated fatty acids may be achieved by expressing RNA from the entire cDNA found in pCF3. Similarly, the isolated nucleic acid fragments encoding glycerolipid desaturases from Arabidopsis, rapeseed, and soybean can also be used by one skilled in the art to obtain substantially homologous full-length cDNAs, if not already obtained, as well as the corresponding genes as fragments of the invention. These, in turn, may be used to overexpress the corresponding desaturases in plants. One skilled in the art can also isolate the coding sequencers) from the fragment(s) of the invention by using and/or creating sites for restriction endonucleases, as described in Sambrook et al., (Molecular Cloning, A Laboratory Manual, 2nd ed. (1989), Cold Spring Harbor Laboratory Press). For example, the fragment in SEQ ID NO:1 in plasmid pCF3 is flanked by Not I sites and can be isolated as a Not I fragment that can be introduced in the sense orientation relative to suitable plant regulatory sequences. Alternatively, sites for Nco I (5'-CCATGG-3') or Sph I (5'-GCATGC-3') that allow precise removal of coding sequences starting with the initiating codon "ATG" may be engineered into the fragment(s) of the invention. For example, for utilizing the coding sequence of delta-15 desaturase from pCF3, an Sph I site can be engineered by substituting nucleotides at positions 44, 45, and 49 of SEQ ID NO:1 with G, C, and C, respectively. Inhibition of Plant Target Genes by Use of Antisense RNA Antisense RNA has been used to inhibit plant target genes in a tissue-specific manner (see van der Krol et al., Biotechniques (1988) 6:958-976). Antisense inhibition has been shown using the entire cDNA sequence (Sheehy et al., Proc. Natl. Acad. Sci. USA (1988) 85:8805-8809) as well as a partial cDNA sequence (Cannon et al., Plant Molec. Biol. (1990) 15:39-47). There is also evidence that the 3' non-coding sequences (Ch'ng et al., Proc. Natl. Acad. Sci. USA (1989) 86:10006-10010) and fragments of 5' coding sequence, containing as few as 41 base-pairs of a 1.87 kb cDNA (Cannon et al., Plant Molec. Biol. (1990) 15:39-47), can play important roles in antisense inhibition. The use of antisense inhibition of the glycerolipid desaturases may require isolation of the transcribed sequence for one or more target glycerolipid desaturase genes that are expressed in the target tissue of the target plant. The genes that are most highly expressed are the best targets for antisense inhibition. These genes may be identified by determining their levels of transcription by techniques, such as quantitative analysis of mRNA levels or nuclear run-off transcription, known to one skilled in the art. For example, antisense inhibition of delta-15 desaturase in Brassica napus resulting in altered levels of polyunsaturated fatty acids may be achieved by expressing antisense RNA from the entire or partial cDNA found in pBNSF3-2. |
| PATENT EXAMPLES | This data is not available for free |
| PATENT PHOTOCOPY | Available on request |
|
Want more information ? Interested in the hidden information ? Click here and do your request. |