WO1999045122A1 - Modification of fatty acid metabolism in plants - Google Patents

Abstract

Methods and systems to modify fatty acid biosynthesis and oxidation in plants to make new polymers are provided. Two enzymes are essential: a hydratase such as D-specific enoyl-CoA hydratase, for example, the hydratase obtained from Aeromonas caviae, and a β-oxidation enzyme system. Some plants have a β-oxidation enzyme system which is sufficient to modify polymer synthesis when the plants are engineered to express the hydratase. Examples demonstrate production of polymer by expression of these enzymes in transgenic plants. Examples also demonstrate that modifications in fatty acid biosynthesis can be used to alter plant phenotypes, decreasing or eliminating seed production and increasing green plant biomass, as well as producing polyhydroxyalkanoates.

Description

MODIFICATION OF FATTY ACID METABOLISM IN PLANTS

Background Of The Invention

The present invention is generally in the field of transgenic plant systems for the production of polyhydroxyalkanoate materials, modification of triglycerides and fatty acids, and methods for altering seed production in plants.

Methods for producing stable transgenic plants for agronomic crops have been developed over the last 15 years. Crops have been genetically modified for improvements in both input and output traits. In the former traits, tolerance to specific agrochemicals has been engineered into crops, and specific natural pesticides, such as the Bacillus thuringenesis toxin, have been expressed directly in the plant. There also has been significant progress in developing male sterility systems for the production of hybrid plants. With respect to output traits, crops are being modified to increase the value of the product, generally the seed, grain, or fiber of the plant. Critical metabolic targets include the modification of starch, fatty acid, and oil biosynthetic pathways.

There is considerable commercial interest in producing microbial polyhydroxyalkanoate (PHA) biopolymers in plant crops. See, for example, U.S. Patent Nos. 5,245,023 and 5,250,430 to Peoples and Sinskey; U.S. Patent No. 5,502,273 to Bright et al.; U.S. Patent No. 5,534,432 to Peoples and Sinskey; U.S. Patent No. 5,602,321 to John; U.S. Patent No. 5,610,041 to Somerville et al; PCT WO 91/00917; PCT WO 92/19747; PCT WO 93/02187; PCT WO 93/02194; PCT WO 94/12014; Poirier et al., Science 256:520-23 (1992); van der Leij & Witholt, Can. J. Microbiol. 41(supplement):222-38 (1995); Nawrath & Poirier, 77ze International Symposium on Bacterial Polyhydroxyalkanoates, (Eggink et al., eds.) Davos Switzerland (August 18-23, 1996); and Williams and Peoples, CHEMTECH 26: 38-44 (1996). PHAs are natural, thermoplastic polyesters and can be processed by traditional polymer techniques for use in an enormous variety of applications, including consumer packaging, disposable diaper linings and garbage bags, food and medical products.

Early studies on the production of polyhydroxybutyrate in the chloroplasts of the experimental plant system Arabidopsis thaliana resulted in the accumulation of up to 14% of the leaf dry weight as PHB (Nawrath et al., 1993). Arabidopsis, however, has no agronomic value. Moreover, in order to economically produce PHAs in agronomic crops, it is desirable to produce the PHAs in the seeds, so that the current infrastructure for harvesting and processing seeds can be utilized. The options for recovery of the PHAs from plant seeds (PCT WO 97/15681 ) and the end use applications (Williams & Peoples, CHEMTECH 26:38-44 (1996)) are significantly affected by the polymer composition. Therefore, it would be advantageous to develop transgenic plant systems that produce PHA polymers having a well-defined composition. Careful selection of the PHA biosynthetic enzymes on the basis of their substrate specificity allows for the production of PHA polymers of defined composition in transgenic systems (U.S. Patent Nos. 5,229,279; 5,245,023; 5,250,430; 5,480,794; 5,512,669; 5,534,432; 5,661,026; and 5,663,063). In bacteria, each PHA group is produced by a specific pathway. In the case of the short pendant group PHAs, three enzymes are involved: β- ketothiolase, acetoacetyl-CoA reductase, and PHA synthase. The homopolymer PHB, for example, is produced by the condensation of two molecules of acetyl-coenzyme A to give acetoacetyl-coenzyme A. The latter then is reduced to the chiral intermediate R-3-hydroxybutyryl-coenzyme A by the reductase, and subsequently polymerized by the PHA synthase enzyme. The PHA synthase notably has a relatively wide substrate specificity which allows it to polymerize C3-C5 hydroxy acid monomers including both 4-hydroxy and 5-hydroxy acid units. This biosynthetic pathway is found in a number of bacteria such as Alcaligenes eutrophus, A. latus, Azotobacter vinlandii, and Zoogloea ramigera. Long pendant group PHAs are produced for example by many different Pseudomonas bacteria. Their biosynthesis involve the β-oxidation of fatty acids and fatty acid synthesis as routes to the hydroxyacyl-coenzyme A monomeric units. The latter then are converted by PHA synthases which have substrate specificities favoring the larger C6-C14 monomeric units (Peoples & Sinskey, 1990). In the case of the PHB-co-HX copolymers which usually are produced from cells grown on fatty acids, a combination of these routes can be responsible for the formation of the different monomeric units. Indeed, analysis of the DNA locus encoding the PHA synthase gene in Aeromonas caviae, which produces the copolymer PHB-co-3-hydroxyhexanoate, was used to identify a gene encoding a D-specific enoyl-CoA hydratase responsible for the production of the D-β-hydroxybutyryl-CoA and D-β - hydroxyhexanoyl-CoA units (Fukui & Doi, J. Bacteriol. 179:4821-30 (1997); Fukui et. al., J. Bacteriol. 180:667-73 (1998)). Other sources of such hydratase genes and enzymes include Alcaligenes, Pseudomonas, and Rhodospir ilium.

The enzymes PHA synthase, acetoacetyl-CoA reductase, and β- ketothiolase, which produce the short pendant group PHAs in A. eutrophus, are coded by an operon comprising the phbC-phbA-phbB genes; Peoples et al, 1987; Peoples & Sinskey, 1989). In the Pseudomonas organisms, the PHA synthases responsible for production of the long pendant group PHAs have been found to be encoded on thepha locus, specifically by the pha A andphaC genes (U.S. Patent Nos. 5,245,023 and 5,250,430; Huisman et. al., J. Biol. Chem. 266:2191-98 (1991)). Since these earlier studies, a range of PHA biosynthetic genes have been isolated and characterized or identified from genome sequencing projects. Examples ofknown PHA biosynthetic genes are disclosed in the following references: Aeronomas caviae (Fukui & Doi, 1997, J. Bacteriol. 11 A.21-30); Alcaligenes eutrophus (U.S. Patent Nos. 5,245,023; 5,250,430; 5,512,669; and 5,661,026; Peoples & Sinskey, J. Biol. Chem. 264:15298-03 (1989)); Acinetobacter (Schembri et. al., FEMS Microbiol. Lett. 118: 145-52 (1994)); Chromatium vinosum (Liebergesell & Steinbuchel, Eur. J. Biochem. 209:135-50 (1992)); Methylobacterium extorquens (Valentin & Steinbuchel, Appl. Microbiol. Biotechnol. 39:309-17 (1993)); Nocardia corallina (GENBANK Accession No. AF019964; Hall et. al., 1998, Can. J. Microbiol. 44:687-69); Paracoccus denitriflcans (Ueda et al., J. Bacteriol. 178:774-79 (1996); Yabutani et. al, FEMS Microbiol. Lett. 133:85-90 (1995)); Pseudomonas acidophila (Umeda et. al., 1998, Applied Biochemistry and Biotechnology, 70-72:341-52): Pseudomonas sp. 61-3 (Matsusaki et al, 1998, J. Bacteriol. 180:6459-67); Nocardia corallina; Pseudomonas aeruginosa (Timm & Steinbuchel, Ewr. J. Biochem. 209: 15-30 (1992)); P. oleovorans (U.S. Patent Nos. 5,245,023 and 5,250,430; Huisman et. al., J. Biol. Chem. 266(4):2191-98 (1991); Rhizobium etli (Cevallos et. al., J. Bacteriol. 178: 1646-54 (1996)); R. meliloti (Tombolini et. al,

Microbiology 141:2553-59 (1995)); Rhodococcus ruber (Pieper-Furst & Steinbuchel, FΕMS Microbiol. Lett. 75:73-79 (1992)); Rhodospirillum rubrum (Hustede et. al., FEMS Microbiol. Lett 93:285-90 (1992)); Rhodobacter sphaeroides (Hustede et. al., FEMS Microbiol. Rev. 9:217-30 (1992); Biotechnol. Lett. 15:709-14 (1993); Synechocystis sp. (DNA Res. 3_: 109-36 (1996)); Thiocapsiae violacea (Appl. Microbiol. Biotechnol. 38:493-501 (1993)) and Zoogloea ramigera (Peoples et. al., J. Biol. Chem. 262:97-102 (1987); Peoples & Sinskey, Molecular Microbiology 3:349-57 (1989)). The availability of these genes or their published DNA sequences should provide a range of options for producing PHAs.

PHA synthases suitable for producing PHB-co-HH copolymers comprising from 1-99% HH monomers are encoded by the Rhodococcus ruber, Rhodospirillum rubrum, Thiocapsiae violacea, and Aeromonas caviae PHA synthase genes. PHA synthases useful for incorporating 3- hydroxyacids of 6-12 carbon atoms in addition to R-3-hydroxybutyrate i.e. for producing biological polymers equivalent to the chemically synthesized copolymers described in PCT WO 95/20614, PCT WO 95/20615, and PCT WO 95/20621 have been identified in a number of Pseudomonas and other bacteria (Steinbuchel & Wiese, Appl. Microbiol. Biotechnol. 37:691-97 (1992); Valentin et al., Appl. Microbiol. Biotechnol. 36:507-14 (1992); Valentin et al., Appl. Microbiol. Biotechnol. 40:710-16 (1994); Lee et al., Appl. Microbiol. Biotechnol. 42:901-09 (1995); Kato et al, Appl. Microbiol. Biotechnol. 45:363-70 (1996); Abe et al., Int. J. Biol. Macromol. 16: 115-19 (1994); Valentin et al., Appl. Microbiol. Biotechnol. 46:261-67 (1996)) and can readily be isolated as described in U.S. Patent Nos. 5,245,023 and 5,250,430. The PHA synthase from P. oleovorans (U.S. Patent Nos. 5,245,023 and 5,250,430; Huisman et. al., J. Biol. Chem. 266(4): 2191-98 (1991)) is suitable for producing the long pendant group PHAs. Plant genes encoding β-ketothiolase also have been identified (Vollack & Bach, Plant Physiol. JO : 1097-107 (1996)).

Despite this ability to modify monomer composition by selection of the syntheses and substrates, it is desirable to modify other features of polymer biosynthesis, such as that which involves fatty acid metabolism.

It is therefore an object of the present invention to provide a method and DNA constructs to introduce fatty acid oxidation enzyme systems for manipulating the cellular metabolism of plants. It is another object of the present invention to provide methods for enhancing the production of PHAs in plants, preferably in the oilseeds thereof.

Summary Of The Invention Methods and systems to modify fatty acid biosynthesis and oxidation in plants to make new polymers are described. Two enzymes are essential: a hydratase such as D-specific enoyl-CoA hydratase, for example, the hydratase obtained ΪXOXΆ Aeromonas caviae, and a β-oxidation enzyme system. Some plants have a β-oxidation enzyme system which is sufficient to modify polymer synthesis when the plants are engineered to express the hydratase.

Examples demonstrate production of polymer by expression of these enzymes in transgenic plants. Examples also demonstrate that modifications in fatty acid biosynthesis can be used to alter plant phenotypes, decreasing or eliminating seed production and increasing green plant biomass, as well as producing PHAs. Brief Description of the Drawings

Figure 1 is a schematic of fatty acid β-oxidation routes to produce polyhydroxyalkanoate monomers.

Figure 2 is a schematic showing plasmid constructs pSBS2024 and pSBS2025.

Figures 3 A and 3B are schematics showing plasmid constructs pCGmfl24 and pCGmfl25.

Figures 4A and 4B are schematics showing plasmid constructs pmfl249 and pmfl254. Figures 5 A and 5B are schematics showing plasmid constructs pCGmf224 and pCGmf225.

Figures 6A and 6B are schematics showing plasmid constructs pCGmflP2S and pCGmf2PlS.

Detailed Description Of The Invention

Methods and DNA constructs for manipulating the cellular metabolism of plants by introducing fatty acid oxidation enzyme systems into the cytoplasm or plastids of developing oilseeds or green tissue are provided. Fatty acid oxidation systems typically comprise several enzyme activities including a β-ketothiolase enzyme activity which utilizes a broad range of β-ketoacyl-CoA substrates.

It surprisingly was found that expression of at least one of these transgenes from the bean phaseolin promoter results in male sterility. Interestingly, these plants did not set seed, but instead produced higher than normal levels of biomass (e.g., leafs, stems, stalks). Therefore the methods and constructs described herein also can be used to create male sterile plants, for example, for hybrid production or to increase the production of biomass of forage, such as alfalfa or tobacco. Plants generated using these methods and DNA constructs are useful for producing polyhydroxyalkanoate biopolymers or for producing novel oil compositions.

The methods described herein include the subsequent incorporation of additional transgenes, in particular encoding additional enzymes involved in fatty acid oxidation or polyhydroxyalkanoate biosynthesis. For polyhydroxyalkanoate biosynthesis, the methods include the incorporation of transgenes encoding enzymes, such as NADH and/or NADPH acetoacetyl- CoenzymeA reductases, PHB synthases, PHA synthases, acetoacetyl-CoA thiolase, hydroxyacyl-CoA epimerases, delta3-cis-delta2-trans enoyl-CoA isomerases, acyl-CoA dehydrogenase, acyl-CoA oxidase and enoyl-CoA hydratases by subsequent transformation of the transgenic plants produced using the methods and DNA constructs described herein or by traditional plant breeding methods. I. Plant Expression Systems

In a preferred embodiment, the fatty acid oxidation transgenes are expressed from a seed specific promoter, and the proteins are expressed in the cytoplasm of the developing oilseed. In an alternate preferred embodiment, fatty acid oxidation transgenes are expressed from a seed specific promoter and the expressed proteins are directed to the plastids using plastid targeting signals. In another preferred embodiment, the fatty acid oxidation transgenes are expressed directly from the plastid chromosome where they have been integrated by homologous recombination. The fatty acid oxidation transgenes may also be expressed throughout the entire plant tissue from a constitutive promoter. It is also useful to be able to control the expression of these transgenes by using promoters that can be activated following the application of an agrochemical or other active ingredient to the crop in the field. Additional control of the expression of these genes encompassed by the methods described herein include the use of recombinase technologies for targeted insertion of the transgenes into specific chromosomal sites in the plant chromosome or to regulate the expression of the transgenes.

The methods described herein involve a plant seed having a genome including (a) a promoter operably linked to a first DNA sequence and a 3'- untranslated region, wherein the first DNA sequence encodes a fatty acid oxidation polypeptide and optionally (b) a promoter operably linked to a second DNA sequence and a 3 '-untranslated region, wherein the second DNA sequence encodes a fatty acid oxidation polypeptide. Expression of the two transgenes provides the plant with a functional fatty acid β-oxidation system having at least β-ketothiolase, dehydrogenase and hydratase activities in the cytoplasm or plastids other than peroxisomes or glyoxisomes. The first and/or second DNA sequence may be isolated from bacteria, yeast, fungi, algae, plants, or animals. It is preferable that at least one of the DNA sequences encodes a polypeptide with at least two, and preferably three, enzyme activities.

Transformation Vectors DNA constructs useful in the methods described herein include transformation vectors capable of introducing transgenes into plants. Several plant transformation vector options are available, including those described in "Gene Transfer to Plants" (Potrykus, et al., eds.) Springer- Verlag Berlin Heidelberg New York (1995); "Transgenic Plants: A Production System for Industrial and Pharmaceutical Proteins" (Owen, et al., eds.) John Wiley & Sons Ltd. England (1996); and "Methods in Plant Molecular Biology: A Laboratory Course Manual" (Maliga, et al. eds.) Cold Spring Laboratory Press, New York (1995). Plant transformation vectors generally include one or more coding sequences of interest under the transcriptional control of 5' and 3' regulatory sequences, including a promoter, a transcription termination and/or polyadenylation signal, and a selectable or screenable marker gene. The usual requirements for 5' regulatory sequences include a promoter, a transcription termination and/or a polyadenylation signal. For the expression of two or more polypeptides from a single transcript, additional RNA processing signals and ribozyme sequences can be engineered into the construct (U.S. Patent No. 5,519,164). This approach has the advantage of locating multiple transgenes in a single locus, which is advantageous in subsequent plant breeding efforts. An additional approach is to use a vector to specifically transform the plant plastid chromosome by homologous recombination (U.S. Patent No. 5,545,818), in which case it is possible to take advantage of the prokaryotic nature of the plastid genome and insert a number of transgenes as an operon. Promoters

A large number of plant promoters are known and result in either constitutive, or environmentally or developmentally regulated expression of the gene of interest. Plant promoters can be selected to control the expression of the transgene in different plant tissues or organelles for all of which methods are known to those skilled in the art (Gasser & Fraley, Science 244: 1293-99 (1989)). The 5' end of the transgene may be engineered to include sequences encoding plastid or other subcellular organelle targeting peptides linked in-frame with the transgene. Suitable constitutive plant promoters include the cauliflower mosaic virus 35S promoter (CaMV) and enhanced CaMN promoters (Odell et. al., Nature, 313: 810 (1985)), actin promoter (McElroy et al., Plant Cell 2:163-71 (1990)), Adhl promoter (Fromm et. al, Bio/Technology 8:833-39 (1990); Kyozuka et al, Mol. Gen. Genet. 228:40-48 (1991)), ubiquitin promoters, the Figwort mosaic virus promoter, mannopine synthase promoter, nopaline synthase promoter and octopine synthase promoter. Useful regulatable promoter systems include spinach nitrate-inducible promoter, heat shock promoters, small subunit of ribulose biphosphate carboxylase promoters and chemically inducible promoters (U.S. Patent No. 5,364,780 to Hershey et al.).

In a preferred embodiment of the methods described herein, the transgenes are expressed only in the developing seeds. Promoters suitable for this purpose include the napin gene promoter (U.S. Patent Nos. 5,420,034 and 5,608,152), the acetyl-CoA carboxylase promoter (U.S. Patent Nos. 5,420,034 and 5,608,152), 2S albumin promoter, seed storage protein promoter, phaseolin promoter (Slightom et. al., Proc. Nail. Acad. Sci. USA 80: 1897-1901 (1983)), oleosin promoter (Plant et. al., Plant Mol. Biol. 25: 193-205 (1994); Rowley et al., Biochim. Biophys. Acta. 1345:1-4 (1997); U.S. Patent No. 5,650,554; and PCT WO 93/20216), zein promoter, glutelin promoter, starch synthase promoter, and starch branching enzyme promoter.

The transformation of suitable agronomic plant hosts using these vectors can be accomplished with a variety of methods and plant tissues. Representative plants useful in the methods disclosed herein include the Brassica family including napus, rappa, sp. carinata andjuncea; maize; soybean; cottonseed; sunflower; palm; coconut; safflower; peanut; mustards including Sinapis alba; and flax. Crops harvested as biomass, such as silage corn, alfalfa, or tobacco, also are useful with the methods disclosed herein. Representative tissues for transformation using these vectors include protoplasts, cells, callus tissue, leaf discs, pollen, and meristems Representative transformation procedures include Agrobacterium-mediated transformation, biolistics, microinjection, electroporation, polyethylene glycol-mediated protoplast transformation, liposome-mediated transformation, and silicon fiber-mediated transformation (U.S. Patent No. 5,464,765; "Gene Transfer to Plants" (Potrykus, et al, eds.) Springer-Verlag Berlin Heidelberg New York (1995); "Transgenic Plants: A Production System for Industrial and Pharmaceutical Proteins" (Owen, et al., eds.) John Wiley & Sons Ltd. England ( 1996); and "Methods in Plant Molecular Biology: A Laboratory Course Manual" (Maliga, et al. eds.) Cold Spring Laboratory Press, New York (1995)). π. Methods for Making and Screening for Transgenic Plants

In order to generate transgenic plants using the constructs described herein, the following procedures can be used to obtain a transformed plant expressing the transgenes subsequent to transformation: select the plant cells that have been transformed on a selective medium; regenerate the plant cells that have been transformed to produce differentiated plants; select transformed plants expressing the transgene at such that the level of desired polypeptide is obtained in the desired tissue and cellular location.

For the specific crops useful for practicing the described methods, transformation procedures have been established, as described for example, in "Gene Transfer to Plants" (Potrykus, et al., eds.) Springer-Nerlag Berlin Heidelberg New York (1995); "Transgenic Plants: A Production System for Industrial and Pharmaceutical Proteins" (Owen, et al., eds.) John Wiley & Sons Ltd. England (1996); and "Methods in Plant Molecular Biology: A Laboratory Course Manual" (Maliga, et al. eds.) Cold Spring Laboratory

10 Press, New York (1995).

Brassica napus can be transformed as described, for example, in U.S.

Patent Nos. 5,188,958 and 5,463,174. Other Brassica such as rappa, carinata and juncea as well as Sinapis alba can be transformed as described by Moloney et. al., Plant Cell Reports 8:238-42 (1989). Soybean can be transformed by a number of reported procedures (U.S. Patent Nos.

5,015,580; 5,015,944; 5,024,944; 5,322,783; 5,416,011; and 5,169,770).

Several transformation procedures have been reported for the production of transgenic maize plants including pollen transformation (U.S. Patent No. 5,629,183), silicon fiber-mediated transformation (U.S. Patent No.

5,464,765), electroporation of protoplasts (U.S. Patent Nos. 5,231,019;

5,472,869; and 5,384,253) gene gun (U.S. Patent Nos. 5,538,877 and

5,538,880 and Agrobacterium-mediated transformation (EP 0 604 662 Al;

PCT WO 94/00977). The Agrobαcterium-mediated procedure is particularly preferred, since single integration events of the transgene constructs are more readily obtained using this procedure, which greatly facilitates subsequent plant breeding. Cotton can be transformed by particle bombardment (U.S.

Patent Nos. 5,004,863 and 5,159,135). Sunflower can be transformed using a combination of particle bombardment and Agrobαcterium infection (EP 0 486 233 A2; U.S. Patent No. 5,030,572). Flax can be transformed by either particle bombardment or Agrobacterium-mediated transformation.

Recombinase technologies include the cre-lox, FLP/FRT, and Gin systems.

Methods for utilizing these technologies are described for example in U.S.

Patent No. 5,527,695 to Hodges et al.; Dale & Ow, Proc. Nαtl. Acαd. Sci. USA 88: 10558-62 (1991); Medberry et. al., Nucleic Acids Res. 23:485-90

(1995).

Selectable Marker Genes

Selectable marker genes useful in practicing the methods described herein include the neomycin phosphotransferase gene nptll (U.S. Patent Nos. 5,034,322 and 5,530, 196), hygromycin resistance gene (U.S. Patent No.

5,668,298), bar gene encoding resistance to phosphinothricin (U.S. Patent

No. 5,276,268). EP 0 530 129 Al describes a positive selection system

11 which enables the transformed plants to outgrow the non-transformed lines by expressing a transgene encoding an enzyme that activates an inactive compound added to the growth media. Screenable marker genes useful in the methods herein include the β-glucuronidase gene (Jefferson et. al., EMBO J. 6:3901-07 (1987); U.S. Patent No. 5,268,463) and native or modified green fluorescent protein gene (Cubitt et. al., Trends Biochem Sci. 20:448-55 (1995); Pang et. al, Plant Physiol. H2:893-900 (1996)). Some of these markers have the added advantage of introducing a trait, such as herbicide resistance, into the plant of interest, thereby providing an additional agronomic value on the input side.

In a preferred embodiment of the methods described herein, more than one gene product is expressed in the plant. This expression can be achieved via a number of different methods, including (1) introducing the encoding DNAs in a single transformation event where all necessary DNAs are on a single vector; (2) introducing the encoding DNAs in a co- transformation event where all necessary DNAs are on separate vectors but introduced into plant cells simultaneously; (3) introducing the encoding DNAs by independent transformation events successively into the plant cells i.e. transformation of transgenic plant cells expressing one or more of the encoding DNAs with additional DNA constructs; and (4) transformation of each of the required DNA constructs by separate transformation events, obtaining transgenic plants expressing the individual proteins and using traditional plant breeding methods to incorporate the entire pathway into a single plant. El. β-Oxidation Enzyme Pathways

Production of PHAs in the cytosol of plants requires the cytosolic localization of enzymes that are able to produce R-3-hydroxyacyl CoA thioesters as substrates for PHA synthases. Both eukaryotes and prokaryotes possess a β-oxidation pathway for fatty acid degradation that consists of a series of enzymes that convert fatty acyl CoA thioesters to acetyl CoA. While these pathways proceed via intermediate 3-hydroxyacyl CoA, the stereochemistry of this intermediate varies among organisms. For example,

12 the β-oxidation pathways of bacteria and the peroxisomal pathway of higher eukaryotes degrade fatty acids to acetyl CoA via S-3-hydroxyacyl CoA (Schultz, "Oxidation of Fatty Acids" in Biochemistry ofLipids, Lipoproteins andMembranes (Nance et al., eds.) pp. 101-06 (Elsevier, Amsterdam 1991)). In Escherichia coli, an epimerase activity encoded by the β-oxidation multifunctional enzyme complex is capable of converting S-3-hydroxyacyl CoA to R-3-hydroxyacyl CoA. Yeast possesses a peroxisomal localized fatty acid degradation pathway that proceeds via intermediate R-3- hydroxyacyl CoA (Hiltunen, et al. J. Biol. Chem. 267: 6646-53 (1992); Filppula, et al. J. Biol. Chem. 270:27453-57 (1995)), such that no epimerase activity is required to produce PHAs.

Plants, like other higher eukaryotes, possesses a β-oxidation pathway for fatty acid degradation localized subcellularly in the peroxisomes (Gerhardt, "Catabolism of Fatty Acids [α and β Oxidation]" in Lipid Metabolism in Plants (Moore, Jr., ed.) pp. 527-65 (CRC Press, Boca Raton, Florida 1993)). Production of PHAs in the cytosol of plants therefore necessitates the cytosolic expression of a β-oxidation pathway, for conversion of fatty acids to R-3-hydroxyacyl CoA thioesters of the correct chain length, as well as cytosolic expression of an appropriate PHA synthase, to polymerize R-3-hydroxyacyl CoA to polymer.

Fatty acids are synthesized as saturated acyl-ACP thioesters in the plastids of plants (Hartwood, "Plant Lipid Metabolism" in Plant Biochemistry (Dey et al., eds.) pp. 237-72 (Academic Press, San Diego 1997)). Prior to export from the plastid into the cytosol, the majority of fatty acids are desaturated via a Δ9 desaturase. The pool of newly synthesized fatty acids in most oilseed crops consists predominantly of oleic acid (cis 9- octadecenoic acid), stearic acid (octadecanoic acid), and palmitic acid (hexadecanoic acid). However, some plants, such as coconut and palm kernel, synthesize shorter chain fatty acids (C8-14). The fatty acid is released from ACP via a thioesterase and subsequently converted to an acyl- CoA thioester via an acyl CoA synthetase located in the plastid membrane (Andrews, et al., "Fatty acid and lipid biosynthesis and degradation" in Plant

13 Physiology, Biochemistry, and Molecular Biology (Dennis et al., eds.) pp. 345-46 (Longman Scientific & Technical, Essex, England 1990); Harwood, "Plant Lipid Metabolism" in Plant Biochemistry (Dey et al., eds) p. 246 (Academic Press, San Diego 1997)). The cytosolic conversion of the pool of newly synthesized acyl CoA thioesters via fatty acid degradation pathways and the conversion of intermediates from these series of reactions to R-3-hydroxyacyl-CoA substrates for PHA synthases can be achieved via the enzyme reactions outlined in Figure 1. The PHA synthase substrates are C4-C16 R-3- hydroxyacyl CoAs. For saturated fatty acyl CoAs, conversion to R-3- hydroxyacyl CoA thioesters using fatty acids degradation pathways necessitates the following sequence of reactions: conversion of the acyl CoA thioester to trαrø-2-enoyl-CoA (reaction 1), hydration of tra«s-2-enoyl-CoA to R-3-hyddroxy acyl CoA (reaction 2a, e.g. yeast system operates through this route and the Aeromonas caviae D-specific hydratase yields C4-C7 R-3- hydroxyacyl-CoAs), hydration of tr «s^,-2-enoyl-Co A to S-3-hydroxy acyl CoA (reaction 2b), and epimerization of S -3 -hydroxyacyl CoA to R-3- hydroxyacyl CoA (reaction 5, e.g. cucumber tetrafunctional protein, bacterial systems). If 3 -hydroxyacyl CoA is not polymerized by PHA synthase forming PHA, it can proceed through the remainder of the β-oxidation pathway as follows: oxidation of 3 -hydroxyacyl CoA to form β-keto acyl CoA (reaction 3) followed by thio lysis in the presence of CoA to yield acetyl CoA and a saturated acyl CoA thioester shorter by two carbon units (reaction 4). The acyl CoA thioester produced in reaction 4 is free to re-enter the β- oxidation pathway at reaction 1 and the acetyl-CoA produced can be converted to R-3 -hydroxyacyl CoA by the action of β-ketothiolase (reaction 7) and NADH or NADPH acetoacetyl-CoA reductase (reaction 6). This latter route is useful for producing R-3-hydroxybutyryl-CoA, R-3- hydroxyvaleryl-CoA and R-3-hydroxyhexanoyl-CoA. The R-3- hydroxyacids of four to sixteen carbon atoms produced by this series of enzymatic reactions can be polymerized by PHA synthases expressed from a transgene, or transgenes in the case of the two subunit synthase enzymes,

14 into PHA polymers.

For Δ9 unsaturated fatty acyl CoAs, a variation of the reaction sequences described is required. Three cycles of β-oxidation, as detailed in Figure 1, will remove six carbon units yielding an unsaturated acyl CoA thioester with a cis double bond at position 3. Conversion of the cis double bond at position 3 to a trans double bond at position 2, catalyzed by A³-cis- A²-trans-enoy\ CoA isomerase will allow the β-oxidation reaction sequences outlined in Figure 1 to proceed. This enzyme activity is present on the microbial β-oxidation complexes and the plant tetrafunctional protein, but not on the y east fox 1.

Acyl CoA thioesters also can be degraded to a β-keto acyl CoA and converted to R-3 -hydroxyacyl CoA via a NADH or NADPH dependent reductase (reaction 6).

Multifunctional enzymes that encode S-specific hydratase, S-specific dehydrogenase, β-ketothiolase, epimerase and Δ³ -cis-A² -frans-enoyl CoA isomerase activities have been found in bacteria such as Escherichia coli (Spratt, et al., J. Bacteriol. 158:535-42 (1984)) and Pseudomonas fragi (Immure, et al., J. Biochem. 107: 184-89 (1990)). The multifunctional enzyme complexes consist of two copies of each of two subunits such that catalytically active protein forms a heterotetramer. The hydratase, dehydrogenase, epimerase, and A³-cis-A²-trans-enoy\ CoA isomerase activities are located on one subunit, whereas the thiolase is located on another subunit. The genes encoding the enzymes from organisms such as E. coli (Spratt, et al., J. Bacteriol. 158:535-42 (1984); DiRusso, J. Bacteriol. 172:6459-68 (1990)) and P. fragi (Sato, et al., J. Biochem. UΛ :8-15 (1992)) have been isolated and sequenced and are suitable for practicing the methods described herein. Furthermore, the E. coli enzyme system has been subjected to site-directed mutagenesis analysis to identify amino acid residues critical to the individual enzyme activities (He & Yang, Biochemistry 35:9625-30 (1996); Yang et. al., Biochemistry 34:6641-47 (1995); He & Yang, Biochemistry 36:11044-49 (1997); He et. al., Biochemistry 36:261-68 (1997); Yang & Εlzinga, J. Biol. Chem. 268:6588-

15 92 (1993)). These mutant genes also could be used in some embodiments of the methods described herein.

Mammals, such as rat, possess a trifunctional β-oxidation enzyme in their peroxisomes that contains hydratase, dehydrogenase, and A³-cis-A²- trans-en yl CoA isomerase activities. The trifunctional enzyme from rat liver has been isolated and has been found to be monomeric with a molecular weight of 78 kDa (Palosaari, et al., J. Biol. Chem. 265:2446-49 (1990)). Unlike the bacterial system, thiolase activity is not part of the multienzyme protein (Schultz, "Oxidation of Fatty Acids" in Biochemistry ofLipids, Lipoproteins and Membranes (Vance et al., eds) p. 95 (Elsevier, Amsterdam (1991)). Epimerization in rat occurs by the combined activities of two distinct hydratases, one which converts R-3 -hydroxyacyl CoA to trans-2- enoyl CoA, and another which converts tr «5-2-enoyl CoA to S-3- hydroxyacyl CoA (Smeland, et al., Biochemical and Biophysical Research Communications 160:988-92 (1989)). Mammals also possess β-oxidation pathways in their mitochondria that degrade fatty acids to acetyl CoA via intermediate S-3-hydroxyacyl CoA (Schultz, "Oxidation of Fatty Acids" in Biochemistry ofLipids, Lipoproteins and Membranes (Vance et al., eds) p. 96 (Elsevier, Amsterdam (1991)). Genes encoding mitochondrial β- oxidation activities have been isolated from several animals including a Rat mitochondrial long chain acyl CoA hydratase/3-hydroxy acyl CoA dehydrogenase (GENBANK Accession # D 16478) and a Rat mitochondrial thiolase (GENBANK Accession #s D13921, D00511).

Yeast possesses a multifunctional enzyme, Fox2, that differs from the β-oxidation complexes of bacteria and higher eukaryotes in that it proceeds via a R-3 -hydroxyacyl CoA intermediate instead of S -3 -hydroxyacyl CoA (Hiltunen, et al, J. Biol. Chem. 267:6646-53 (1992)). Fox2 possesses R- specific hydratase and R-specific dehydrogenase enzyme activities. This enzyme does not possess the Δ^-cz^'s-Δ^trflra-enoyl CoA isomerase activity needed for degradation of Δ9-c s-hydroxyacyl CoAs to form R-3- hydroxyacyl CoAs. The gene encoding/ x2 from yeast has been isolated and sequenced and encodes a 900 amino acid protein. The DNA sequence of

16 the structural gene and ammo acid sequence of the encoded polypeptide is shown in SEQ ID NO 1 and SEQ ID NO 2

Plants have a tetrafunctional protein similar to the yeast Fox2, but also encoding a Δ -cιs-A²-trans-enoyl CoA isomerase activity (Muller et , al , J Biol Chem 269 20475-81 (1994)) The DNA sequence of the cDNA and ammo acid sequence of the encoded polypeptide is shown in SEQ ID NO 3 and SEQ ID NO 4 IV. Targeting of Enzymes to the Cytoplasm of Oil Seed Crops

Engineering PHA production in the cytoplasm of plants requires directing the expression of β-oxidation to the cytosol of the plant No targeting signals are present in the bacterial systems, such as faoAB In fungi, yeast, plants, and mammals, β-oxidation occurs in subcellular organelles Typically, the genes are expressed from the nuclear chromosome, and the polypeptides synthesized in the cytoplasm are directed to these organelles by the presence of specific ammo acid sequences To practice the methods described herein using genes isolated from eukaryotic sources, e g , fatty acid oxidation enzymes from eukaryotic sources, such as yeast, fungi, plants, and mammals, the removal or modification of subcellular targeting signals is required to direct the enzymes to the cytosol It may be useful to add signals for directing proteins to the endoplasmic reticulum Peptides useful in this process are well known in the art The general approach is to modify the transgene by inserting a DNA sequence specifying an ER targeting peptide sequence to form a chimeπc gene

Eukaryotic acyl CoA dehydrogenase s, as well as other mitrochondπal proteins, are targeted to the mitochondria via leader peptides on the N- terminus of the protein that are usually 20-60 ammo acids long (Horwich, Current Opinion in Cell Biology, 2 625-33 (1990)) Despite the lack of an obvious consensus sequence for mitochondrial import leader peptides, mutagenesis of key residues in the leader sequence have been demonstrated to prevent the import of the mitochrondπal protein For example, the import of Saccharomyces cerevisiae Fl-ATPase was prevented by mutagenesis of its leader sequence, resulting in the accumulation of the modified precursor

17 protein in the cytoplasm (Bedwell, et al., Mol. Cell Biol. 9:1014-25 (1989)) Three eukaryotic peroxisomal targeting signals have been reported (Gould, et al., J. Cell Biol. 108: 1657-64 (1989); Brickner, et al, J. Plant Physiol, 111: 1213-21 (1997)). The tripeptide targeting signal S/A/C- K H/R-L occurs at the C-terminal end of many peroxisomal proteins (Gould, et al., J. Cell Biol. 108: 1657-64 (1989)). Mutagenesis of this sequence has been shown to prevent import of proteins into peroxisomes. Some peroxisomal proteins do not contain the tripeptide at the C-terminal end of the protein. For these proteins, it has been suggested that targeting occurs via the tripeptide in an internal position within the protein sequence (Gould, et al., J. Cell Biol. 108:1657-64 (1989)) or via an unknown, unrelated sequence (Brickner, et al, J. Plant Physiol. Hi: 1213-21 (1997)). The results of in vitro peroxisomal targeting experiments with fragments of acyl CoA oxidase from Candida tropicalis appear to support the latter theory and suggest that there are two separate targeting signals within the internal amino acid sequence of the polypeptide (Small, et al., The EJMBO Journal 7: 1167- 73 (1988)). In the aforementioned study, the targeting signals were localized to two regions of 118 amino acids in length, and neither of regions was found to contain the targeting signal S/A/C-K/H/R-L. A small number of peroxisomal proteins appear to contain an amino terminal leader sequence for import into peroxisomes (Brickner, et al., J. Plant Physiol. 113: 1213-21 (1997)). These targeting signals can be deleted or altered by site directed mutagenesis. V. Cultivation and Harvesting of Transgenic Plant The transgenic plants can be grown using standard cultivation techniques. The plant or plant part also can be harvested using standard equipment and methods. The PHAs can be recovered from the plant or plant part using known techniques such as solvent extraction in conjunction with traditional seed processing technologies, as described in PCT WO 97/15681, or can be used directly, for example, as animal feed, where it is unnecessary to extract the PHA from the plant biomass.

Several lines which did not produce seed, produced much higher

18 levels of biomass. produced much higher levels of biomass. This phenotype therefore may be useful as a means to increase the amount of green biomass produced per acre for silage, forage, or other biomass crops. End uses include the more cost effective production of forage crops for animal feed or as energy crops for electric power generation. Other uses include increasing biomass levels in crops, such as alfalfa or tobacco, for subsequent recovery of industrial products, such as PHAs by extraction.

The compositions and methods of preparation and use thereof described herein are further described by the following non-limiting examples.

Example 1: Isolation and Characterization of the Pseudomonas putidafaoAB Genes and Fao Enzyme

All DNA manipulations, including PCR, DNA sequencing E. coli transformation, and plasmids purification, were performed using standard procedures, as described, for example, by Sambrook et. al., Molecular Cloning: A Laboratory Manual (Cold Spring Harbor Laboratory Press, New York (1989)). The genes encoding faoAB from Pseudomonas putida were isolated using a probe generated from P. putida genomic DNA by PCR (polymerase chain reaction) using primers 1 and 2 possessing homology to faoB from Pseudomonas fragi (Sato, et al., J. Biochem. H 8-15 (1992)). Primer 1 :

5' gat ggg ccg etc caa ggg tgg 3' (SEQ ID NO:5)

Primer 2: 5' caa ccc gaa ggt gcc gcc att 3' (SEQ ID NO:6)

A 1.1 kb DNA fragment was purified from the PCR reaction and used as a probe to screen a P. putida genomic library constructed in plasmid pBKCMV using the lambda ZAP expression system (Stratagene). Plasmid pMFXl was selected from the positive clones and the DNA sequence of the insert containing the/αoAB genes and flanking sequences determined. This is shown in SEQ ID NO:7. A fragment containing faoAB was subcloned with the native P. putida ribosome binding site intact into the expression

19 vector pTRCN forming plasmid pMFX3 as follows. Plasmid pMFXl was digested with R rG I. The resulting protruding ends were filled in with Klenow. Digestion withHwd III yielded a 3.39 kb blunt ended! Hind III fragment encoding FaoAB. The expression vector pTRCN was digested with Sma llHind Til and ligated with the faoAB fragment forming the 7.57 kb plasmid pMFX3.

Enzymes in the FaoAB multienzyme complex were assayed as follows. Hydratase activity was assayed by monitoring the conversion of NAD to NADH using the coupling enzyme L-β-hydroxyacyl CoA dehydrogenase as previously described, except that assays were run in the presence of CoA (Filppula, et al., J. Biol. Chem. 270:27453-57 (1995)). Severe product inhibitation of the coupling enzyme was observed in the absence of CoA. The assay contained (1 mL final volume) 60 μM crotonyl CoA, 50 μM Tris-CI, pH 9, 50 μg bovine serum albumin per mL, 50 mM KCl, 1 mM NAD, 7 μg L-specific β-hydroxyacyl CoA dehydrogenase from porcine heart per mL, and 0.25 mM CoA. The assay was initiated with the addition of FaoAB to the assay mixture. A control assay was performed without substrate to determine the rate of consumption of NAD in the absence of the hydratase generated product, S-hydroxybutyryl CoA. One unit of activity is defined as the consumption of one μMol of NAD per min (ε₃₄₀ = 6220 M^"1cm^"1).

Hydroxyacyl CoA dehydrogenase was assayed in the reverse direction with acetoacetyl CoA as the substrate by monitoring the conversion of NADH to NAD at 340 nm (Binstock, et al, Methods in Enzymology, 71:403 (1981)). The assay contained (1 mL final volume) 0.1 M KH₂PO₄, pH 7, 0.2 mg bovine serum albumin per mL, 0.1 mM NADH, and 33 μM acetoacetyl CoA. The assay was initiated with the addition of FaoAB to the assay mixture. When necessary, enzyme samples were diluted in 0.1 M KH₂PO₄, pH 7, containing 1 mg bovine serum albumin per mL. A control assay was performed without substrate acetoacetyl CoA to detect the rate of consumption of NADH in the crude due to enzymes other than hydroxyacyl CoA dehydrogenase. One unit of activity is defined as the consumption of

20 one μMol of NADH per minute (ε_{3 0}= 6220 M^cm^"1).

HydroxyacylCoA dehydrogenase was assayed in the forward direction with crotonyl CoA as a substrate by monitoring the conversion of NAD to NADH at 340 nm (Binstock, et al., Methods in Enzymology, 71:403 (1981)). The assay mixture contained (1 mL final volume) 0.1 M KH₂PO₄, pH 8, 0.3 mg bovine serum albumin per mL, 2 mM β-mercaptoethanol, 0.25 mM CoA, 30 μM crotonyl CoA, and an aliquot of FaoAB. The reaction was preincubated for a couple of minutes to allow in situ formation of S- hydroxybutyryl CoA. The assay then was initiated by the addition of NAD (0.45 mM). A control assay was performed without substrate to detect the rate of consumption of NAD due to enzymes other than hydroxyacyl CoA dehydrogenase. One unit of activity is defined as the consumption of one μMol of NAD per minute (ε₃₄₀ = 6220 M^cm^"1).

Thiolase activity was determined by monitoring the decrease in absorption at 304 nm due to consumption of substrate acetoacetyl CoA as previously described with some modifications (Palmer, et al., J. Biol. Chem. 266: 1-7 (1991)). The assay contained (final volume 1 mL) 62.4 mM Tris-Cl, pH 8.1, 4.8 mM MgCl₂, 62.5 μM CoA, and 62.5 μM acetoacetyl CoA. The assay was initiated with the addition of FaoAB to the assay mixture. A control sample without enzyme was performed for each assay to detect the rate of substrate degradation of pH 8.1 in the absence of enzyme. One unit of activity is defined as the consumption of one μMol of substrate acetoacetyl CoA per minute (ε₃₄o= 16900 M^cm^"1).

Epimerase activity was assayed as previously described (Binstock, et al., Methods in Enzymology, 71:403 (1981)) except that R-3 -hydroxyacyl CoA thioesters were utilized instead of D,L-3 -hydroxyacyl CoA mixtures. The assay contained (final volume 1 mL) 30 μM R-3 -hydroxyacyl CoA, 150 mM KH₂PO₄ (pH 8), 0.3 mg/mL BSA, 0.5 mM NAD, 0.1 mM CoA, and 7 μg/mL L-specific β-hydroxyacyl CoA dehydrogenase from porcine heart. The assay was initiated with the addition of FaoAB.

For expression of FaoAB in DH5α/pMFX3, cultures were grown in 2xTY medium at 30 °C. 2xTY medium contains (per L) 16 g tryptone, 10 g

21 yeast, and 5 g NaCl. A starter culture was grown overnight and used to inoculate (1% inoculum) fresh medium (100 mL in a 250 mL Erienmeyer flask for small scale growths; 1.5L in a 2.8L flask for large scale growths). Cells were induced with 0.4 mM IPTG when the absorbance at 600 nm was in the range of 0.4 to 0.6. Cells were cultured an additional 4 h prior to harvest. Cells were lysed by sonication, and the insoluble matter was removed from the soluble proteins by centrifugation. Acyl CoA dehydrogenase activity was monitored in the reverse direction to ensure activity of the Fao A subunit (SEQ ID NO: 31) and thiolase activity was assayed to determine activity of the Fao subunit. FaoAB in DH5α/pMFX3 contained dehydrogenase and thiolase activity values of 4.3 and 0.99 U/mg, respectively, which is significantly more than the 0.0074 and 0.0033 U/mg observed for dehydrogenase and thiolase, respectively, in control strain DH5α/pTRCN. FaoAB was purified from DH5α/pMFX3 using a modified procedure previously described for the purification of FaoAB from Pseudomonas fragi (Imamura, et al., J. Biochem. 107:184-89 (1990)). Thiolase activity (assayed in the forward direction) and dehydrogenase activities (assayed in the reverse direction) were monitored throughout the purification. Three liters of DH5α/pMFX3 cells (2 X 1.5 L aliquots in 2.8 L Erienmeyer flasks) were grown in 2 x TY medium using the cell growth procedure previously described for preparing cells for enzyme activity analysis. Cells (15.8 g) were resuspended in 32 mL of 10 mM KH PO , pH 7, and lysed by sonication. Soluble proteins were removed from insoluble cells debris by centrifugation (18,000 RPM, 30 min., 4 °C). The soluble extract was made 50% in acetone and the precipitated protein was isolated by centrifugation and redissolved in 10 mM KH₂PO₄, pH 7. The sample was adjusted to 33% saturation with (NHt) SO₄ and the soluble and insoluble proteins were separated by centrifugation. The resulting supernatant was adjusted to 56% saturation with (NH₄)₂SO₄ and the insoluble pellet was isolated by centrifugation and dissolved in 10 mM KH₂PO₄, pH 7. The sample was heated at 50°C for 30 min. and the soluble proteins were isolated by

22 centrifugation and dialyzed in a 6,000 to 8,000 molecular weight cut off membrane in 10 mM KH₂PO₄, pH 7 (2 X 3L, 20 h) The sample was loaded on a Toyo Jozo DEAE FPLC column (3 cm x 14 cm) that previously had been equilibrated in 10 mM KH₂PO₄, pH 7 The protein was eluted with a linear gradient (100 mL by 100 mL, 0 to 500 mM NaCl in 10 KH₂PO₄, pH 7) at a flow of 3 mL/min FaoAB eluted between 300 and 325 mM NaCl The sample was dialyzed in a 50,000 molecular weight cut off membrane m 10 mM KH₂PO₄, pH 7 (1 X 2L, 15h) prior to loading on a macro-prep hydroxylapatite 18/30 (Biorad) FPLC column (2 cm x 15 cm) that previously had been equilibrated in 10 mM KH PO , pH 7 The protein was eluted with a linear gradient (250 mL by 250 mL, 10 to 500 mM KH₂PO₄, pH 7) at a flow rate of 3 mL/min FaoAB eluted between 70 and 130 mM KH PO₄ The fractions containing activity were concentrated to 9 mL using a MILLIPORE™ 100,000 molecular weight cutoff concentrator The buffer was exchanged 3 times with 10 mM KH₂PO , pH 7 containing 20% sucrose and frozen at -70°C Enzyme activities of the hydroxylapatite purified fraction were assayed with a range of substrates The results are shown in Table 1 below Table 1: Enzyme Substrates and Activities

Enzyme Substrate Activity (U/mg) hydratase crotonyl CoA 8 8 dehydrogenase (forward) crotonyl CoA 0 46 dehydrogenase (reverse) acetoacetyl CoA 29 thiolase acetoacetyl CoA 9 9 epimerase R-3-hydroxyoctanyl CoA 0 022 epimerase R-3- hydroxyhexanyl CoA 0 0029

epimerase R-3- hydroxybutyryl CoA 0 000022

Example 2: Production of Antibodies to the FaoAB and FaoAB Polypeptides

Following purification of the FaoAB protein as described in Example 1, a sample was separated by SDS-PAGE The protein band corresponding to the FaoA (SEQ ID NO 31) and FaoB (SEQ ID NO 26) was excised and used to immunize New Zealand white rabbits with complete Freunds adjuvant Boosts were performed using incomplete Freunds at three week

23 intervals Antibodies were recovered from serum by affinity chromatography on Protein A columns (Pharmacia) and tested against the antigen by Western blotting procedures Control extracts of Brassica seeds were used to test for cross reactivity to plant proteins No cross reactivity was detected

Example 3: Construction of Plasmids for Expression of the

Pseudomonas puήdo fao AB Genes in Transgenic Oilseeds

Construction of pSBS2024 Oligonucleotide primers GVR471

5 '-CGGTA CCATTGTACTCCCAGTATC AT-3 ' (SEQ ID NO 8) and GVR472 y-CA TTTAAA 7AGTAGAGTATTGAATATG-3 ' (SEQ ID NO 9) homologous to sequences flanking the 5' and 3' ends (underlined), respectively, of the bean phaseolin promoter (SEQ ID NO 10, Slightom et al , 1983) were designed with the addition oϊKpnl (in italics, nucleotides 1-7 in SEQ ID NO 8) and Swal (in italics, nucleotides 1-9 in SEQ ID NO 9) at the 5' ends of GVR471 and GVR472, respectively These restriction sites were incorporated to facilitate cloning The primers were used to amplify a 1 4 kb phaseolin promoter, which was cloned at the Sm l site in pUC19 by blunt ended ligation The designated plasmid, pCPPI (see Figure 2) was cut with Sail and Svrøl and ligated to a SaWSwal phaseolin terminator (SEQ ID NO 27) The bean phaseolin terminator sequence encompassing the polyadenylation signals was amplified using the following PCR primers GVR396

5'-GATTTAAATGCAAGCTTAAATAAGTATGAACTAAAATGC-3'

(SEQ ID NO 22) and GVR397

5 '-CGGTACCTTAGTTGGTAGGGTGCTA-3 ' (SEQ ID NO 23) and the 1 2Kb fragment (SEQ ID NO 27) cloned into Sail -Sal site of pCCPl to obtain pSBS2024 (Figure 2) The resulting plasmid which contains a unique Hindlll site for cloning was called pSBS2024 (Figure 2)

24 Construction of pSBS2025

A soybean oleosin promoter fragment (SEQ ID NO:l 1; Rowley et al, 1997) was simplified with primers that flank the DNA sequence.

Primer JA408 5 ' -rcr^G^TACATCCATTTCTTAATATAATCCTCTTATTC-3 '

(SEQ ID NO: 12) contains sequences that are complementary to the 5' end (underlined).

Primer npl

5' -CA TTTAAA 7GGTT AAGGTGAAGGT AGGGCT-3 ' (SEQ ID NO: 13) contains sequences homologous to the 3' end (underlined) of the promoter fragment. The restriction sites Xbal (in italics) and Swal (in italics) were incorporated at the 5' end of JA408 and npl, respectively, to facilitate cloning. The primers were used to amplify a 975 bp promoter fragment, which then was cloned into Small site of pUC 19 (see Figure 2). The resulting plasmid, pCSPI, was cut with Sαtl and Swal and ligated to the soybean terminator (SEQ ID NO:28). The soybean oleosin terminator was amplified by PCR using the following primers:

JA410: 5 '-AAGCTTACGTGATGAGTATTAATGTGTTGTTATG-3 ' (SEQ ID NO:29) and

JA411 :

5 '-TCTAGACAATTCATCAAATACAAATCACATTGCC-3 ' (SEQ ID NO:30) and the 225 bp fragment cloned into the Sall-Swal site of pCSPI to obtain plasmid pSBS2025 (Figure 6). The designated plasmid, pSBS2025, carried a unique Hindlϊl site for cloning (Figure 2).

25 Construction of Promoter-coding Sequence Fusions Two oligonucleotide primers were synthesized: np2

5 JL GCT7AAAATGATTTACGAAGGTAAAGCC-3 ' (SEQ ID NO: 14) homologous to nucleotides 553 to 573 of the 5' flanking sequences, and np3

5 ' A 77UC77TCAGTTGAAGCGCTG-3 ' (SEQ ID NO: 15) complementary to nucleotides 2700 to 2683 flanking the 3' end of mfl (faoA, SEQ ID NO:24) of plasmid pmfx3. A Hindlll (in italics) site was introduced at the 5' end of primers np2 and np3 to facilitate cloning. In addition, a 3 bp AAA sequence (bold) was incorporated to obtain a more favorable sequence surrounding the plant translational initiation codon. Primers np2 and np3 were used to amplify the fragment and cloned into Sm l site of pUC 19. The resulting plasmid was called pCmfl (Figures 3A and 3B). Plasmid pBmf2 was constructed in a similar process (Figures 5A and 5B). In order to generate a Hindlll (in italics) at 5' and 3' ends of the mf2 (faoB) gene (SEQ ID NO:25) for cloning, a second set of synthetic primers were designed. Primers np4

5' -AA GCTTAAA ATGAGCCTGAATCC AAGAGAC-3 ' (SEQ ID NO: 16) complementary to 5' (nucleotides 2132-2152 bp) and np5 5 J1 GC77TCAGACGCGTTCGAAGACAGTG -3 ' (SEQ ID NO: 17) homologous to 3' (nucleotides 3907-3886 bp) sequences of mf2 (faoB, SEQ ID NO:25) of plasmid pmfx3 were used in a PCR reaction to amplify the 1.17 kb DNA fragment. The resulting PCR product was cloned into the EcoRV site of pBluescript. The plasmid was referred to as pBmf2. Both plasmids were individually cut with Hindlll and their inserts cloned in plasmids pSBS2024 and pSBS2025, which had previously been linearized with the same restriction enzyme. As a result, the following

26 plasmids were generated: pmfl24 and pmfl25 (Figures 3A and 3B) and pmf224 and pmf225 (Figures 5 A and 5B) containing the Fao genes (mfl and mf2) fused to either the phaseolin or soybean promoters. DNA sequence analysis confirmed the correct promoter-coding sequence-termination sequence fusions for pmfl24, pmfl25, pmf224, and pmf225.

Example 4: Assembly of Promoter-coding Sequence

Fusions into Plant Transformation Vectors

After obtaining plasmids pmfl24, pmfl25, pmf224, and pmf225, promoter-coding sequence fusions were independently cloned into the binary vectors, pCGN1559 (McBride and Summerfelt, 1990) containing the CaMV 35S promoter driving the expression of NPTII gene (conferring resistance to the antibiotic kanamycin) and pSBS2004 containing a parsley ubiquitin promoter driving the PPT gene, which confers resistance to the herbicide phosphinothricine. Binary vectors suitable for this purpose with a variety of selectable markers can be obtained from several sources.

The phaseolin-mf21 fusion cassette was released from the parent plasmid with Xbal and ligated with pCGN1559, which had been linearized with the same restriction enzyme. The resulting plasmid was designated pCGmfl24 (Figures 3 A and 3B). Plasmid pCGmfl25 containing the soybean-mfl fusion was constructed in a similar way (Figures 3 A and 3B), except that both pmfl25 and pCGNl 559 were cut with BamHI before ligation.

Construction of pmf!249 an pmf!254 The plasmid pSBS2004 was linearized with BamHI fragment containing the soybean-mfl fusion. This plasmid was designated pmfl254 (Figures 4 A and 4B). Similarly, the Xbal phaseolin-mfl fusion fragment was ligated to pSBS2004 which had been linearized with the same restriction enzyme. The resulting plasmid was designated pmfl249 (Figures 4A and 4B).

Construction of pCGmf224 and pCGmf225

The phaseolin-mf2 and soybean-mf2 fusions were constructed by

27 excising the fusions from the vector by cutting with either BamHI or Xbal, and cloned into pCGN1559 which had been linearized with either restriction enzyme (Figures 5A and 5B). Construction of pCGmflP2S and pCGmf2PlS The two expression cassettes containing the promoter-coding sequence fusions were assembled on the same binary vector as follows: Plasmid pmfl24 containing the phaseolin-mfl fusion was cut with BamHI and cloned into the BamHI site of pCGN1559 to create pCGmfB124. This plasmid then was linearized with Xbal and ligated to the Xbal fragment of pmf225 containing the soybean-m£2 fusion. The final plasmid was designated pCGmflP2S (Figures 6A and 6B). Plasmid pCGmf2PlS was assembled in similar manner. The phaseolin-mf2 fusion was released from pmf!224 by cutting with BamHI and cloned at the BamHI site of pCGN1559. The resulting plasmid, pCGmfB224, was linearized with Xbal and ligated to the Xbal fragment of pmfl25 containing the soybean-mfl fusion (Figures 6 A and 6B).

Example 5: Transformation of Brassica

Brassica seeds were surface sterilized in 10% commercial bleach (Javex, Colgate-Palmolive) for 30 min. with gentle shaking. The seeds were washed three times in sterile distilled water. Seeds were placed in germination medium comprising Murashige-Skoog (MS) salts and vitamins, 3% (w/v) sucrose and 0.7% (w/v) phytagar, pH 5.8 at a density of 20 per plate and maintained at 24 °C and a 16 h light / 8 h dark photoperiod at a light intensity of 60-80 //Em^'V¹ for four to five days.

Each of the constructs, pCGmfl24, pCGmfl25, pCGmf224, pCGmflP2S, and pCGmf2PlS were introduced into Agrobacterium tumefacians strain EHAIOI (Hood et al., J. Bacteriol. 168:1291-1301 (1986)) by electroporation. Prior to transformation of cotyledonary petioles, single colonies of strain EHAIOI harboring each construct were grown in 5 ml of minimal medium supplemented with 100 mg kanamycin per liter and 100 mg gentamycin per liter for 48 hr at 28 °C. One milliliter of bacterial

28 suspension was pelletized by centrifugation for 1 min in a microfuge. The pellet was resuspended in 1 ml minimal medium.

For transformation, cotyledons were excised from 4 day old, or in some cases 5 day old, seedlings, so that they included approximately 2 mm of petiole at the base. Individual cotyledons with the cut surface of their petioles were immersed in diluted bacterial suspension for 1 s and immediately embedded to a depth of approximately 2 mm in co-cultivation medium, MS medium with 3% (w/v) sucrose and 0.7% phytagar and enriched with 20 μM benzyladenine. The inoculated cotyledons were plated at a density of 10 per plate and incubated under the same growth conditions for 48 h. After co-cultivation, the cotyledons then were transferred to regeneration medium comprising MS medium supplemented with 3% sucrose, 20 μM benzyladenine, 0.7% (w/v) phytagar, pH 5.8, 300 mg timentinin per liter, and 20 mg kanamycin sulfate per liter. After two to three weeks, regenerant shoots obtained were cut and maintained on "shoot elongation" medium (MS medium containing, 3% sucrose, 300 mg timentin per liter, 0.7% (w/v) phytagar, 300 mg timentinin per liter, and 20 mg kanamycin sulfate per liter, pH 5.8) in Magenta jars. The elongated shoots were transferred to "rooting" medium comprising MS medium, 3% sucrose, 2 mg indole butyric acid per liter, 0.7% phytagar, and 500 mg carbenicillin per liter. After roots emerged, plantlets were transferred to potting mix (Redi Earth, W.R. Grace and Co.). The plants were maintained in a misting chamber (75% relative humidity) under the same growth conditions. Two to three weeks after growth, leaf samples were taken for neomycin phosphotransferase (NPTII) assays (Moloney et al., Plant Cell Reports 8:238-42 (1989)).

Seeds from the FaoA and FaoB transgenic lines can be analyzed for expression of the fatty acid oxidation polypeptides by western blotting using the anti-FaoA and anti-FaoB antibodies. The FaoB polypeptide (SEQ ID NO:26) is not functional in the absence of the FaoA gene product; however, the FaoAB gene product has enzyme activity.

Transgenic lines expressing the FaoA and FaoB complex are obtained

29 by crossing the FaoA and FaoB transgenic lines expressing the individual polypeptides and seeds analyzed by western blotting and enzymes assays as described.

Example 6: Transformation of B. napus cv. Westar and Analysis of Transgenic Lines

Transformation

The protocol used was adopted from a procedure described by Moloney et al (1989). Seeds of Brassica napus cv. Westar were surface sterilized in 10% commercial bleach (Javex, Colgate-Palmolive Canada Inc.) for 30 min with gentle shaking. The seeds were washed three times in sterile distilled water. Seeds were placed on germination medium comprising Murashige-Skoog (MS) salts and vitamins, 3% sucrose and 0.7% phytagar, pH 5.8 at a density of 20 per plate and maintained at 24 °C in a 16 h light/8 h dark photoperiod at a light intensity of 60-80 μEm^'V¹ for four to five days. Each of the constructs, pCGmfl24, pCGmfl25, pCGmf224, pCGmf225, pCGmflP2S, and pCGmf2PlS were introduced into Agrobacterium tumefaciens strain EHAIOI (Hood et al. 1986) by electroporation. Prior to transformation of cotyledonary petioles, single colonies of strain EHA101 harboring each construct were grown in 5 mL of minimal medium supplemented with 100 mg kanamycin per liter, and 100 mg gentamycin per liter for 48 h at 28 °C. One milliliter of bacterial suspension was pelletized by centrifugation for 1 min in a microfuge. The pellet was resuspended in 1 mL minimal medium. For transformation, cotyledons were excised from four-day-old, or in some cases five-day-old, seedlings so that they included approximately 2 mm of petiole at the base. Individual cotyledons with the cut surface of their petioles were immersed in diluted bacterial suspension for 1 s and immediately embedded to a depth of approximately 2 mm in co-cultivation medium, MS medium with 3% sucrose and 0.7% phytagar, enriched with 20 μM benzyladenine. The inoculated cotyledons were plated at a density of 10 per plate and incubated under the same growth conditions for 48 h. After

30 co-cultivation, the cotyledons then were transferred to regeneration medium, which comprised MS medium supplemented with 3% sucrose, 20 μM benzyladenine, 0.7% phytagar, pH 5.8, 300 mg timentinin per liter, and 20 mg kanamycin sulfate per liter.

After two to three weeks, regenerant shoots were obtained, cut, and maintained on "shoot elongation" medium (MS medium containing 3% sucrose, 300mg timentin per liter, 0.7% phytagar, and 20 mg kanamycin per liter, pH 5.8) in Magenta jars. The elongated shoots then were transferred to "rooting" medium, which comprised MS medium, 3% sucrose, 2 mg indole butyric acid per liter, 0.7% phytagar and 500 mg carbenicillin per liter. After roots emerged, the plantlets were transferred to potting mix (Redi Earth, W.R. Grace and Co. Canada Ltd.). The plants were maintained in a misting chamber (75% RH) under the same growth conditions. Two to three weeks after growth, leaf samples were taken for neomycin phosphotransferase (NPT II) assays (Moloney et al. 1989). The results are shown in Table 2 below. The data show the number of plants that were confirmed to be transformed.

Table 2: NPT H Activity in Transformed Plants

Constructs No. of NPTII NPTII No. of plants plants assayed confirmed confirmed transformed

¹pCGmfl24 47 27 23 33 pCGmfl25 37 28 18 18

³pCGmf224 49 40 30 39

⁴pCGmf225 52 37 28 34

⁵pCGmflP2S 27 27 21 21

⁶pCGmf2PlS

'pCGmf 124 - bean phaseolin regulatory sequences driving FaoA gene

²pCGmfl25 - soybean oleosin regulatory sequences driving FaoA gene

³pCGmf224 - bean phaseolin regulatory sequences driving FaoB gene

⁴pCGm£225 - soybean oleosin regulatory sequences driving FaoB gene

⁵pCGmfl92S - bean phaseolin and soybean oleosin regulatory sequences driving FaoA &

FaoB genes, respectively

⁶pCGmf2PlS - bean phaseolin and soybean oleosin regulatory sequences driving FaoB &

FaoA genes, respectively

The fate of the transforming DNA was investigated for sixteen randomly selected transgenic lines. Southern DNA hybridization analysis showed that the FaoA and/or FaoB were integrated into the genomes of the

31 transgenic lines tested.

Approximately 80% of the pmfl24 transgenic plants in which the FaoA gene is expressed from the strong bean phaseolin promoter were observed to be male sterile. Clearly high level expression of the FaoA gene from this promoter results in functional expression of the FaoA gene product which impairs seed and/or pollen development. This result was very unexpected, since it was not anticipated that the plant cells would be capable of carrying out the first step in the β-oxidation pathway in the cytosol. This result, however, provides additional applications for expressing β-oxidation genes in plants for male sterility for hybrid production or to prevent the production of seed. It was also note that in a side-by-side comparison with normal transgenic lines, the pmfl24 lines produced much higher levels of biomass, presumably due to the elimination of seed development. This phenotype therefore may be useful as a means to increase the amount of green biomass produced per acre for silage, forage, or other biomass crops. Here, the use of an inducible promoter system or recombinase technology could be used to produce seed for planting. Seven of the sterile plants were successfully cross-pollinated with pollen from pmf225 transgenic lines and set seeds. Northern analysis on RNA from seeds from pmf224 lines containing the phaseolin promoter-FaoB constructs showed a signal indicative of the expected 1.2 kb transcript in all the samples tested except the control. Northern analysis on RNA from seeds from pmfl25 lines containing the weak soybean oleolsin promoter-FaoA constructs revealed a transcript of the expected size of 2.1 kb. Western blotting on 300-500 μg of protein from approximately 80% of seeds of pmfl25 plants where the FaoA gene is expressed from the relatively weak soybean oleosin promoter were inconclusive, although a weak signal was detected in one transgenic line. Fatty Acid Analysis Given the unexpected results indicating a strong metabolic effect of expressing the FaoA gene from the strong bean phaseolin promoter in seeds, the fatty acid profile of the seeds from transgenic lines expressing the FaoA

32 gene from the weak soybean oleosin promoter was analyzed. Seeds expressing only the FaoA gene or also expressing the FaoB gene from the bean phaseolin promoter were examined. The analysis was carried out as described in Millar et al., The Plant Cell U ΛS89-902 (1998). Seed fatty acid methyl esters (FAMES) were prepared by placing ten seeds of B. napus in 15 x 45-mm screw capped glass tubes and heating at 80 °C in 0.75 mL of IN methanolic HC1 reagent (Supelco, PA) and 10 μL of 1 mg 17:0 methyl ester (internal standard) per mL overnight. After cooling the samples, the FAMES were extracted with 0.3 mL hexane and 0.5 mL 0.9% NaCl by vortexing vigorously. The samples were allowed to stand to separate the phases, and 300 μL of the organic phase was drawn and analyzed on a Hewlett-Packard gas chromatograph.

Fatty acid profile analysis indicated the presence of an additional component or enhanced component in the lipid profile in all of the transgenic plants expressing the FaoA gene SEQ ED NO:24 which was absent from the control plants. This result again proves conclusively that the FaoA gene is being transcribed and translated and that the FaoA polypeptide SEQ ID NO: 27 is catalytically active. This peak also was observed in eleven additional transgenic plants harboring SoyP-FaoA, PhaP-FaoA-SoyP-FaoB, SoyP- FaoA-PhaP-FaoB genes and a sterile (PhaP-FaoA) plant cross-pollinated with SoyP-FaoB. These data clearly demonstrate functional expression of the FaoA gene and that even the very low levels of expression are sufficient to change the lipid profile of the seed. Adapting the methods described herein, one of skill in the art can express these genes at levels intermediate between that obtained with the phaseolin promoter and the soybean oleosin promoter using other promoters such as the Arabidopsis oleosin promoter, napin promoter, or cruciferin promoter, and can use inducible promoter systems or recombinase technologies to control when fatty acid oxidation transgenes are expressed.

Example 7: Yeast β-oxidation Multi-functional Enzyme Complex

S. cerevisiae contains a β-oxidation pathway that proceeds via R-

33 hydroxyacyl CoA rather than the S-3-hydroxyacyl CoA observed in bacteria and higher eukaryotes. The/ox2 gene from yeast encodes a hydratase that produces R-3 -hydroxyacyl CoA from tr raOenoyl-CoA and a dehydrogenase that utilizes R-3-hydroxyacyl-CoA to produce β-keto acyl CoAs.

The fox2 gene (sequence shown in SEQ ED NO: 1) was isolated from S. cerevisiae genomic DNA by PCR in two pieces. Primers N-fox2b and N- bamfox2b were utilized to PCR a 1 J kb Smal/BamHI fragment encoding the N-terminal region of Fox2, and primers C-fox2 and C-bamfox2 were utilized to PCR a 1.6 kb BamHI/Xbal fragment encoding the C-terminal Fox2 region. The full fox2 gene was reconstructed via subcloning in vector pTRCN. N-fox2b fox2 tec ccc ggg agg agg ttt tta tta tgc ctg gaa att tat cct tea aag ata gag tt (SEQ ID NO: 18)

N-bamfox2b fox2 aaggatccttgatgtcatttacaactacc (SEQ ED NO: 19)

C-fox2 fox2 get eta gat agg gaa aga tgt atg taa g (SEQ ED NO: 20) C-bamfox2 fox2 tgacatcaaggatcctttt (SEQ ED NO:21)

The fox 1 gene, however, does not possess a β-ketothiolase activity and this activity must be supplied by a second transgene. Representative sources of such a gene include algae, bacteria, yeast, plants, and mammals. The bacterium Alcaligenes eutrophus possesses a broad specificity β-ketothiolase gene suitable for use in the methods described herein. It can be readily isolated using the acetoacetyl-CoA thiolase gene as a hybridization probe, as described in U.S. Patent 5,661,026 to Peoples et al. This enzyme also has been purified (Hay wood et al., FEMS Micro. Lett. 52:91 (1988)), and the purified enzyme is useful for preparing antibodies or determining protein sequence information as a basis for the isolation of the gene.

34 Example 8: Plant β-Oxidation Gene

The DNA sequence of the cDNA encoding β-oxidation tetrafunctional protein, shown in SEQ ID NO:4, can be isolated as described in Preisig-Muller et al., J. Biol. Chem. 269:20475-81 (1994). The equivalent gene can be isolated from other plant species including Arabidopsis, Brassica, soybean, sunflower, and corn using similar procedures or by screening genomic libraries, many of which are commercially available, for example from Clontech Laboratories Inc., Palo Alto, California, USA. A peroxisomal targeting sequence P-R-M was identified at the carboxy terminus of the protein. Constructs suitable for expressing in the plant cytosol can be prepared by PCR amplification of this gene using primers designed to delete this sequence.

35

Claims

We claim:

1. A method for manipulating the metabolism of a plant, comprising expressing heterologous genes encoding fatty acid oxidation enzymes in cytosol or plastids other than the peroxisomes, glyoxisomes or mitochondria of the plant.

2. The method of claim 1 wherein the fatty acid ╬▓-oxidation enzymes are expressed from genes selected from the group consisting of bacterial, yeast, fungal, plant, and mammalian.

3. The method of claim 2 wherein the fatty acid oxidation enzymes are expressed from genes from bacteria selected from the group consisting of Escherichia, Pseudomonas, Alcaligenes, and Coryneform.

4. The method claim 3 wherein the genes are Pseudomonas putida faoAB.

5. The method of claim 1 further comprising expressing genes encoding enzymes selected from the group consisting of polyhydroxyalkanoate synthases, acetoacetyl-CoA reductases, ╬▓-ketoacyl-CoA thiolases, and enoyl-CoA hydratases.

6. A DNA construct for use in a method of manipulating the metabolism of a plant cell comprising, in phase,

(a) a promoter region functional in a plant;

(b) a structural DNA sequence encoding at least one fatty acid oxidation enzyme activity; and

(c) a 3 ' nontranslated region of a gene naturally expressed in a plant, wherein the nontranslated region encodes a signal sequence for polyadenylation of mRNA.

7. The DNA construct of claim 6 wherein the promoter is a seed specific promoter.

8. The DNA construct of claim 7 wherein the seed specific promoter is selected from the group consisting of napin promoter, phaseolin promoter, oleosin promoter, 2S albumin promoter, zein promoter, ╬▓-conglycinin promoter, acyl-carrier protein promoter, and fatty acid desaturase promoter.

36 9 The DNA construct of claim 6 wherein the promoter is a constitutive promoter

10 The DNA construct of claim 6 wherein the promoter is selected from the group consisting of CaMV 35S promoter, enhanced CaMV 35S promoter, and ubiquitin promoter

11 A method for enhancing the biological production of polyhydroxyalkanoates in a transgenic plant, comprising expressing genes encoding heterologous fatty acid oxidation enzymes in cytosol or plastids other than the peroxisomes, glyoxisomes or mitochondria of the plant

12 The method of claim 11 wherein the transgenic plant is selected from the group consisting of Brassica, maize, soybean, cottonseed, sunflower, palm, coconut, safflower, peanut, mustards, flax, tobacco, and alfalfa

13 A transgenic plant or part thereof comprising heterologous genes encoding fatty acid oxidation enzymes in cytosol or plastids other than the peroxisomes, glyoxisomes or mitochondria of the plant

14 The plant or part thereof of claim 13 wherein the fatty acid ╬▓- oxidation enzymes are expressed from genes selected from the group consisting of bacterial, yeast, fungal, plant, and mammalian

15 The plant or part thereof of claim 14 wherein the fatty acid oxidation enzymes are expressed from genes from bacteria selected from the group consisting of Escherichia, Pseudomonas, Alcaligenes, and Coryneform

16 The plant or part thereof of claim 15 wherein the genes are Pseudomonas putida fao AΑ

17 The plant or part thereof of claim 13 further comprising genes encoding enzymes selected from the group consisting of polyhydroxyalkanoate synthases, acetoacetyl-CoA reductases, ╬▓-ketoacyl- CoA thiolases, and enoyl-CoA hydratases

37 18 The plant or part thereof of claim 13 wherein the plant is selected from the group consisting of Brassica, maize, soybean, cottonseed, sunflower, palm, coconut, safflower, peanut, mustards, flax, tobacco, and alfalfa

19 The plant or part thereof comprising a DNA construct comprising, in phase,

(a) a promoter region functional in a plant,

(b) a structural DNA sequence encoding at least one fatty acid oxidation enzyme activity, and

(c) a 3 ' nontranslated region of a gene naturally expressed in a plant, wherein the nontranslated region encodes a signal sequence for polyadenylation of mRNA

20 The plant or part thereof of claim 19 wherein the promoter is a seed specific promoter

21 The plant or part thereof of claim 20 wherein the seed specific promoter is selected from the group consisting of napm promoter, phaseolin promoter, oleosin promoter, 2S albumin promoter, zein promoter, ╬▓- conglycimn promoter, acyl-carrier protein promoter, and fatty acid desaturase promoter

22 The plant or part thereof of claim 19 wherein the promoter is a constitutive promoter

23 The plant or part thereof of claim 20 wherein the promoter is selected from the group consisting of CaMV 35S promoter, enhanced CaMV 35S promoter, and ubiquitin promoter

24 A method of preventing or suppressing seed production in a plant, compnsmg expressing heterologous genes encoding fatty acid oxidation enzymes in cytosol or plastids other than the peroxisomes, glyoxisomes or mitochondria of the plant

38