WO2004101757A2 - Production of polyunsaturated fatty acids in oleaginous yeasts - Google Patents

Abstract

The present invention relates to methods for the production of ω-3 and/or ω-6 fatty acids in oleaginous yeast. Thus, desaturases and elongases able to catalyze the conversion of linoleic acid (LA) to Ϝ-linolenic acid (GLA); α-linoleic acid (ALA) to stearidonic acid (STA); GLA to dihomo-Ϝ- linoleic acid (DGLA); STA to eicosatetraenoic acid (ETA); DGLA to arachidonic acid (ARA); ETA to eicosapentaenoic acid (EPA); DGLA to ETA; EPA to docosapentaenoic acid (DPA); and ARA to EPA have been introduced into the genome of Yarrowia for synthesis of ARA and EPA.

Description

TITLE PRODUCTION OF POLYUNSATURATED FATTY ACIDS IN

OLEAGINOUS YEASTS This application claims the benefit of U.S. Provisional Application No. 60/468677, filed May 7, 2003.

FIELD OF THE INVENTION This invention is in the field of biotechnology. More specifically, this invention pertains to the production of long chain polyunsaturated fatty acids (PUFAs) in oleaginous yeasts. BACKGROUND OF THE INVENTION

It has long been recognized that certain polyunsaturated fatty acids, or PUFAs, are important biological components of healthy cells. For example, such PUFAs are recognized as:

• "Essential" fatty acids that can not be synthesized de novo in mammals and instead must be obtained either in the diet or derived by further desaturation and elongation of linoleic acid (LA) or - linolenic acid (ALA);

• Constituents of plasma membranes of cells, where they may be found in such forms as phospholipids or triglycerides; • Necessary for proper development, particularly in the developing infant brain, and for tissue formation and repair; and,

• Precursors to several biologically active eicosanoids of importance in mammals, including prostacyclins, eicosanoids, leukotrienes and prostaglandins. In the 1970's, observations of Greenland Eskimos linked a low incidence of heart disease and a high intake of long-chain ω-3 PUFAs (Dyerberg, J. et al., Amer. J. Clin Nutr. 28:958-966 (1975); Dyerberg, J. et al., Lancet 2(8081): 117-119 (July 15, 1978)). More recent studies have confirmed the cardiovascular protective effects of ω-3 PUFAs (Shimokawa, H., World Rev NutrDiet, 88:100-108 (2001); von Schacky, C, and Dyerberg, J., World Rev Nutr Diet, 88:90-99 (2001)). Further, it has been discovered that several disorders respond to treatment with ω-3 fatty acids, such as the rate of restenosis after angioplasty, symptoms of inflammation and rheumatoid arthritis, asthma, psoriasis and eczema. γ-linolenic acid (GLA, an ω-6 PUFA) has been shown to reduce increases in blood pressure associated with stress and to improve performance on arithmetic tests. GLA and dihomo-γ-linolenic acid (DGLA, another ω-6 PUFA) have been shown to inhibit platelet aggregation, cause vasodilation, lower cholesterol levels and inhibit proliferation of vessel wall smooth muscle and fibrous tissue (Brenner et al., Adv. Exp. Med. Biol. 83:85-101 (1976)). Administration of GLA or DGLA, alone or in combination with eicosapentaenoic acid (EPA, an ω-3 PUFA), has been shown to reduce or prevent gastrointestinal bleeding and other side effects caused by non-steroidal anti-inflammatory drugs (U.S. 4,666,701). Further, GLA and DGLA have been shown to prevent or treat endometriosis and premenstrual syndrome (U.S. 4,758,592) and to treat myalgic encephalomyelitis and chronic fatigue after viral infections

(U.S. 5,116,871). Other evidence indicates that PUFAs may be involved in the regulation of calcium metabolism, suggesting that they may be useful in the treatment or prevention of osteoporosis and kidney or urinary tract stones. Finally, PUFAs can be used in the treatment of cancer and diabetes (U.S. 4,826,877; Horrobin et al., Am. d. Clin. Nutr. 57 (Suppl.): 732S-737S (1993)).

PUFAs are generally divided into two major classes (consisting of the ω-6 and the ω-3 fatty acids) that are derived by desaturation and elongation of the essential fatty acids, linoleic acid (LA) and α-linolenic acid (ALA), respectively. Despite this common derivation from "essential" fatty acids, it is becoming increasingly apparent that the ratio of ω-6 to ω-3 fatty acids in the diet is important for maintenance of good health. Due to changes in human dietary habits, the current ratio of ω-6 to ω-3 fatty acids is approximately 10:1 , whereas the preferred ratio is 2:1 (Kris-Etherton, P.M. et al., Am. J. Clin. Nutr. 71(1 Suppl.): 179S-88S (2000); Simopoulos, A. P. et al., Ann. Nutr. Metab. 43:127-130 (1999); Krauss, R. M. et al. AHA Circulation 102:2284-2299 (2000)).

The main sources of ω-6 fatty acids are vegetable oils (e.g., corn oil, soy oil) that contain high amounts of LA. GLA is found in the seeds of a number of plants, including evening primrose (Oenothera biennis), borage (Borago officinalis) and black currants (Ribes nigrum). Microorganisms in the genera Mortierella (filamentous fungus), Entomophthora, Pythium and Porphyridium (red alga) can be used for commercial production of the ω-6 fatty acid, arachidonic acid (ARA). The fungus Mortierella alpina, for example, is used to produce an oil containing ARA, while U.S. 5,658,767 (Martek Corporation) teaches a method for the production of an oil containing ARA comprising cultivating Pythium insidiuosum in a culture medium containing a carbon and nitrogen source. The ω-3 PUFAs of importance include EPA and docosahexaenoic acid (DHA), both of which are found in different types of fish oil and marine plankton. U.S. 5,244,921 (Martek Corporation) describes a process for producing an edible oil containing EPA by cultivating heterotrophic diatoms in a fermentor, specifically Cyclotella sp. and Nitzschia sp. DHA can be obtained from cold water marine fish, egg yolk fractions and by cultivation of certain heterotrophic microalgae of the class Dinophyceae, specifically, Crypthecodinium sp. such as C. cohnii (U.S. 5,492,938 and U.S. 5,407,957). Stearidonic acid (STA), a precursor to EPA and DHA, can be found in marine oils and plant seeds; its commercial sources include production in the genera Trichodesma and Echium. Other sources of ω-3 acids are found in flaxseed oil and walnut oil, each containing predominantly ALA.

Despite a variety of commercial sources of PUFAs from natural sources, there are several disadvantages associated with these methods of production. First, natural sources such as fish and plants tend to have highly heterogeneous oil compositions. The oils obtained from these sources therefore can require extensive purification to separate or enrich one or more of the desired PUFAs. Fish oils commonly have unpleasant tastes and odors, which may be impossible to separate economically from the desired product and can render such products unacceptable as food supplements. Unpleasant tastes and odors can make medical regimens based on ingestion of high dosages undesirable, and may inhibit compliance by the patient. Furthermore, fish may accumulate environmental pollutants and ingestion of fish oil capsules as a dietary supplement may result in ingestion of undesired contaminants. Natural sources are also subject to uncontrollable fluctuations in availability (e.g., due to weather, disease, or over-fishing in the case of fish stocks); and, crops that produce PUFAs often are not competitive economically with hybrid crops developed for food production. Large-scale fermentation of some organisms that naturally produce PUFAs (e.g., Porphyridium, Mortierella) can also be expensive and/or difficult to cultivate on a commercial scale.

As a result of the limitations described above, extensive work has been conducted toward: 1.) the development of recombinant sources of PUFAs that are easy to produce commercially; and 2.) modification of fatty acid biosynthetic pathways, to enable production of desired PUFAs.

Advances in the isolation, cloning and manipulation of fatty acid desaturase and elongase genes from various organisms have been made over the last several years. Knowledge of these gene sequences offers the prospect of producing a desired fatty acid and/or fatty acid composition in novel host organisms that do not naturally produce PUFAs. The literature reports a number of examples in Saccharomyces cerevisiae, such as: 1. Domergue, F. et al. (Eur. J. Biochem. 269:4105-4113 (2002)), wherein two desaturases from the marine diatom Phaeodactylum tricornutum were cloned into S. cerevisiae, leading to the production of EPA;

2. Beaudoin F., et al. (Proc. Natl. Acad. Sci. U.S.A. 97(12):6421-6 (2000)), wherein the ω-3 and ω-6 PUFA biosynthetic pathways were reconstituted in S. cerevisiae, using genes from Caenorhabditis elegans;

3. Dyer, J.M. et al. (Appl. Eniv. Microbiol., 59:224-230 (2002)), wherein plant fatty acid desaturases (FAD2 and FAD3) were expressed in S. cerevisiae, leading to the production of ALA; and

4. U.S. 6,136,574 (Knutzon et al., Abbott Laboratories), wherein one desaturase from Brassica napus and two desaturases from the fungus Mortierella alpina were cloned into S. cerevisiae, leading to the production of LA, GLA, ALA and STA. There remains a need, however, for an appropriate microbial system in which these types of genes can be expressed to provide for economical production of commercial quantities of one or more PUFAs. Additionally, a need exists for oils enriched in specific PUFAs, notably EPA and DHA.

Many microorganisms (including algae, bacteria, molds and yeasts) can synthesize oils in the ordinary course of cellular metabolism. Thus, oil production involves cultivating the microorganism in a suitable culture medium to allow for oil synthesis, followed by separation of the microorganism from the fermentation medium and treatment for recovery of the intracellular oil. Attempts have been made to optimize production of fatty acids by fermentive means involving varying such parameters as microorganisms used, media and conditions that permit oil production. However, these efforts have proved largely unsuccessful in improving yield of oil or the ability to control the characteristics of the oil composition produced.

One class or microorganisms that has not been previously examined as a production platform for PUFAs, however, are the oleaginous yeasts. These organisms can accumulate oil up to 80% of their dry cell weight. The technology for growing oleaginous yeast with high oil content is well developed (for example, see EP 0 005 277B1; Ratledge, C, Prog. Ind. Microbiol. 16:119-206 (1982)), and may offer a cost advantage compared to commercial micro-algae fermentation for production of ω-3 or ω-6 PUFAs. Whole yeast cells may also represent a convenient way of encapsulating ω-3 or ω-6 PUFA-enriched oils for use in functional foods and animal feed supplements.

Despite the advantages noted above, oleaginous yeast are naturally deficient in ω-6 and ω-3 PUFAs, since naturally produced PUFAs in these organisms are limited to 18:2 fatty acids (and less commonly, 18:3 fatty acids). Thus, the problem to be solved is to develop an oleaginous yeast that accumulates oils enriched in ω-3 and/or ω-6 fatty acids. Toward this end, it is necessary to introduce desaturases and elongases that allow for the synthesis and accumulation of ω-3 and/or ω-6 fatty acids in oleaginous yeasts. Although advances in the art of genetic engineering have been made, such techniques have not been developed for oleaginous yeasts. Thus, one must overcome problems associated with the use of these particular host organisms for the production of PUFAs. Applicants have solved the stated problem by demonstrating production of PUFAs in the host Yarrowia lipolytica, following the introduction of a heterologous ω-6 and/or ω-3 biosynthetic pathway. Specifically, ARA (representative of ω-6 fatty acids) and EPA (representative of ω-3 fatty acids) were produced herein, to exemplify the techniques of the invention. SUMMARY OF THE INVENTION

The present invention provides methods for the expression of enzymes comprising the ω-3/ω-6 fatty acid biosynthetic pathway in an oleaginous yeast host for the production of ω-3 and/or ω-6 fatty acids. Accordingly, the invention provides a method for the production of ω-3 and/or ω-6 fatty acids comprising: a) providing an oleaginous yeast comprising a functional ω-3/ ω-6 fatty acid biosynthetic pathway; b) growing the yeast of step (a) in the presence of a fermentable carbon source whereby an ω-3 or ω-6 fatty acid is produced; and c) optionally recovering the ω-3 or ω-6 fatty acid.

In one specific embodiment the invention provides a method for the production of linoleic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ12 desaturase polypeptide; and (ii) an endogenous source of oleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a Δ12 desaturase polypeptide is expressed and the oleic acid is converted to linoleic acid; and c) optionally recovering the linoleic acid of step (b).

In specific embodiments the invention provides for the production of specific ω-6 fatty acids such as linoleic acid (LA), γ-linolenic acid (GLA), dihomo-γ- linoleic acid (DGLA) and arachidonic acid (ARA) by de novo biosynthesis or single step enzymatic reactions from the appropriate precursors. Similarly the invention provides for the production of specific ω-3 fatty acids such as α-linoleic acid (ALA), stearidonic acid (STA), eicosatetraenoic acid (ETA), eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA) and docosahexaenoic acid by single step enzymatic reactions from the appropriate precursors.

BRIEF DESCRIPTION OF THE DRAWINGS AND SEQUENCE DESCRIPTIONS Figure 1 shows a schematic illustration of the biochemical mechanism for lipid accumulation in oleaginous yeast.

Figure 2 illustrates the ω-3/ω-6 fatty acid biosynthetic pathway. Figure 3 illustrates the construction of plasmid vector pY5 for gene expression in Yarrowia lipolytica. Figure 4 illustrates the construction of plasmid vectors pY5-4 and pY5-13 for gene expression in Y. lipolytica.

Figure 5 is a schematic presentation of the construction of intermediate vector pYZM5CHPPA.

Figure 6 show a comparison between the DNA sequence of the Saprolegnia diclina Δ17 desaturase gene and the synthetic gene codon- optimized for expression in Y. lipolytica. Figure 7 illustrates the favored consensus sequences around the translation initiation codon 'ATG' in Y. lipolytica.

Figure 8 illustrates the strategy utilized for in vitro synthesis of the codon-optimized Δ17 desaturase gene. Figure 9 shows plasmids for expression of the synthetic codon- optimized and wildtype Δ17 desaturase genes in Y. lipolytica.

Figures 10A and 10B show the results of gas chromatographic analysis of fatty acids produced in Y. lipolytica transformed with the wildtype and synthetic codon-optimized Δ17 desaturase genes, respectively.

Figure 11 is a schematic presentation of the construction of intermediate vector pY24-4.

Figure 12 is a schematic presentation of the construction of intermediate vector pYZV16. Figure 13 is a schematic presentation of the construction of integration vector pYZM5EL6.

Figure 14 is a schematic presentation of the construction of integration vectors pYZV5EL6 and pYZV5EL6/17.

Figure 15 is a chromatogram illustrating the production of ARA from an engineered Y. lipolytica.

Figure 16 is a chromatogram illustrating the production of EPA from an engineered Y. lipolytica.

The invention can be more fully understood from the following detailed description and the accompanying sequence descriptions, which form a part of this application.

The following sequences comply with 37 C.F.R. §1.821-1.825 ("Requirements for Patent Applications Containing Nucleotide Sequences and/or Amino Acid Sequence Disclosures - the Sequence Rules") and are consistent with World Intellectual Property Organization (WIPO) Standard ST.25 (1998) and the sequence listing requirements of the EPO and PCT (Rules 5.2 and 49.5(a-bis), and Section 208 and Annex C of the Administrative Instructions). The symbols and format used for nucleotide and amino acid sequence data comply with the rules set forth in 37 C.F.R. §1.822. SEQ ID NO:1 shows the DNA sequence of the Mortierella alpina Δ6 desaturase gene, while SEQ ID NO:2 shows the amino acid sequence of the M. alpina Δ6 desaturase. SEQ ID NO:3 shows the DNA sequence of the Mortierella alpina Δ5 desaturase gene, while SEQ ID NO:4 shows the amino acid sequence of the M. alpina Δ5 desaturase.

SEQ ID NO:5 shows the DNA sequence of the Saprolegnia diclina Δ17 desaturase gene, while SEQ ID NO:6 shows the corresponding amino acid sequence of the S. diclina Δ17 desaturase.

SEQ ID NO:7 shows the DNA sequence of the Mortierella alpina high affinity elongase gene, while SEQ ID NO:8 shows the amino acid sequence of the M. alpina high affinity elongase. SEQ ID NO:9 shows the DNA sequence of the synthetic Δ17 desaturase gene codon-optimized for expression in Yarrowia lipolytica.

SEQ ID NOs:10-31 correspond to the 11 pairs of oligonucleotides that together comprise the entire codon-optimized coding region of the S. diclina Δ17 desaturase gene (e.g., D17-1A, D17-1 B, D17-2A, D17-2B, D17-3A, D17-3B, D17-4A, D17-4B, D17-5A, D17-5B, D17-6A, D17-6B, D17-7A, D17-7B, D17-8A, D17-8B, D17-9A, D17-9B, D17-10A, D17-10B, D17-11A and D17-11B, respectively).

SEQ ID NOs:32-37 correspond to primers D17-1 , D17-4R, D17-5, D17-8D, D17-8U and D17-11 , respectively, used for PCR amplification during synthesis of the codon-optimized Δ17 desaturase gene.

SEQ ID NOs:38 and 39 correspond to primers TEF5' and TEF3', respectively, used to isolate the TEF promoter.

SEQ ID NOs:40 and 41 correspond to primers XPR5' and XPR3', respectively, used to isolate the XPR2 transcriptional terminator. SEQ ID NOs:42 and 43 correspond to primers YL21A and YL22, used for amplifying the wild type Δ17 desaturase gene of S. diclina from plasmid pRSP19.

SEQ ID NOs:44 and 45 correspond to primers YL53 and YL54, respectively, used for site-directed mutagenesis to generate pYSD17M. SEQ ID NOs:46 and 47 correspond to primers KU5 and KU3, respectively, used for amplifying a 1.7 kB DNA fragment (SEQ ID NO:48; amino acid sequence provided as SEQ ID NO:49) containing the Yarrowia URA3 gene.

SEQ ID NOs:50 and 51 correspond to primers KI5 and KI3, respectively, used for amplifying a 1.1 kB DNA fragment (SEQ ID NO:52; amino acid sequence provided as SEQ ID NO:53) containing the conjugase gene of Impatients balsama. SEQ ID NOs:54 and 55 correspond to primers KTI5 and KTI3, respectively, used for amplifying a 1.7 kB DNA fragment (SEQ ID NO:56; amino acid sequence provided as SEQ ID NO:57) containing a TEF::conjugase::XPR chimeric gene. SEQ ID NOs:58 and 59 correspond to primers KH5 and KH3, respectively, used for amplifying a 1 kB DNA fragment (SEQ ID NO:60; amino acid sequence provided as SEQ ID NO:61) containing the E. coli hygromycin resistance gene.

SEQ ID NOs:62 and 63 correspond to primers KTH5 and KTH3, respectively, used for amplifying a 1.6 kB DNA fragment (SEQ ID NO:64; amino acid sequence provided as SEQ ID NO:65) containing the TEF::HPT::XPR fusion gene.

SEQ ID NOs:66 and 67 correspond to the 401 bp of 5'-sequence and 568 bp of 3'-sequence of the Yarrowia lipolytica URA3 gene, respectively, used to direct integration of expression cassettes into the Ura loci of the Yarrowia genome.

SEQ ID NOs:68-71 correspond to primers YL63, YL64, YL65 and YL66, respectively, used for site-directed mutagenesis to generate pY24- 4. SEQ ID NOs:72 and 73 correspond to primers YL11 and YL12, respectively, used for amplifying the M. alpina Δ5 desaturase.

SEQ ID NOs:74-77 correspond to primers YL81 , YL82, YL83 and YL84, respectively, used for site-directed mutagenesis to generate pYZM5CH. SEQ ID NOs:78 and 79 correspond to primers YL105 and YL106, respectively, used for site-directed mutagenesis to generate pYZM5CHPP.

SEQ ID NOs:80 and 81 correspond to primers YL119 and YL120, respectively, used for site-directed mutagenesis to generate pYZMδCHPPA. SEQ ID NOs:82 and 83 correspond to primers YL121 and YL122, respectively, used for amplifying 440 bp of 5'-non-coding DNA sequence (SEQ ID NO:84) upstream from the Y. lipolytica URA3 gene.

SEQ ID NOs:85 and 86 correspond to primers YL114 and YL115, respectively, used for site-directed mutagenesis to generate pYZV5 and pYZV5P. SEQ ID NO:87 corresponds to a 5.2 kB DNA fragment suitable for integration and expression of the M. alpina Δ5 desaturase gene in the Yarrowia lipolytica genome.

SEQ ID NOs:88-91 correspond to primers YL61 , YL62, YL69 and YL70, respectively, used for site-directed mutagenesis to generate pY58BH.

SEQ ID NOs:92-95 correspond to primers YL77, YL78, YL79A and YL80A, respectively, used for site-directed mutagenesis to generate pY54PC.

SEQ ID NO:96 corresponds to a 8.9 kB DNA fragment suitable for integration and coordinate expression of the M. alpina Δ6 desaturase, M. alpina elongase and M. alpina Δ5 desaturase genes in the Yarrowia lipolytica genome.

SEQ ID NOs:97-100 correspond to primers YL101 , YL102, YL103 and YL104, respectively, used for site-directed mutagenesis to generate pYSD17SPC.

SEQ ID NO: 101 corresponds to a 10.3 kB DNA fragment suitable for integration and coordinate expression of the M. alpina Δ6 desaturase, M. alpina elongase, M. alpina Δ5 desaturase and codon-optimized Δ17 desaturase genes in the Yarrowia lipolytica genome. SEQ ID NOs: 102-113 correspond to primers YL1 , YL2, YL3, YL4,

YL5, YL6, YL7, YL8, YL9, YL10, YL23 and YL24, respectively, used for plasmid construction.

SEQ ID NO:114 shows the DNA sequence of the Saprolegnia diclina Δ5 desaturase gene, while SEQ ID NO:115 shows the amino acid sequence of the S. diclina Δ5 desaturase. '

SEQ ID NOs:116, 117, 120, 121 , 124 and 125 correspond to primers YL13A, YL14, YL19A, YL20, YL15 and YL16B, respectively, used for cloning various Δ5 desaturases.

SEQ ID NO:118 shows the DNA sequence of the Isochrysis galbana Δ5 desaturase gene, while SEQ ID NO:119 shows the amino acid sequence of the /. galbana Δ5 desaturase.

SEQ ID NO: 122 shows the DNA sequence of the Thraustochytrium aureum Δ5 desaturase gene, while SEQ ID NO: 123 shows the amino acid sequence of the T. aureum Δ5 desaturase. SEQ ID NO: 126 corresponds to the codon-optimized translation initiation site for genes optimally expressed in Yarrowia sp. DETAILED DESCRIPTION OF THE INVENTION In accordance with the subject invention, Applicants provide methods for the production of ω-3 and/or ω-6 fatty acids in oleaginous yeasts. Specifically, Applicants provide methods for production of linoleic acid, γ-linolenic acid, dihomo- γ -linolenic acid, arachidonic acid, α-linolenic acid, stearidonic acid, eicosatetraenoic acid, eicosapentaenoic acid, docosapentaenoic acid and docosahexaenoic acid. This is accomplished by introduction of functional ω-3/ ω-6 fatty acid biosynthetic pathway encoded by genes conferring Δ17 desaturase, Δ6 desaturase, Δ5 desaturase, Δ9 desaturase, Δ12 desaturase, Δ15 desaturase, Δ4 desaturase and elongase activities into oleaginous yeast hosts for recombinant expression. Thus, this disclosure demonstrates that oleaginous yeasts can be engineered to enable production of any PUFA composition that is desired. The subject invention finds many applications. PUFAs, or derivatives thereof, made by the methodology disclosed herein can be used as dietary substitutes, or supplements, particularly infant formulas, for patients undergoing intravenous feeding or for preventing or treating malnutrition. Alternatively, the purified PUFAs (or derivatives thereof) may be incorporated into cooking oils, fats or margarines formulated so that in normal use the recipient would receive the desired amount for dietary supplementation. The PUFAs may also be incorporated into infant formulas, nutritional supplements or other food products and may find use as anti-inflammatory or cholesterol lowering agents. Optionally, the compositions may be used for pharmaceutical use (human or veterinary). In this case, the PUFAs are generally administered orally but can be administered by any route by which they may be successfully absorbed, e.g., parenterally (e.g., subcutaneously, intramuscularly or intravenously), rectally, vaginally or topically (e.g., as a skin ointment or lotion). Supplementation of humans or animals with PUFAs produced by recombinant means can result in increased levels of the added PUFAs, as well as their metabolic progeny. For example, treatment with arachidonic acid (ARA) can result not only in increased levels of ARA, but also downstream products of ARA such as prostaglandins. Complex regulatory mechanisms can make it desirable to combine various PUFAs, or add different conjugates of PUFAs, in order to prevent, control or overcome such mechanisms to achieve the desired levels of specific PUFAs in an individual.

Definitions

In this disclosure, a number of terms and abbreviations are used. The following definitions are provided.

"Open reading frame" is abbreviated ORF. "Polymerase chain reaction" is abbreviated PCR. "American Type Culture Collection" is abbreviated ATCC. "Polyunsaturated fatty acid(s)" is abbreviated PUFA(s). The term "fatty acids" refers to long chain aliphatic acids (alkanoic acids) of varying chain lengths, from about C^₂ to C22 (although both longer and shorter chain-length acids are known). The predominant chain lengths are between C ₆ and C 2- ^Tne structure of a fatty acid is represented by a simple notation system of "X:Y", where X is the total number of carbon (C) atoms and Y is the number of double bonds.

Generally, fatty acids are classified as saturated or unsaturated. The term "saturated fatty acids" refers to those fatty acids that have no "double bonds" between their carbon backbone. In contrast, "unsaturated fatty acids" are c/s-isomers that have "double bonds" along their carbon backbones. "Monounsaturated fatty acids" have only one "double bond" along the carbon backbone (e.g., usually between the 9^th and 10^th carbon atom as for palmitoleic acid (16:1) and oleic acid (18:1)), while "polyunsaturated fatty acids" (or "PUFAs") have at least two double bonds along the carbon backbone (e.g., between the 9^th and 10^th, and 12^th and 13^th carbon atoms for linoleic acid (18:2); and between the 9^th and 10^th, 12*h and 13th, _anα 15th _{and 16}th for α-linolenic acid (18:3)).

"PUFAs" can be classified into two major families (depending on the position (n) of the first double bond nearest the methyl end of the fatty acid carbon chain). Thus, the "ω-6 fatty acids" (ω-6 or n ) have the first unsaturated double bond six carbon atoms from the omega (methyl) end of the molecule and additionally have a total of two or more double bonds, with each subsequent unsaturation occurring 3 additional carbon atoms toward the carboxyl end of the molecule. In contrast, the "ω-3 fatty acids" (ω-3 or n-3) have the first unsaturated double bond three carbon atoms away from the omega end of the molecule and additionally have a total of three or more double bonds, with each subsequent unsaturation occurring 3 additional carbon atoms toward the carboxyl end of the molecule. For the purposes of the present disclosure, the omega-reference system will be used to indicate the number of carbons, the number of double bonds and the position of the double bond closest to the omega carbon, counting from the omega carbon (which is numbered 1 for this purpose). This nomenclature is shown below in Table 1 , in the column titled "Shorthand Notation". The remainder of the Table summarizes the common names of ω-3 and ω-6 fatty acids, the abbreviations that will be used throughout the specification and each compounds' chemical name.

Table 1 Nomenclature of Polyunsaturated Fatty Acids

The term "essential fatty acid" refers to a particular PUFA that an organism must ingest in order to survive, being unable to synthesize the particular essential fatty acid de novo. For example, mammals can not synthesize the essential fatty acid LA (18:2, ω-6). Other essential fatty acids include GLA (ω-6), DGLA (ω-6), ARA (ω-6), EPA (ω-3) and DHA (ω-3).

The term "fat" refers to a lipid substance that is solid at 25 °C and usually saturated.

The term "oil" refers to a lipid substance that is liquid at 25 °C and usually polyunsaturated. PUFAs are found in the oils of some algae, oleaginous yeasts and filamentous fungi. "Microbial oils" or "single cell oils" are those oils naturally produced by microorganisms during their lifespan. Such oils can contain long chain PUFAs.

The term "PUFA biosynthetic pathway enzyme" refers to any of the following enzymes (and genes which encode said enzymes) associated with the biosynthesis of a PUFA, including: a Δ4 desaturase, a Δ5 desaturase, a Δ6 desaturase, a Δ12 desaturase, a Δ15 desaturase, a Δ17 desaturase, a Δ9 desaturase and/or an elongase.

The term "ω-3/ω-6 fatty acid biosynthetic pathway" refers to a set of genes which, when expressed under the appropriate conditions encode enzymes that catalyze the production of either or both ω-3 and ω-6 fatty acids. Typically the genes involved in the ω-3/ω-6 fatty acid biosynthetic pathway encode some or all of the following enzymes: Δ12 desaturase, Δ6 desaturase, elongase, Δ5 desaturase, Δ17 desaturase, Δ15 desaturase, Δ9 desaturase and Δ4 desaturase. A representative pathway is illustrated in Figure 2, which demonstrates how both ω-3 and ω-6 fatty acids may be produced from a common source. The term "functional" as used herein in context with the ω-3/ω-6 fatty acid biosynthetic pathway means that some or all of the genes in the pathway express active enzymes. It should be understood that "ω-3/ω-6 fatty acid biosynthetic pathway" or "functional ω- 3/ω-6 fatty acid biosynthetic pathway" does not imply that all the genes listed in this paragraph are required as a number of fatty acid products will only require the expression of a subset of the genes of this pathway.

The term "desaturase" refers to a polypeptide component of a multi- enzyme complex that can desaturate one or more fatty acids to produce a mono- or polyunsaturated fatty acid or precursor of interest. Despite use of the omega-reference system throughout the specification to refer to specific fatty acids, it is more convenient to indicate the activity of a desaturase by counting from the carboxyl end of the substrate using the delta-system. Of particular interest herein are: 1.) Δ17 desaturases that desaturate a fatty acid between the 17^th and 18^th carbon atom numbered from the carboxyl-terminal end of the molecule and which, for example, catalyze the conversion of ARA to EPA and/or DGLA to ETA; 2.) Δ6 desaturases that catalyze the conversion of LA to GLA and/or ALA to STA; 3.) Δ5 desaturases that catalyze the conversion of DGLA to ARA and/or ETA to EPA; 4.) Δ4 desaturases that catalyze the conversion of DPA to DHA; 5.) Δ12 desaturases that catalyze the conversion of oleic acid to LA; 6.) Δ15 desaturases that catalyze the conversion of LA to ALA; and 7.) Δ9 desaturases that catalyze the conversion of palmitate to palmitoleic acid (16:1) and/or stearate to oleic acid (18:1).

The term "elongase" refers to a polypeptide component of a multi- enzyme complex that can elongate a fatty acid carbon chain to produce a mono- or polyunsaturated fatty acid that is 2 carbons longer than the fatty acid substrate that the elongase acts upon. This process of elongation occurs in a multi-step mechanism in association with fatty acid synthase, whereby CoA is the acyl carrier (Lassner et al., The Plant Cell 8:281-292 (1996)). Briefly, malonyl-CoA is condensed with a long-chain acyl-CoA to yield CO₂ and a β-ketoacyl-CoA (where the acyl moiety has been elongated by two carbon atoms). Subsequent reactions include reduction to β-hydroxyacyl-CoA, dehydration to an enoyl-CoA and a second reduction to yield the elongated acyl-CoA. Examples of reactions catalyzed by elongases are the conversion of GLA to DGLA, STA to ETA and EPA to DPA. Accordingly, elongases can have different specificities (e.g., a C_16/18 elongase will prefer a C₁₆ substrate, a C_18/20 elongase will prefer a C₁₈ substrate and a C_20/22 elongase will prefer a C₂₀ substrate). The term "high affinity elongase" refers to an elongase whose substrate specificity is preferably for GLA (with DGLA as a product of the elongase reaction). One such elongase is described in WO 00/12720. The terms "conversion efficiency" and "percent substrate conversion" refer to the efficiency by which a particular enzyme (e.g., a desaturase or elongase) can convert substrate to product. The conversion efficiency is measured according to the following formula: ([product]/[substrate+product])*100, where 'product' includes the immediate product and all products in the pathway derived from it. The term "oleaginous" refers to those organisms that tend to store their energy source in the form of lipid (Weete, In: Fungal Lipid Biochemistry, 2^nd Ed., Plenum, 1980). Generally, the cellular PUFA content of these microorganisms follows a sigmoid curve, wherein the concentration of lipid increases until it reaches a maximum at the late logarithmic or early stationary growth phase and then gradually decreases during the late stationary and death phases (Yongmanitchai and Ward, Appl. Environ. Microbiol. 57:419-25 (1991)).

The term "oleaginous yeast" refers to those microorganisms classified as yeasts that can accumulate at least 25% of their dry cell weight as oil. Examples of oleaginous yeast include, but are no means limited to, the following genera: Yarrowia, Candida, Rhodotorula, Rhodosporidium, Cryptococcus, Trichosporon and Lipomyces.

The term "fermentable carbon source" means a carbon source that a microorganism will metabolize to derive energy. Typical carbon sources of the invention include, but are not limited to: monosaccharides, oligosaccharides, polysaccharides, alkanes, fatty acids, esters of fatty acids, monoglycerides, diglycerides, triglycerides, carbon dioxide, methanol, formaldehyde, formate and carbon-containing amines.

As used herein, an "isolated nucleic acid fragment" is a polymer of RNA or DNA that is single- or double-stranded, optionally containing synthetic, non-natural or altered nucleotide bases. An isolated nucleic acid fragment in the form of a polymer of DNA may be comprised of one or more segments of cDNA, genomic DNA or synthetic DNA.

A "substantial portion" of an amino acid or nucleotide sequence is that portion comprising enough of the amino acid sequence of a polypeptide or the nucleotide sequence of a gene to putatively identify that polypeptide or gene, either by manual evaluation of the sequence by one skilled in the art, or by computer-automated sequence comparison and identification using algorithms such as BLAST (Basic Local Alignment Search Tool; Altschul, S. F., et al., J. Mol. Biol. 215:403-410 (1993). In general, a sequence of ten or more contiguous amino acids or thirty or more nucleotides is necessary in order to identify putatively a polypeptide or nucleic acid sequence as homologous to a known protein or gene. Moreover, with respect to nucleotide sequences, gene-specific oligonucleotide probes comprising 20-30 contiguous nucleotides may be used in sequence-dependent methods of gene identification (e.g., Southern hybridization) and isolation (e.g., in situ hybridization of bacterial colonies or bacteriophage plaques). In addition, short oligonucleotides of 12-15 bases may be used as amplification primers in PCR in order to obtain a particular nucleic acid fragment comprising the primers. Accordingly, a "substantial portion" of a nucleotide sequence comprises enough of the sequence to specifically identify and/or isolate a nucleic acid fragment comprising the sequence. The term "complementary" is used to describe the relationship between nucleotide bases that are capable of hybridizing to one another. For example, with respect to DNA, adenosine is complementary to thymine and cytosine is complementary to guanine.

"Codon degeneracy" refers to the nature in the genetic code permitting variation of the nucleotide sequence without affecting the amino acid sequence of an encoded polypeptide. The skilled artisan is well aware of the "codon-bias" exhibited by a specific host cell in usage of nucleotide codons to specify a given amino acid. Therefore, when synthesizing a gene for improved expression in a host cell, it is desirable to design the gene such that its frequency of codon usage approaches the frequency of preferred codon usage of the host cell.

"Chemically synthesized", as related to a sequence of DNA, means that the component nucleotides were assembled in vitro. Manual chemical synthesis of DNA may be accomplished using well-established procedures; or, automated chemical synthesis can be performed using one of a number of commercially available machines. "Synthetic genes" can be assembled from oligonucleotide building blocks that are chemically synthesized using procedures known to those skilled in the art. These building blocks are ligated and annealed to form gene segments that are then enzymatically assembled to construct the entire gene. Accordingly, the genes can be tailored for optimal gene expression based on optimization of nucleotide sequence to reflect the codon bias of the host cell. The skilled artisan appreciates the likelihood of successful gene expression if codon usage is biased towards those codons favored by the host. Determination of preferred codons can be based on a survey of genes derived from the host cell, where sequence information is available.

"Gene" refers to a nucleic acid fragment that expresses a specific protein, including regulatory sequences preceding (5' non-coding sequences) and following (3' non-coding sequences) the coding sequence. "Native gene" refers to a gene as found in nature with its own regulatory sequences. "Chimeric gene" refers to any gene that is not a native gene, comprising regulatory and coding sequences that are not found together in nature. Accordingly, a chimeric gene may comprise regulatory sequences and coding sequences that are derived from different sources, or regulatory sequences and coding sequences derived from the same source, but arranged in a manner different than that found in nature. "Endogenous gene" refers to a native gene in its natural location in the genome of an organism. A "foreign" gene refers to a gene not normally found in the host organism, but that is introduced into the host organism by gene transfer. Foreign genes can comprise native genes inserted into a non-native organism, or chimeric genes. A "transgene" is a gene that has been introduced into the genome by a transformation procedure. A "codon-optimized gene" is a gene having its frequency of codon usage designed to mimic the frequency of preferred codon usage of the host cell.

"Coding sequence" refers to a DNA sequence that codes for a specific amino acid sequence. "Suitable regulatory sequences" refer to nucleotide sequences located upstream (5' non-coding sequences), within, or downstream (3' non-coding sequences) of a coding sequence, and which influence the transcription, RNA processing or stability, or translation of the associated coding sequence. Regulatory sequences may include promoters, translation leader sequences, introns, polyadenylation recognition sequences, RNA processing sites, effector binding sites and stem-loop structures.

"Promoter" refers to a DNA sequence capable of controlling the expression of a coding sequence or functional RNA. In general, a coding sequence is located 3' to a promoter sequence. Promoters may be derived in their entirety from a native gene, or be composed of different elements derived from different promoters found in nature, or even comprise synthetic DNA segments. It is understood by those skilled in the art that different promoters may direct the expression of a gene in different tissues or cell types, or at different stages of development, or in response to different environmental or physiological conditions. Promoters that cause a gene to be expressed in most cell types at most times are commonly referred to as "constitutive promoters". It is further recognized that since in most cases the exact boundaries of regulatory sequences have not been completely defined, DNA fragments of different lengths may have identical promoter activity. The term "3' non-coding sequences" or "transcription terminator" refers to DNA sequences located downstream of a coding sequence. This includes polyadenylation recognition sequences and other sequences encoding regulatory signals 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 3' region can influence the transcription, RNA processing or stability, or translation of the associated coding 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 post-transcriptional processing of the primary transcript and is referred to as the mature RNA. "Messenger RNA" or "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 and so can be translated into protein by the cell. "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 (U.S. 5,107,065; WO 99/28508). 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, or the coding sequence. "Functional RNA" refers to antisense RNA, ribozyme RNA, or other RNA that is not translated and yet has an effect on cellular processes.

The term "operably linked" refers to the association of nucleic acid sequences on a single nucleic acid fragment so that the function of one is affected by the other. For example, a promoter is operably linked with a coding sequence when it is capable of affecting the expression of that coding sequence (i.e., the coding sequence is under the transcriptional control of the promoter). Coding sequences can be operably linked to regulatory sequences in sense or antisense orientation.

The term "expression", as used herein, refers to the transcription and stable accumulation of sense (mRNA) or antisense RNA derived from the nucleic acid fragments of the invention. Expression may also refer to translation of mRNA into a polypeptide. "Transformation" refers to the transfer of a nucleic acid molecule into a host organism, resulting in genetically stable inheritance. The nucleic acid molecule may be a plasmid that replicates autonomously, for example; or, it may integrate into the genome of the host organism. Host organisms containing the transformed nucleic acid fragments are referred to as "transgenic" or "recombinant" or "transformed" organisms.

The terms "plasmid", "vector" and "cassette" refer to an extra chromosomal element often carrying genes that are not part of the central metabolism of the cell, and usually in the form of circular double-stranded DNA fragments. Such elements may be autonomously replicating sequences, genome integrating sequences, phage or nucleotide sequences, linear or circular, of a single- or double-stranded DNA or RNA, derived from any source, in which a number of nucleotide sequences have been joined or recombined into a unique construction which is capable of introducing a promoter fragment and DNA sequence for a selected gene product along with appropriate 3' untranslated sequence into a cell. "Transformation cassette" refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that facilitates transformation of a particular host cell. "Expression cassette" refers to a specific vector containing a foreign gene and having elements in addition to the foreign gene that allow for enhanced expression of that gene in a foreign host.

The term "sequence analysis software" refers to any computer algorithm or software program that is useful for the analysis of nucleotide or amino acid sequences. "Sequence analysis software" may be commercially available or independently developed. Typical sequence analysis software will include, but is not limited to: 1.) the GCG suite of programs (Wisconsin Package Version 9.0, Genetics Computer Group (GCG), Madison, WI); 2.) BLASTP, BLASTN, BLASTX (Altschul et al., d. Mol. Biol. 215:403-410 (1990)); 3.) DNASTAR (DNASTAR, Inc.

Madison, WI); 4.) Sequencher (Gene Codes Corporation, Ann Arbor, Ml); and 5.) the FASTA program incorporating the Smith-Waterman algorithm (W. R. Pearson, Comput. Methods Genome Res., [Proc. Int. Symp.] (1994), Meeting Date 1992, 111-20. Editor(s): Suhai, Sandor. Plenum: New York, NY). Within the context of this application it will be understood that where sequence analysis software is used for analysis, that the results of the analysis will be based on the "default values" of the program referenced, unless otherwise specified. As used herein "default values" will mean any set of values or parameters that originally load with the software when first initialized.

Standard recombinant DNA and molecular cloning techniques used herein are well known in the art and are described by Sambrook, J., Fritsch, E. F. and Maniatis, T., Molecular Cloning: A Laboratory Manual. 2nd ed., Cold Spring Harbor Laboratory: Cold Spring Harbor, NY (1989) (hereinafter "Maniatis"); by Silhavy, T. J., Bennan, M. L. and Enquist, L. W., Experiments with Gene Fusions. Cold Spring Harbor Laboratory: Cold Spring Harbor, NY (1984); and by Ausubel, F. M. et al., Current Protocols in Molecular Biology, published by Greene Publishing Assoc. and Wiley-lnterscience (1987). Microbial Biosynthesis Of Fatty Acids

In general, lipid accumulation in oleaginous microorganisms is triggered in response to the overall carbon to nitrogen ratio present in the growth medium (Figure 1). When cells have exhausted available nitrogen supplies (e.g., when the carbon to nitrogen ratio is greater than about 40), the depletion of cellular adenosine monophosphate (AMP) leads to the cessation of AMP-dependent isocitrate dehydrogenase activity in the mitochondria and the accumulation of citrate, transport of citrate into the cytosol and subsequent cleavage of the citrate by ATP-citrate lyase to yield acetyl-CoA and oxaloacetate. Acetyl-CoA is the principle building block for de novo biosynthesis of fatty acids. Although any compound that can effectively be metabolized to produce acetyl-CoA can serve as a precursor of fatty acids, glucose is the primary source of carbon in this type of reaction (Figure 1). Glucose is converted to pyruvate via glycolysis and pyruvate is then transported into the mitochondria where it can be converted to acetyl-CoA by pyruvate dehydrogenase ("PD"). Since acetyl- CoA cannot be transported directly across the mitochondrial membrane into the cytoplasm, the two carbons from acetyl-CoA condense with oxaloacetate to yield citrate (catalyzed by citrate synthase). Citrate is transported directly into the cytoplasm, where it is cleaved by ATP-citrate lyase to regenerate acetyl-CoA and oxaloacetate. The oxaloacetate reenters the tricarboxylic acid cycle, via conversion to malate. The synthesis of malonyl-CoA is the first committed step of fatty acid biosynthesis, which takes place in the cytoplasm. Malonyl-CoA is produced via carboxylation of acetyl-CoA by acetyl-CoA carboxylase ("ACC"). Fatty acid synthesis is catalyzed by a multi-enzyme fatty acid synthase complex ("FAS") and occurs by the condensation of eight two- carbon fragments (acetyl groups from acetyl-CoA) to form a 16-carbon saturated fatty acid, palmitate. More specifically, FAS catalyzes a series of 7 reactions, which involve the following (Smith, S. FASEB d., 8(15): 1248-59 (1994)):

1. Acetyl-CoA and malonyl-CoA are transferred to the acyl carrier peptide (ACP) of FAS. The acetyl group is then transferred to the malonyl group, forming β-ketobutyryl-ACP and releasing C0₂. 2. The β-ketobutyryl-ACP undergoes reduction (via β-ketoacyl reductase) and dehydration (via β-hydroxyacyl dehydratase) to form a trans-monounsaturated fatty acyl group.

3. The double bond is reduced by NADPH, yielding a saturated fatty- acyl group two carbons longer than the initial one. The butyryl- group's ability to condense with a new malonyl group and repeat the elongation process is then regenerated.

4. When the fatty acyl group becomes 16 carbons long, a thioesterase activity hydrolyses it, releasing free palmitate.

Palmitate (16:0) is the precursor of longer chain saturated and unsaturated fatty acids (e.g., stearic (18:0), palmitoleic (16:1) and oleic (18:1) acids) through the action of elongases and desaturases present in the endoplasmic reticulum membrane. Palmitate and stearate are converted to their unsaturated derivatives, palmitoleic (16:1) and oleic (18:1) acids, respectively, by the action of a Δ9 desaturase. Triacylglycerols (the primary storage unit for fatty acids) are formed by the esterification of two molecules of acyl-CoA to glycerol-3-phosphate to yield ,2-diacylglycerol phosphate (commonly identified as phosphatidic acid) (Figure 1). The phosphate is then removed, by phosphatidic acid phosphatase, to yield 1 ,2-diacylglycerol. Triacylglycerol is formed upon the addition of a third fatty acid by the action of a diacylglycerol-acyl transferase. Biosynthesis Of Omega-3 And Omega-6 Fatty Acids

Simplistically, the metabolic process that converts LA to GLA, DGLA and ARA (the ω-6 pathway) and ALA to STA, ETA, EPA and DHA (the ω-3 pathway) involves elongation of the carbon chain through the addition of carbon atoms and desaturation of the molecule through the addition of double bonds (Figure 2). This requires a series of special desaturation and elongation enzymes present in the endoplasmic reticulim membrane. ω-6 Fatty Acids

Oleic acid is converted to LA (18:2), the first of the ω-6 fatty acids, by the action of a Δ12 desaturase. Subsequent ω-6 fatty acids are produced as follows: 1.) LA is converted to GLA by the activity of a Δ6 desaturase; 2.) GLA is converted to DGLA by the action of an elongase; and 3.) DGLA is converted to ARA by the action of a Δ5 desaturase. ω-3 Fatty Acids Linoleic acid (LA) is converted to ALA, the first of the ω-3 fatty acids, by the action of a Δ15 desaturase. Subsequent ω-3 fatty acids are produced in a series of steps similar to that for the ω-6 fatty acids. Specifically: 1.) ALA is converted to STA by the activity of a Δ6 desaturase; 2.) STA is converted to ETA by the activity of an elongase; and 3.) ETA is converted to EPA by the activity of a Δ5 desaturase. Alternatively, ETA and EPA can be produced from DGLA and ARA, respectively, by the activity of a Δ17 desaturase. EPA can be further converted to DHA by the activity of an elongase and a Δ4 desaturase. Genes Involved In Omega Fatty Acid Production Many microorganisms, including algae, bacteria, molds and yeasts, can synthesize PUFAs and omega fatty acids in the ordinary course of cellular metabolism. Particularly well-studied are fungi including Schizochythum aggregatm, species of the genus Thraustochytrium and Morteriella alpina. Additionally, many dinoflagellates (Dinophyceaae) naturally produce high concentrations of PUFAs. As such, a variety of genes involved in oil production have been identified through genetic means and the DNA sequences of some of these genes are publicly available (non-limiting examples are shown below in Table 2):

Table 2 Some Publicly Available Genes Involved In PUFA Production

Additionally, the patent literature provides many additional DNA sequences of genes (and/or details concerning several of the genes above and their methods of isolation) involved in PUFA production. See, for example: U.S. 5,968,809 (Δ6 desaturases); U.S. 5,972,664 and

U.S. 6,075,183 (Δ5 desaturases); WO 91/13972 and U.S. 5,057,419 (Δ9 desaturases); WO 93/11245 (Δ15 desaturases); WO 94/11516, U.S. 5,443,974 and WO 03/099216 (Δ12 desaturases); U.S. 2003/0196217 A1 (Δ17 desaturase); WO 02/090493 (Δ4 desaturases); and, WO 00/12720 and U.S. 2002/0139974A1 (elongases). Each of these patents and applications are herein incorporated by reference in their entirety.

As will be obvious to one skilled in the art, the particular functionalities required to be introduced into a host organism for production of a particular PUFA final product will depend on the host cell (and its native PUFA profile and/or desaturase profile), the availability of substrate and the desired end product(s). As shown in Figure 2, LA, GLA, DGLA, ARA, ALA, STA, ETA, EPA, DPA and DHA may all be produced in oleaginous yeasts, by introducing various combinations of the following PUFA enzyme functionalities: a Δ4 desaturase, a Δ5 desaturase, a Δ6 desaturase, a Δ12 desaturase, a Δ15 desaturase, a Δ17 desaturase, a Δ9 desaturase and/or an elongase. One skilled in the art will be able to identify various candidate genes encoding each of the above enzymes, according to publicly available literature (e.g., GenBank), the patent literature, and experimental analysis of microorganisms having the ability to produce PUFAs. The sequences may be derived from any source, e.g., isolated from a natural source (from bacteria, algae, fungi, plants, animals, etc.), produced via a semi-synthetic route or synthesized de novo. In some embodiments, manipulation of genes endogenous to the host is preferred; for other purposes, it is necessary to introduce heterologous genes. Although the particular source of the desaturase and elongase genes introduced into the host is not critical to the invention, considerations for choosing a specific polypeptide having desaturase or elongase activity include: 1.) the substrate specificity of the polypeptide; 2.) whether the polypeptide or a component thereof is a rate-limiting enzyme; 3.) whether the desaturase or elongase is essential for synthesis of a desired PUFA; and/or 4.) co-factors required by the polypeptide. The expressed polypeptide preferably has parameters compatible with the biochemical environment of its location in the host cell. For example, the polypeptide may have to compete for substrate with other enzymes in the host cell. Analyses of the K^ and specific activity of the polypeptide are therefore considered in determining the suitability of a given polypeptide for modifying PUFA production in a given host cell. The polypeptide used in a particular host cell is one that can function under the biochemical conditions present in the intended host cell but otherwise can be any polypeptide having desaturase or elongase activity capable of modifying the desired PUFA.

Endogenous PUFA Genes

In some cases, the host organism in which it is desirable to produce PUFAs will possess endogenous genes encoding some PUFA biosynthetic pathway enzymes. For example, oleaginous yeast can typically produce 18:2 fatty acids (and some have the additional capability of synthesizing 18:3 fatty acids); thus, oleaginous yeast typically possess native Δ12 desaturase activity and may also have Δ15 desaturases. In some embodiments, therefore, expression of the native desaturase enzyme is preferred over a heterologous (or "foreign") enzyme since: 1.) the native enzyme is optimized for interaction with other enzymes and proteins within the cell; and 2.) heterologous genes are unlikely to share the same codon preference in the host organism. Additionally, advantages are incurred when the sequence of the native gene is known, as it permits facile disruption of the endogenous gene by targeted disruption. Heterologous PUFA Genes

In many instances, the appropriate desaturases and elongases are not present in the host organism of choice to enable production of the desired PUFA products. Thus, it is necessary to introduce heterologous genes. For the purposes of the present invention herein, it was desirable to demonstrate an example of the introduction of an ω-3 and/or ω-6 biosynthetic pathway into an oleaginous host organism; and thus, a Mortierella alpina Δ5 desaturase, M. alpina Δ6 desaturase, Saprolegnia diclina Δ17 desaturase and M. alpina elongase were introduced into Yarrowia lipolytica. However, the specific enzymes (and genes encoding those enzymes) introduced into the host organism and the specific PUFAs produced are by no means limiting to the invention herein.

If one desired to produce EPA, as demonstrated herein, it will be obvious to one skilled in the art that numerous other genes derived from different sources would be suitable to introduce Δ5 desaturase, Δ6 desaturase, Δ17 desaturase and elongase activity into the preferred microbial host. Thus, in one embodiment of the present invention, other DNAs which are substantially identical to the M. alpina Δ6 desaturase, Δ5 desaturase and high-affinity PUFA elongase and the S. diclina Δ17 desaturase also can be used for production of ω-6 and/or ω-3 fatty acids (e.g., EPA) in oleaginous yeast. By "substantially identical" is intended an amino acid sequence or nucleic acid sequence exhibiting in order of increasing preference at least 80%, 90% or 95% homology to the selected polypeptides, or nucleic acid sequences encoding the amino acid sequence. For polypeptides, the length of comparison sequences generally is at least 16 amino acids, preferably at least 20 amino acids or most preferably 35 amino acids. For nucleic acids, the length of comparison sequences generally is at least 50 nucleotides, preferably at least 60 nucleotides, more preferably at least 75 nucleotides, and most preferably 110 nucleotides.

Homology typically is measured using sequence analysis software, wherein the term "sequence analysis software" refers to any computer algorithm or software program that is useful for the analysis of nucleotide or amino acid sequences. "Sequence analysis software" may be commercially available or independently developed. Typical sequence analysis software will include, but is not limited to: 1.) the GCG suite of programs (Wisconsin Package Version 9.0, Genetics Computer Group (GCG), Madison, WI); 2.) BLASTP, BLASTN, BLASTX (Altschul et al., d. Mol. Biol. 215:403-410 (1990)); 3.) DNASTAR (DNASTAR, Inc., Madison, WI); and 4.) the FASTA program incorporating the Smith- Waterman algorithm (W. R. Pearson, Comput. Methods Genome Res., [Proc. Int. Symp.] (1994), Meeting Date 1992, 111-20. Suhai, Sandor, Ed. Plenum: New York, NY). Within the context of this application it will be understood that where sequence analysis software is used for analysis, the results of the analysis will be based on the "default values" of the program referenced, unless otherwise specified. As used herein "default values" will mean any set of values or parameters that originally load with the software when first initialized. In general, such computer software matches similar sequences by assigning degrees of homology to various substitutions, deletions, and other modifications.

Additionally it will be appreciated by one of skill in the art that polypeptides may have amino acids conservatively substituted in a manner such that the function of the polypeptide is not altered or compromised. Polypeptides having the desaturase and elongase activities as described herein and possessing such conservative substitutions are considered within the scope of the present invention. Conservative substitutions typically include substitutions within the following groups: 1.) glycine and alanine; 2.) valine, isoleucine and leucine; 3.) aspartic acid, glutamic acid, asparagine and glutamine; 4.) serine and threonine; 5.) lysine and arginine; and 6.) phenylalanine and tyrosine. Substitutions may also be made on the basis of conserved hydrophobicity or hydrophilicity (Kyte and Doolittle, d. Mol. Biol.

157:105-132 (1982)), or on the basis of the ability to assume similar polypeptide secondary structure (Chou and Fasman, Adv. Enzymol. 47:45-148 (1978)).

In alternate embodiments of the invention, other DNAs which, although not substantially identical to the M. alpina Δ6 desaturase, Δ5 desaturase and high-affinity PUFA elongase and the S. diclina Δ17 desaturase, also can be used for the purposes herein (e.g., for production of ω-3 and/or ω-6 PUFAs such as ARA and EPA). For example, DNA sequences encoding Δ6 desaturase polypeptides that would be useful for introduction into an oleaginous yeast according to the teachings of the present invention may be obtained from microorganisms having an ability to produce GLA or STA. Such microorganisms include, for example, those belonging to the genera Mortierella, Conidiobolus, Pythium, Phytophathora, Penicillium, Porphyhdium, Coidosporium, Mucor, Fusarium, Aspergillus, Rhodotorula and Entomophthora. Within the genus Porphyridium, of particular interest is P. cruentum. Within the genus

Mortierella, of particular interest are M. elongata, M. exigua, M. hygrophila, M. ramanniana var. angulispora and M. alpina. Within the genus Mucor, of particular interest are M. circinelloides and M. javanicus.

Alternatively, a related desaturase that is not substantially identical to the M. alpina Δ6 desaturase, but which can desaturate a fatty acid molecule at carbon 6 from the carboxyl end of the molecule would also useful in the present invention as a Δ6 desaturase, assuming the desaturase can still effectively convert LA to GLA and/or ALA to STA. As such, related desaturases and elongases can be identified by their ability of function substantially the same as the desaturases and elongase disclosed herein.

In summary, genes encoding PUFA biosynthetic pathway enzymes suitable for the purposes herein may be isolated from a variety of sources. Desaturases for the purposes herein are characterized by the ability to: 1.) desaturate a fatty acid between the 17^th and 18^th carbon atom numbered from the carboxyl-terminal end of the molecule and catalyze the conversion of ARA to EPA and DGLA to ETA (Δ17 desaturases); 2.) catalyze the conversion of LA to GLA and/or ALA to STA (Δ6 desaturases); 3.) catalyze the conversion of DGLA to ARA and/or ETA to EPA (Δ5 desaturases); 4.) catalyze the conversion of DPA to DHA (Δ4 desaturases); 5.) catalyze the conversion of oleic acid to LA (Δ12 desaturases); 6.) catalyze the conversion of LA to ALA (Δ15 desaturases); and/or 7.) catalyze the conversion of palmitate to palmitoleic acid and/or stearate to oleic acid (Δ9 desaturases). In like manner, suitable elongases for the purposes herein are not limited to those from a specific source; instead, the enzymes having use for the purposes herein are characterized by their ability to elongate a fatty acid carbon chain by 2 carbons relative to the substrate the elongase acts upon, to thereby produce a mono- or polyunsaturated fatty acid. Optimization Of Omega Fatty Acid Genes For Expression In Particular Organisms Although the particular source of a PUFA desaturase or elongase is not critical in the invention herein, it will be obvious to one of skill in the art that heterologous genes will be expressed with variable efficiencies in an alternate host. Thus, ω-3 and/or ω-6 PUFA production may be optimized by selection of a particular desaturase or elongase whose level of expression in a heterologous host is preferred relative to the expression of an alternate desaturase or elongase in the host organism of interest. Furthermore, it may be desirable to modify the expression of particular PUFA biosynthetic pathway enzymes to achieve optimal conversion efficiency of each, according to the specific PUFA product composition of interest. A variety of genetic engineering techniques are available to optimize expression of a particular enzyme. Two such techniques include codon optimization and gene mutation, as described below. Genes produced by. e.g., either of these two methods, having desaturase and/or elongase activity(s) would be useful in the invention herein for synthesis of ω-3 and/or ω-6 PUFAs.

Codon Optimization

As will be appreciated by one skilled in the art, it is frequently useful to modify a portion of the codons encoding a particular polypeptide that is to be expressed in a foreign host, such that the modified polypeptide uses codons that are preferred by the alternate host. Use of host-preferred codons can substantially enhance the expression of the foreign gene encoding the polypeptide.

In general, host-preferred codons can be determined within a particular host species of interest by examining codon usage in proteins (preferably those expressed in the largest amount) and determining which codons are used with highest frequency. Then, the coding sequence for a polypeptide of interest having desaturase or elongase activity can be synthesized in whole or in part using the codons preferred in the host species. All (or portions) of the DNA also can be synthesized to remove any destabilizing sequences or regions of secondary structure that would be present in the transcribed mRNA. All (or portions) of the DNA also can be synthesized to alter the base composition to one more preferable in the desired host cell.

In the present invention, it was desirable to modify a portion of the codons encoding the polypeptide having Δ17 desaturase activity, to enhance the expression of the gene \r\Yarrowia lipolytica. The nucleic acid sequence of the native gene (e.g., the Saprolegnia diclina Δ17 desaturase) was modified to employ host-preferred codons. The skilled artisan will appreciate that this optimization method will be equally applicable to other genes in the ω-3/ω-6 fatty acids biosynthetic pathway (see for example, co-pending U.S. Provisional Application No. 60/468718, herein incorporated entirely by reference). Furthermore, modulation of the S. diclina Δ17 desaturase is only exemplary. Gene Mutation

Methods for synthesizing sequences and bringing sequences together are well established in the literature. For example, in vitro mutagenesis and selection, site-directed mutagenesis, error prone PCR (Melnikov et al., Nucleic Acids Research, 27(4): 1056-1062 (February 15, 1999)), "gene shuffling" or other means can be employed to obtain mutations of naturally occurring desaturase or elongase genes. This would permit production of a polypeptide having desaturase or elongase activity, respectively, in vivo with more desirable physical and kinetic parameters for function in the host cell such as a longer half-life or a higher rate of production of a desired PUFA.

If desired, the regions of a polypeptide of interest (i.e., a desaturase or an elongase) important for enzymatic activity can be determined through routine mutagenesis, expression of the resulting mutant polypeptides and determination of their activities. Mutants may include deletions, insertions and point mutations, or combinations thereof. A typical functional analysis begins with deletion mutagenesis to determine the N- and C-terminal limits of the protein necessary for function, and then internal deletions, insertions or point mutants are made to further determine regions necessary for function. Other techniques such as cassette mutagenesis or total synthesis also can be used. Deletion mutagenesis is accomplished, for example, by using exonucleases to sequentially remove the 5' or 3' coding regions. Kits are available for such techniques. After deletion, the coding region is completed by ligating oligonucleotides containing start or stop codons to the deleted coding region after the 5' or 3' deletion, respectively. Alternatively, oligonucleotides encoding start or stop codons are inserted into the coding region by a variety of methods including site-directed mutagenesis, mutagenic PCR or by ligation onto DNA digested at existing restriction sites. Internal deletions can similarly be made through a variety of methods including the use of existing restriction sites in the DNA, by use of mutagenic primers via site-directed mutagenesis or mutagenic PCR. Insertions are made through methods such as linker-scanning mutagenesis, site-directed mutagenesis or mutagenic PCR, while point mutations are made through techniques such as site-directed mutagenesis or mutagenic PCR.

Chemical mutagenesis also can be used for identifying regions of a desaturase or elongase polypeptide important for activity. A mutated construct is expressed, and the ability of the resulting altered protein to function as a desaturase or elongase is assayed. Such structure-function analysis can determine which regions may be deleted, which regions tolerate insertions, and which point mutations allow the mutant protein to function in substantially the same way as the native desaturase or native elongase. All such mutant proteins and nucleotide sequences encoding them that are derived from the codon-optimized gene described herein are within the scope of the present invention.

In the present invention, it was desirable to modify a portion of the codons encoding the polypeptide having Δ17 desaturase activity, to enhance the expression of the gene in Yarrowia lipolytica. The nucleic acid sequence of the native gene (e.g., the S. diclina Δ17 desaturase) was modified to employ host preferred codons. The skilled artisan will appreciate that these optimization methods will be equally applicable to other genes in the ω-3/ω-6 fatty acids biosynthetic pathway and that modulation of the S. diclina Δ17 desaturase and M. alpina Δ6 desaturase, Δ5 desaturase are only exemplary. Microbial Production Of ω-3 And/Or ω-6 Fattv Acids

Microbial production of ω-3 and/or ω-6 fatty acids has several advantages over purification from natural sources such as fish or plants. For example: 1.) Many microbes are known with greatly simplified oil compositions compared with those of higher organisms, making purification of desired components easier; 2.) Microbial production is not subject to fluctuations caused by external variables, such as weather and food supply; 3.) Microbially produced oil is substantially free of contamination by environmental pollutants; 4.) Microbes can provide PUFAs in particular forms which may have specific uses; and 5.) Microbial oil production can be manipulated by controlling culture conditions, notably by providing particular substrates for microbially expressed enzymes, or by addition of compounds/genetic engineering to suppress undesired biochemical pathways. In addition to these advantages, production of ω-3 and/or ω-6 fatty acids from recombinant microbes provides the ability to alter the naturally occurring microbial fatty acid profile by providing new biosynthetic pathways in the host or by suppressing undesired pathways, thereby increasing levels of desired PUFAs, or conjugated forms thereof, and decreasing levels of undesired PUFAs. For example, it is possible to modify the ratio of ω-3 to ω-6 fatty acids so produced, produce either ω-3 or ω-6 fatty acids exclusively while eliminating production of the alternate omega fatty acid, or engineer production of a specific PUFA without significant accumulation of other PUFA downstream or upstream products. Expression Systems, Cassettes And Vectors The genes and gene products described herein may be produced in heterologous microbial host cells, particularly in the cells of oleaginous yeasts (e.g., Yarrowia lipolytica). Expression in recombinant microbial hosts may be useful for the production of various PUFA pathway intermediates, or for the modulation of PUFA pathways already existing in the host for the synthesis of new products heretofore not possible using the host.

Microbial expression systems and expression vectors containing regulatory sequences that direct high level expression of foreign proteins are well known to those skilled in the art. Any of these could be used to construct chimeric genes for production of any of the gene products of the preferred desaturase and/or elongase sequences. These chimeric genes could then be introduced into appropriate microorganisms via transformation to provide high-level expression of the encoded enzymes. Accordingly, it is expected that introduction of chimeric genes encoding a PUFA biosynthetic pathway (e.g., the Δ5 desaturase, Δ6 desaturase, Δ17 desaturase and elongase described herein), under the control of the appropriate promoters will result in increased production of ω-3 and/or ω-6 fatty acids. It is contemplated that it will be useful to express various combinations of these PUFA desaturase and elongase genes together in a host microorganism. It will be obvious to one skilled in the art that the particular genes included within a particular expression cassette(s) will depend on the host cell, its ability to synthesize PUFAs using native desaturases and elongases, the availability of substrate and the desired end product(s). For example, it may be desirable for an expression cassette to be constructed comprising genes encoding one or more of the following enzymatic activities: a Δ4 desaturase, a Δ5 desaturase, a Δ6 desaturase, a Δ12 desaturase, a Δ15 desaturase, a Δ17 desaturase, a Δ9 desaturase and/or an elongase. As such, the present invention encompasses a method of producing PUFAs comprising exposing a fatty acid substrate to the PUFA enzyme(s) described herein, such that the substrate is converted to the desired fatty acid product. Thus, each PUFA gene and corresponding enzyme product described herein (e.g., a wildtype, codon-optimized, synthetic and/or mutant enzyme having appropriate desaturase or elongase activity) can be used directly or indirectly for the production of PUFAs. Direct production of PUFAs occurs wherein the fatty acid substrate is converted directly into the desired fatty acid product without any intermediate steps or pathway intermediates. For example, production of ARA would occur in a host cell which produces or which is provided DLGA, by adding or introducing into said cell an expression cassette that provides Δ5 desaturase activity.

In contrast, multiple genes encoding the PUFA biosynthetic pathway may be used in combination, such that a series of reactions occur to produce a desired PUFA. For example, expression cassette(s) encoding elongase, Δ5 desaturase, Δ17 desaturase and Δ4 desaturase activity would enable a host cell that naturally produces GLA, to instead produce DHA (such that GLA is converted to DGLA by an elongase; DGLA may then be converted to ARA by a Δ5 desaturase; ARA is then converted to EPA by a Δ17 desaturase, which may in turn be converted to DPA by an elongase; and DPA would be converted to DHA by a Δ4 desaturase). In a preferred embodiment, wherein the host cell is an oleaginous yeast, expression cassettes encoding each of the enzymes necessary for PUFA biosynthesis will need to be introduced into the organism, since naturally produced PUFAs in these organisms are limited to 18:2 fatty acids (i.e., LA), and less commonly, 18:3 fatty acids (i.e., ALA). Alternatively, substrate feeding may be required.

Vectors or DNA cassettes useful for the transformation of suitable host cells are well known in the art. The specific choice of sequences present in the construct is dependent upon the desired expression products (supra), the nature of the host cell and the proposed means of separating transformed cells versus non-transformed cells. Typically, however, the vector or cassette contains sequences directing transcription and translation of the relevant gene(s), a selectable marker and sequences allowing autonomous replication or chromosomal integration. Suitable vectors comprise a region 5' of the gene that controls transcriptional initiation and a region 3' of the DNA fragment that controls transcriptional termination. It is most preferred when both control regions are derived from genes from the transformed host cell, although it is to be understood that such control regions need not be derived from the genes native to the specific species chosen as a production host. Initiation control regions or promoters which are useful to drive expression of desaturase and/or elongase ORFs in the desired host cell are numerous and familiar to those skilled in the art. Virtually any promoter capable of directing expression of these genes in the selected host cell is suitable for the present invention. Expression in a host cell can be accomplished in a transient or stable fashion. Transient expression can be accomplished by inducing the activity of a regulatable promoter operably linked to the gene of interest. Stable expression can be achieved by the use of a constitutive promoter operably linked to the gene of interest. As an example, when the host cell is yeast, transcriptional and translational regions functional in yeast cells are provided, particularly from the host species. The transcriptional initiation regulatory regions can be obtained, for example, from: 1.) genes in the glycolytic pathway, such as alcohol dehydrogenase, gIyceraldehyde-3-phosphate-dehydrogenase (see U.S. Patent Application Number 60/482263, incorporated herein by reference), phosphoglycerate mutase (see U.S. Patent Application Number 60/482263, incorporated herein by reference), fructose- bisphosphate aldolase (see U.S. Patent Application Number 60/519971 , incorporated herein by reference), phosphoglucose-isomerase, phosphoglycerate kinase, etc.; or, 2.) regulatable genes such as acid phosphatase, lactase, metallothionein, glucoamylase, the translation elongation factor EF1-α (TEF) protein (U.S. 6,265,185), ribosomal protein S7 (U.S. 6,265,185), etc. Any one of a number of regulatory sequences can be used, depending upon whether constitutive or induced transcription is desired, the efficiency of the promoter in expressing the ORF of interest, the ease of construction and the like.

Nucleotide sequences surrounding the translational initiation codon 'ATG' have been found to affect expression in yeast cells. If the desired polypeptide is poorly expressed in yeast, the nucleotide sequences of exogenous genes can be modified to include an efficient yeast translation initiation sequence to obtain optimal gene expression. For expression in yeast, this can be done by site-directed mutagenesis of an inefficiently expressed gene by fusing it in-frame to an endogenous yeast gene, preferably a highly expressed gene. Alternatively, as demonstrated in the invention herein in Yarrowia lipolytica, one can determine the consensus translation initiation sequence in the host and engineer this sequence into heterologous genes for their optimal expression in the host of interest. The termination region can be derived from the 3' region of the gene from which the initiation region was obtained or from a different gene. A large number of termination regions are known and function satisfactorily in a variety of hosts (when utilized both in the same and different genera and species from where they were derived). The termination region usually is selected more as a matter of convenience rather than because of any particular property. Preferably, the termination region is derived from a yeast gene, particularly Saccharomyces, Schizosaccharomyces, Candida, Yarrowia or Kluyveromyces. The 3'- regions of mammalian genes encoding γ-interferon and -2 interferon are also known to function in yeast. Termination control regions may also be derived from various genes native to the preferred hosts. Optionally, a termination site may be unnecessary; however, it is most preferred if included.

As one of skill in the art is aware, merely inserting a gene into a cloning vector does not ensure that it will be successfully expressed at the level needed. In response to the need for a high expression rate, many specialized expression vectors have been created by manipulating a number of different genetic elements that control aspects of transcription, translation, protein stability, oxygen limitation and secretion from the host cell. More specifically, some of the molecular features that have been manipulated to control gene expression include: 1.) the nature of the relevant transcriptional promoter and terminator sequences; 2.) the number of copies of the cloned gene and whether the gene is plasmid- borne or integrated into the genome of the host cell; 3.) the final cellular location of the synthesized foreign protein; 4.) the efficiency of translation in the host organism; 5.) the intrinsic stability of the cloned gene protein within the host cell; and 6.) the codon usage within the cloned gene, such that its frequency approaches the frequency of preferred codon usage of the host cell. Each of these types of modifications are encompassed in the present invention, as means to further optimize expression of the P U FA b iosy nthetic pathway enzymes .

Transformation Of Microbial Hosts

Once the DNA encoding a desaturase or elongase polypeptide suitable for expression in an oleaginous yeast has been obtained, it is placed in a plasmid vector capable of autonomous replication in a host cell or it is directly integrated into the genome of the host cell. Integration of expression cassettes can occur randomly within the host genome or can be targeted through the use of constructs containing regions of homology with the host genome sufficient to target recombination within the host locus. Where constructs are targeted to an endogenous locus, all or some of the transcriptional and translational regulatory regions can be provided by the endogenous locus.

Where two or more genes are expressed from separate replicating vectors, it is desirable that each vector has a different means of selection and should lack homology to the other constructs to maintain stable expression and prevent reassortment of elements among constructs. Judicious choice of regulatory regions, selection means and method of propagation of the introduced construct can be experimentally determined so that all introduced genes are expressed at the necessary levels to provide for synthesis of the desired products.

Constructs comprising the gene of interest may be introduced into a host cell by any standard technique. These techniques include transformation (e.g., lithium acetate transformation [Methods in

Enzymology, 194:186-187 (1991)]), protoplast fusion, holistic impact, electroporation, microinjection, or any other method that introduces the gene of interest into the host cell. More specific teachings applicable for oleaginous yeasts (i.e., Yarrowia lipolytica) include U.S. Patent Nos. 4,880,741 and 5,071 ,764 and Chen, D. C. et al. (Appl Microbiol Biotechnol. 48(2):232-235 (1997)).

For convenience, a host cell that has been manipulated by any method to take up a DNA sequence (e.g., an expression cassette) will be referred to as "transformed" or "recombinant" herein. The transformed host will have at least one copy of the expression construct and may have two or more, depending upon whether the gene is integrated into the genome, amplified or is present on an extrachromosomal element having multiple copy numbers. The transformed host cell can be identified by selection for a marker contained on the introduced construct. Alternatively, a separate marker construct may be co-transformed with the desired construct, as many transformation techniques introduce many DNA molecules into host cells. Typically, transformed hosts are selected for their ability to grow on selective media. Selective media may incorporate an antibiotic or lack a factor necessary for growth of the untransformed host, such as a nutrient or growth factor. An introduced marker gene may confer antibiotic resistance, or encode an essential growth factor or enzyme, thereby permitting growth on selective media when expressed in the transformed host. Selection of a transformed host can also occur when the expressed marker protein can be detected, either directly or indirectly. The marker protein may be expressed alone or as a fusion to another protein. The marker protein can be detected by: 1.) its enzymatic activity (e.g., β-galactosidase can convert the substrate X-gal [5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside] to a colored product; luciferase can convert luciferin to a light-emitting product); or 2.) its light- producing or modifying characteristics (e.g., the green fluorescent protein of Aequorea victoria fluoresces when illuminated with blue light). Alternatively, antibodies can be used to detect the marker protein or a molecular tag on, for example, a protein of interest. Cells expressing the marker protein or tag can be selected, for example, visually, or by techniques such as FACS or panning using antibodies. For selection of yeast transformants, any marker that functions in yeast may be used. Preferred for use herein are resistance to kanamycin, hygromycin and the amino glycoside G418, as well as ability to grow on media lacking uracil or leucine.

Following transformation, substrates suitable for the recombinantly expressed desaturases and/or elongases (and optionally other PUFA enzymes that are expressed within the host cell) may be produced by the host either naturally or transgenically, or they may be provided exogenously.

Metabolic Engineering Of ω-3 And/Or ω-6 Fatty Acid Biosynthesis In Microbes Methods for manipulating biochemical pathways are well known to those skilled in the art; and, it is expected that numerous manipulations will be possible to maximize ω-3 and/or ω-6 fatty acid biosynthesis in oleaginous yeasts, and particularly, in Yarrowia lipolytica. This may require metabolic engineering directly within the PUFA biosynthetic pathway or additional manipulation of pathways that contribute carbon to the PUFA biosynthetic pathway.

In the case of manipulations within the PUFA biosynthetic pathway, it may be desirable to increase the production of LA to enable increased production of ω-6 and/or ω-3 fatty acids. This may be accomplished by introducing and/or amplifying genes encoding Δ9 and/or Δ12 desaturases. To maximize production of ω-6 unsaturated fatty acids, such as ARA, it is well known to one skilled in the art that production is favored in a host microorganism that is substantially free of ALA. Thus, preferably, the host is selected or obtained by removing or inhibiting Δ15 or ω-3 type desaturase activity that permits conversion of LA to ALA. The endogenous desaturase activity can be reduced or eliminated by, for example: 1.) providing a cassette for transcription of antisense sequences to the Δ15 desaturase transcription product; 2.) disrupting the Δ15 desaturase gene through insertion, substitution and/or deletion of all or part of the target gene; or 3.) using a host cell which naturally has [or has been mutated to have] low or no Δ15 desaturase activity. Inhibition of undesired desaturase pathways can also be accomplished through the use of specific desaturase inhibitors such as those described in U.S. 4,778,630.

Alternatively, it may be desirable to maximize production of ω-3 fatty acids (and minimize synthesis of ω-6 fatty acids). Thus, one could utilize a host microorganism wherein the Δ12 desaturase activity that permits conversion of oleic acid to LA is removed or inhibited, using any of the means described above (see also, for example, co-pending U.S. Provisional Application No. 60/484209, herein incorporated entirely by reference). Subsequently, appropriate expression cassettes would be introduced into the host, along with appropriate substrates (e.g., ALA) for conversion to ω-3 fatty acid derivatives of ALA (e.g., STA, ETA, EPA, DPA, DHA).

Beyond the immediate PUFA biosynthetic pathway, it is expected that manipulation of several other enzymatic pathways leading to the biosynthesis of precursor fatty acids may contribute to the overall net biosynthesis of specific PUFAs. Identification and manipulation of these related pathways will be useful in the future.

Techniques To Up-Regulate Desirable Biosynthetic Pathways Additional copies of desaturase and elongase genes may be introduced into the host to increase the output of ω-3 and/or ω-6 fatty acid biosynthetic pathways. Expression of the desaturase or elongase genes also can be increased at the transcriptional level through the use of a stronger promoter (either regulated or constitutive) to cause increased expression, by removing/deleting destabilizing sequences from either the mRNA or the encoded protein, or by adding stabilizing sequences to the mRNA (U.S. 4,910,141). Yet another approach to increase expression of the desaturase or elongase genes, as demonstrated in the instant invention, is to increase the translational efficiency of the encoded mRNAs by replacement of codons in the native gene with those for optimal gene expression in the selected host microorganism.

Techniques To Down-Regulate Undesirable Biosynthetic Pathways Conversely, biochemical pathways competing with the ω-3 and/or ω-6 fatty acid biosynthetic pathways for energy or carbon, or native PUFA biosynthetic pathway enzymes that interfere with production of a particular PUFA end-product, may be eliminated by gene disruption or down- regulated by other means (e.g., antisense mRNA). For gene disruption, a foreign DNA fragment (typically a selectable marker gene) is inserted into the structural gene to be disrupted in order to interrupt its coding sequence and thereby functionally inactivate the gene. Transformation of the disruption cassette into the host cell results in replacement of the functional native gene by homologous recombination with the non- functional disrupted gene (see, for example: Hamilton et al. J. Bacteriol. 171 :4617-4622 (1989); Balbas et al. Gene 136:211-213 (1993); Gueldener et al. Nucleic Acids Res. 24:2519-2524 (1996); and Smith et al. Methods Mol. Cell. Biol. 5:270-277(1996)).

Antisense technology is another method of down-regulating genes when the sequence of the target gene is known. To accomplish this, a nucleic acid segment from the desired gene is cloned and operably linked to a promoter such that the anti-sense strand of RNA will be transcribed. This construct is then introduced into the host cell and the antisense strand of RNA is produced. Antisense RNA inhibits gene expression by preventing the accumulation of mRNA that encodes the protein of interest. The person skilled in the art will know that special considerations are associated with the use of antisense technologies in order to reduce expression of particular genes. For example, the proper level of expression of antisense genes may require the use of different chimeric genes utilizing different regulatory elements known to the skilled artisan. Although targeted gene disruption and antisense technology offer effective means of down-regulating genes where the sequence is known, other less specific methodologies have been developed that are not sequence-based. For example, cells may be exposed to UV radiation and then screened for the desired phenotype. Mutagenesis with chemical agents is also effective for generating mutants and commonly used substances include chemicals that affect nonreplicating DNA (e.g., HN0₂ and NH₂OH), as well as agents that affect replicating DNA (e.g., acridine dyes, notable for causing frameshift mutations). Specific methods for creating mutants using radiation or chemical agents are well documented in the art. See, for example: Thomas D. Brock in Biotechnology: A Textbook of Industrial Microbiology, 2^nd ed. (1989) Sinauer Associates: Sunderland, MA; or Deshpande, Mukund V., Appl. Biochem. Biotechnol., 36:227 (1992). Another non-specific method of gene disruption is the use of transposable elements or transposons. Transposons are genetic elements that insert randomly into DNA but can be later retrieved on the basis of sequence to determine where the insertion has occurred. Both in vivo and in vitro transposition methods are known. Both methods involve the use of a transposable element in combination with a transposase enzyme. When the transposable element or transposon is contacted with a nucleic acid fragment in the presence of the transposase, the transposable element will randomly insert into the nucleic acid fragment. The technique is useful for random mutagenesis and for gene isolation, since the disrupted gene may be identified on the basis of the sequence of the transposable element. Kits for in vitro transposition are commercially available [see, for example: 1.) The Primer Island Transposition Kit, available from Perkin Elmer Applied Biosystems, Branchburg, NJ, based upon the yeast Ty1 element; 2.) The Genome Priming System, available from New England Biolabs, Beverly, MA, based upon the bacterial transposon Tn7; and 3.) the EZ::TN Transposon Insertion Systems, available from Epicentre Technologies, Madison, WI, based upon the Tn5 bacterial transposable element].

Within the context of the present invention, it may be useful to modulate the expression of the fatty acid biosynthetic pathway by any one of the methods described above. For example, the present invention provides methods whereby genes encoding key enzymes in the biosynthetic pathways are introduced into oleaginous yeasts for the production of ω-3 and/or ω-6 fatty acids. These genes encode one or more of the following: Δ6 desaturase, Δ5 desaturase, Δ12 desaturase, Δ15 desaturase, Δ4 desaturase, Δ17 desaturase, Δ9 desaturase and PUFA elongase. It will be particularly useful to express these genes in oleaginous yeasts that do not naturally possess ω-3 and/or ω-6 fatty acid biosynthetic pathways and coordinate the expression of these genes, to maximize production of preferred PUFA products using various means for metabolic engineering of the host organism. Preferred Microbial Hosts For Recombinant Production Of ω-3 And/Or ω-6 Fatty Acids

Host cells for production of omega fatty acids may include microbial hosts that grow on a variety of feedstocks, including simple or complex carbohydrates, organic acids and alcohols, and/or hydrocarbons over a wide range of temperature and pH values.

Preferred microbial hosts, however, are oleaginous yeasts. These organisms are naturally capable of oil synthesis and accumulation, wherein the oil can comprise greater than about 25% of the cellular dry weight, more preferably greater than about 30% of the cellular dry weight, and most preferably greater than about 40% of the cellular dry weight. Genera typically identified as oleaginous yeast include, but are not limited to: Yarrowia, Candida, Rhodotorula, Rhodospoήdium, Cryptococcus, Trichosporon and Lipomyces. More specifically, illustrative oil- synthesizing yeasts include: Rhodosporidium toruloides, Lipomyces starkeyii, L. lipoferus, Candida revkaufi, C. pulcherrima, C. tropicalis, C. utilis, Trichosporon pullans, T. cutaneum, Rhodotorula glutinus, R. graminis, and Yarrowia lipolytica (formerly classified as Candida lipolytica).

Most preferred is the oleaginous yeast Yarrowia lipolytica; and, in a further embodiment, most preferred are the Y. lipolytica strains designated as ATCC #20362, ATCC #8862, ATCC #18944, ATCC #76982 and/or LGAM S(7)1 (Papanikolaou S., and Aggelis G., Bioresour. Technol. 82(1 ):43-9 (2002)).

Historically, various strains of Y. lipolytica have been used for the manufacture and production of: isocitrate lyase (DD259637); lipases (SU 1454852, WO2001083773, DD279267); polyhydroxyalkanoates (WO2001088144); citric acid (RU2096461 , RU2090611 , DD285372, DD285370, DD275480, DD227448, PL160027); erythritol (EP770683); 2-oxoglutaric acid (DD267999); γ-decalactone (U.S. 6,451 ,565, FR2734843); γ-dodecalatone (EP578388); and pyruvic acid (JP09252790). Fermentation Processes For PUFA Production

The transformed microbial host cell is grown under conditions that optimize desaturase and elongase activities and produce the greatest and the most economical yield of the preferred PUFAs. In general, media conditions that may be optimized include the type and amount of carbon source, the type and amount of nitrogen source, the carbon-to-nitrogen ratio, the oxygen level, growth temperature, pH, length of the biomass production phase, length of the oil accumulation phase and the time of cell harvest. Microorganisms of interest, such as oleaginous yeast, are grown in complex media (e.g., yeast extract-peptone-dextrose broth (YPD)) or a defined minimal media that lacks a component necessary for growth and thereby forces selection of the desired expression cassettes (e.g., Yeast Nitrogen Base (DIFCO Laboratories, Detroit, Ml)). Fermentation media in the present invention must contain a suitable carbon source. Suitable carbon sources may include, but are not limited to: monosaccharides (e.g., glucose, fructose), disaccharides (e.g., lactose, sucrose), oligosaccharides, polysaccharides (e.g., starch, cellulose or mixtures thereof), sugar alcohols (e.g., glycerol) or mixtures from renewable feedstocks (e.g., cheese whey permeate, comsteep liquor, sugar beet molasses, barley malt). Additionally, carbon sources may include alkanes, fatty acids, esters of fatty acids, monoglycerides, diglycerides, triglycerides, phospholipids and various commercial sources of fatty acids including vegetable oils (e.g., soybean oil) and animal fats. Additionally, the carbon source may include one-carbon sources (e.g., carbon dioxide, methanol, formaldehyde, formate and carbon-containing amines) for which metabolic conversion into key biochemical intermediates has been demonstrated. Hence it is contemplated that the source of carbon utilized in the present invention may encompass a wide variety of carbon-containing sources and will only be limited by the choice of the host organism. Although all of the above mentioned carbon sources and mixtures thereof are expected to be suitable in the present invention, preferred carbon sources are sugars and/or fatty acids. Most preferred is glucose and/or fatty acids containing between 10-22 carbons.

Nitrogen may be supplied from an inorganic (e.g., (NH4)₂Sθ4) or organic source (e.g., urea or glutamate). In addition to appropriate carbon and nitrogen sources, the fermentation media must also contain suitable minerals, salts, cofactors, buffers, vitamins and other components known to those skilled in the art suitable for the growth of the microorganism and promotion of the enzymatic pathways necessary for PUFA production. Particular attention is given to several metal ions (e.g., Mn⁺², Co⁺², Zn⁺², Mg⁺²) that promote synthesis of lipids and PUFAs (Nakahara, T. et al., Ind. Appl. Single Cell Oils, D. J. Kyle and R. Colin, eds. pp 61-97 (1992)). Preferred growth media in the present invention are common commercially prepared media, such as Yeast Nitrogen Base (DIFCO Laboratories, Detroit, Ml). Other defined or synthetic growth media may also be used and the appropriate medium for growth of the particular microorganism will be known by one skilled in the art of microbiology or fermentation science. A suitable pH range for the fermentation is typically between about pH 4.0 to pH 8.0, wherein pH 5.5 to pH 7.0 is preferred as the range for the initial growth conditions. The fermentation may be conducted under aerobic or anaerobic conditions, wherein microaerobic conditions are preferred.

Typically, accumulation of high levels of PUFAs in oleaginous yeast cells requires a two-stage process, since the metabolic state must be "balanced" between growth and synthesis/storage of fats. Thus, most preferably, a two-stage fermentation process is necessary for the production of PUFAs in oleaginous yeast. In this approach, the first stage of the fermentation is dedicated to the generation and accumulation of cell mass and is characterized by rapid cell growth and cell division. In the second stage of the fermentation, it is preferable to establish conditions of nitrogen deprivation in the culture to promote high levels of lipid accumulation. The effect of this nitrogen deprivation is to reduce the effective concentration of AMP in the cells, thereby reducing the activity of the NAD-dependent isocitrate dehydrogenase of mitochondria. When this occurs, citric acid will accumulate, thus forming abundant pools of acetyl- CoA in the cytoplasm and priming fatty acid synthesis. Thus, this phase is characterized by the cessation of cell division followed by the synthesis of fatty acids and accumulation of oil.

Although cells are typically grown at about 30°C, some studies have shown increased synthesis of unsaturated fatty acids at lower temperatures (Yongmanitchai and Ward, Appl. Environ. Microbiol. 57:419-25 (1991)). Based on process economics, this temperature shift should likely occur after the first phase of the two-stage fermentation, when the bulk of the organisms' growth has occurred. It is contemplated that a variety of fermentation process designs may be applied, where commercial production of omega fatty acids using recombinant expression of desaturase and/or elongase genes is desired. For example, commercial production of PUFAs from a recombinant microbial host may be produced by a batch, fed-batch or continuous fermentation process.

A batch fermentation process is a closed system wherein the media composition is fixed at the beginning of the process and not subject to further additions beyond those required for maintenance of pH and oxygen level during the process. Thus, at the beginning of the culturing process the media is inoculated with the desired organism and growth or metabolic activity is permitted to occur without adding additional sources (i.e., carbon and nitrogen sources) to the medium. In batch processes the metabolite and biomass compositions of the system change constantly up to the time the culture is terminated. In a typical batch process, cells proceed through a static lag phase to a high-growth log phase and finally to a stationary phase, wherein the growth rate is diminished or halted. Left untreated, cells in the stationary phase will eventually die. A variation of the standard batch process is the fed-batch process, wherein the carbon source is continually added to the fermentor over the course of the fermentation process. A fed-batch process is also suitable in the present invention. Fed-batch processes are useful when catabolite repression is apt to inhibit the metabolism of the cells or where it is desirable to have limited amounts of carbon source in the media at any one time. Measurement of the carbon source concentration in fed-batch systems is difficult and therefore may be estimated on the basis of the changes of measurable factors such as pH, dissolved oxygen and the partial pressure of waste gases (e.g., C0₂). Batch and fed-batch culturing methods are common and well known in the art and examples may be found in Thomas D. Brock in Biotechnology: A Textbook of Industrial Microbiology, 2^nd ed., (1989) Sinauer Associates: Sunderland, MA; or Deshpande, Mukund V., Appl. Biochem. Biotechnol., 36:227 (1992), herein incorporated by reference. Commercial production of omega fatty acids using recombinant expression of desaturase and/or elongase genes may also be accomplished by a continuous fermentation process wherein a defined media is continuously added to a bioreactor while an equal amount of culture volume is removed simultaneously for product recovery. Continuous cultures generally maintain the cells in the log phase of growth at a constant cell density. Continuous or semi-continuous culture methods permit the modulation of one factor or any number of factors that affect cell growth or end product concentration. For example, one approach may limit the carbon source and allow all other parameters to moderate metabolism. In other systems, a number of factors affecting growth may be altered continuously while the cell concentration, measured by media turbidity, is kept constant. Continuous systems strive to maintain steady state growth and thus the cell growth rate must be balanced against cell loss due to media being drawn off the culture. Methods of modulating nutrients and growth factors for continuous culture processes, as well as techniques for maximizing the rate of product formation, are well known in the art of industrial microbiology and a variety of methods are detailed by Brock, supra. Purification Of PUFAs

The PUFAs may be found in the host microorganism as free fatty acids or in esterified forms such as acylglycerols, phospholipids, sulfolipids or glycolipids, and may be extracted from the host cell through a variety of means well-known in the art. One review of extraction techniques, quality analysis and acceptability standards for yeast lipids is that of Z. Jacobs (Critical Reviews in Biotechnology 12(5/6) :463-491 (1992)). A brief review of downstream processing is also available by A. Singh and O. Ward (Adv. Appl. Microbiol. 45:271-312 (1997)).

In general, means for the purification of PUFAs may include extraction with organic solvents, sonication, supercritical fluid extraction (e.g., using carbon dioxide), saponification and physical means such as presses, or combinations thereof. Of particular interest is extraction with methanol and chloroform in the presence of water (E. G. Bligh & W.J. Dyer, Can. d. Biochem. Physiol. 37:911-917 (1959)). Where desirable, the aqueous layer can be acidified to protonate negatively-charged moieties and thereby increase partitioning of desired products into the organic layer. After extraction, the organic solvents can be removed by evaporation under a stream of nitrogen. When isolated in conjugated forms, the products may be enzymatically or chemically cleaved to release the free fatty acid or a less complex conjugate of interest, and can then be subject to further manipulations to produce a desired end product. Desirably, conjugated forms of fatty acids are cleaved with potassium hydroxide.

If further purification is necessary, standard methods can be employed. Such methods may include extraction, treatment with urea, fractional crystallization, HPLC, fractional distillation, silica gel chromatography, high-speed centrifugation or distillation, or combinations of these techniques. Protection of reactive groups, such as the acid or alkenyl groups, may be done at any step through known techniques (e.g., alkylation, iodination). Methods used include methylation of the fatty acids to produce methyl esters. Similarly, protecting groups may be removed at any step. Desirably, purification of fractions containing GLA, STA, ARA, DHA and EPA may be accomplished by treatment with urea and/or fractional distillation. DESCRIPTION OF PREFERRED EMBODIMENTS The present invention demonstrates the feasibility of introducing an ω-3 and/or ω-6 biosynthetic pathway into oleaginous yeast for the production of PUFAs. Toward this end, ARA (representative of ω-6 fatty acids) and EPA (representative of ω-3 fatty acids) were selected as desirable products to produce in the oleaginous yeast, Yarrowia lipolytica. Thus, the synthesis of ARA required the introduction of genes encoding Δ6 desaturase, elongase and Δ5 desaturase activities into Yarrowia, whereas the synthesis of EPA required the introduction of genes encoding Δ6 desaturase, elongase, Δ5 desaturase and Δ17 desaturase activities into Yarrowia.

A variety of publicly available Δ5 desaturases from different organisms having the ability to convert DGLA to ARA and ETA to EPA were expressed in Yarrowia lipolytica and screened for activity, in order to identify the gene demonstrating the highest level of activity in the alternate host. On this basis, a Mortierella alpina Δ5 desaturase (SEQ ID NO:4) was selected as the preferred gene for expression in oleaginous yeast, based on its ability to convert -30% of intracellular DGLA to ARA in a substrate feeding trial. Additional substrate feeding trials were conducted to verify the enzymatic activities encoded by the following genes:

• A . alpina Δ6 desaturase (SEQ ID NO:2) converts LA to GLA and ALA to STA (wherein the percent substrate conversion of LA to GLA in Y. lipolytica was ~30%); • A Saprolegnia diclina Δ17 desaturase (SEQ ID NO:6) converts DGLA to ETA and ARA to EPA (wherein the percent substrate conversion of ARA to EPA in Y. lipolytica was -23%); and

• A . alpina high affinity PUFA elongase (SEQ ID NO:8) converts GLA to DGLA, STA to ETA and EPA to DPA

(wherein the percent substrate conversion of GLA to DGLA in Y. lipolytica was ~30%). Based on the lower percent substrate conversion of the S. diclina Δ17 desaturase (relative to the Δ6 and Δ5 desaturase and the elongase), this particular gene was codon-optimized to enhance its expression in

Yarrowia. This was accomplished by determining the codon usage and signature of structural genes in Yarrowia lipolytica, designing a codon- optimized Δ17 desaturase gene, and then synthesizing the gene in vitro to enable its increased efficiency in the alternate host (with respect to the wildtype gene).

To enable synthesis of ARA or EPA (and thereby demonstrate proof-of-concept for the ability of oleaginous hosts to be engineered for production of ω-6 and ω-3 fatty acids (i.e., ARA and EPA)), two different DNA expression constructs were subsequently prepared: 1.) the first contained the Δ6 desaturase, Δ5 desaturase and high-affinity PUFA elongase; and 2.) the second contained the Δ6 desaturase, Δ5 desaturase, high-affinity PUFA elongase and codon-optimized Δ17 desaturase. Both constructs were separately transformed into Yarrowia lipolytica and integrated into the chromosomal URA3 gene encoding the enzyme orotidine-5'-phosphate decarboxylase (EC 4.1.1.23). GC analysis of the host cells fed with appropriate substrates detected production of ARA (Example 5) and EPA (Example 6), respectively. Thus, this is the first demonstration of PUFA biosynthesis in an oleaginous yeast whereby the ω-3 and/or ω-6 biosynthetic pathways have been introduced into an oleaginous yeast.

On the basis of the teachings and results described herein, it is expected that one skilled in the art will recognize the feasability and commercial utility created by using oleaginous yeast as a production platform for the synthesis of a variety of ω-3 and/or ω-6 PUFAs.

EXAMPLES The present invention is further defined in the following Examples. It should be understood that these Examples, while indicating preferred embodiments of the invention, are given by way of illustration only. From the above discussion and these Examples, one skilled in the art can ascertain the essential characteristics of this invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions. GENERAL METHODS

Standard recombinant DNA and molecular cloning techniques used in the Examples are well known in the art and are described by: 1.) Sambrook, J., Fritsch, E. F. and Maniatis, T. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY (1989) (Maniatis); 2.) T. J. Silhavy, M. L. Bennan, and L. W. Enquist, Experiments with Gene Fusions; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY (1984); and 3.) Ausubel, F. M. et al., Current Protocols in Molecular Biology, published by Greene Publishing Assoc. and Wiley- lnterscience (1987). Materials and methods suitable for the maintenance and growth of microbial cultures are well known in the art. Techniques suitable for use in the following examples may be found as set out in Manual of Methods for General Bacteriology (Phillipp Gerhardt, R. G. E. Murray, Ralph N. Costilow, Eugene W. Nester, Willis A. Wood, Noel R. Krieg and G. Briggs Phillips, Eds), American Society for Microbiology: Washington, D.C.

(1994)); or by Thomas D. Brock in Biotechnology: A Textbook of Industrial Microbiology, 2^nd ed., Sinauer Associates: Sunderland, MA (1989). All reagents, restriction enzymes and materials used for the growth and maintenance of microbial cells were obtained from Aldrich Chemicals (Milwaukee, WI), DIFCO Laboratories (Detroit, Ml), GIBCO/BRL

(Gaithersburg, MD), or Sigma Chemical Company (St. Louis, MO), unless otherwise specified.

E. coli (XL1-Blue) competent cells were purchased from the Stratagene Company (San Diego, CA). E. coli strains were typically grown at 37 °C on Luria Bertani (LB) plates.

General molecular cloning was performed according to standard methods (Sambrook et al., supra). Oligonucleotides were synthesized by Sigma-Genosys (Spring, TX). Site-directed mutagenesis was performed using Stratagene's QuickChange™ Site-Directed Mutagenesis kit, per the manufacturers' instructions. When polymerase chain reaction (PCR) or site-directed mutagenesis was involved in subcloning, the constructs were sequenced to confirm that no errors had been introduced to the sequence. PCR products were cloned into Promega's pGEM-T-easy vector (Madison, WI). DNA sequence was generated on an ABI Automatic sequencer using dye terminator technology (U.S. 5,366,860; EP 272,007) using a combination of vector and insert-specific primers. Sequence editing was performed in Sequencher (Gene Codes Corporation, Ann Arbor, Ml). All sequences represent coverage at least two times in both directions. Comparisons of genetic sequences were accomplished using DNASTAR software (DNA Star, Inc.). Alternatively, manipulations of genetic sequences were accomplished using the suite of programs available from the Genetics Computer Group Inc. (Wisconsin Package Version 9.0, Genetics Computer Group (GCG), Madison, WI). The GCG program "Pileup" was used with the gap creation default value of 12, and the gap extension default value of 4. The GCG "Gap" or "Bestfit" programs were used with the default gap creation penalty of 50 and the default gap extension penalty of 3. Unless otherwise stated, in all other cases GCG program default parameters were used.

The meaning of abbreviations is as follows: "sec" means second(s), "min" means minute(s), "h" means hour(s), "d" means day(s), "μL" means microliter(s), "mL" means milliliter(s), "L" means liter(s), "μM" means micromolar, "mM" means millimolar, "M" means molar, "mmol" means millimole(s), "μmole" mean micromole(s), "g" means gram(s), "μg" means microgram(s), "ng" means nanogram(s), "U" means unit(s), "bp" means base pair(s), and "kB" means kilobase(s). Cultivation Of Yarrowia lipolytica

Yarrowia lipolytica strains ATCC #76982 and ATCC #90812 were purchased from the American Type Culture Collection (Rockville, MD). Y. lipolytica strains were usually grown at 28 °C on YPD agar (1% yeast extract, 2% bactopeptone, 2% glucose, 2% agar). For selection of transformants, minimal medium (0.17% yeast nitrogen base (DIFCO Laboratories, Detroit, Ml) without ammonium sulfate or amino acids, 2% glucose, 0.1 % proline, pH 6.1) was used. Supplements of adenine, leucine, lysine and/or uracil were added as appropriate to a final concentration of 0.01 %. Fatty Acid Analysis Of Yarrowia lipolytica

For fatty acid analysis, cells were collected by centrifugation and lipids were extracted as described in Bligh, E. G. & Dyer, W. J. (Can. J. Biochem. Physiol. 37:911-917 (1959)). Fatty acid methyl esters were prepared by transesterification of the lipid extract with sodium methoxide (Roughan, G., and Nishida I. Arch Biochem Biophys. 276(1 ):38-46 (1990)) and subsequently analyzed with a Hewlett-Packard 6890 GC fitted with a 30-m X 0.25 mm (i.d.) HP-INNOWAX (Hewlett-Packard) column. The oven temperature was from 170 °C (25 min hold) to 185 °C at 3.5 °C/min. For direct base transesterification, Yarrowia culture (3 mL) was harvested, washed once in distilled water, and dried under vacuum in a Speed-Vac for 5-10 min. Sodium methoxide (100 μl of 1 %) was added to the sample, and then the sample was vortexed and rocked for 20 min. After adding 3 drops of 1 M NaCI and 400 μl hexane, the sample was vortexed and spun. The upper layer was removed and analyzed by GC as described above.

EXAMPLE 1 Construction Of Plasmids Suitable For Heterologous Gene Expression In

Yarrowia lipolytica The plasmid pY5, a derivative of plNA532 (a gift from Dr. Claude Gaillardin, Insitut National Agronomics, Centre de biotechnologie Agro- Industrielle, laboratoire de Genetique Moleculaire et Cellularie INRA- CNRS, F-78850 Thiverval-Grignon, France), was constructed for expression of heterologous genes in Yarrowia lipolytica, as diagrammed in Figure 3.

First, the partially-digested 3598 bp EcoRI fragment containing the ARS18 sequence and LEU2 gene of plNA532 was subcloned into the EcoRI site of pBluescript (Strategene, San Diego, CA) to generate pY2. The TEF promoter (Muller S., et al. Yeast, 14: 1267-1283 (1998)) was amplified from Yarrowia lipolytica genomic DNA by PCR using TEF5' (SEQ ID NO:38) and TEF3' (SEQ ID NO:39) as primers. PCR amplification was carried out in a 50 μl total volume containing: 100 ng Yarrowia genomic DNA, PCR buffer containing 10 mM KCl, 10 mM

(NH₄)₂S0₄, 20 mM Tris-HCl (pH 8.75), 2 mM MgS0₄, 0.1% Triton X-100, 100 μg/mL BSA (final concentration), 200 μM each deoxyribonucleotide triphosphate, 10 pmole of each primer and 1 μl of PfuTurbo DNA polymerase (Stratagene, San Diego, CA). Amplification was carried out as follows: initial denaturation at 95 °C for 3 min, followed by 35 cycles of the following: 95 ^°C for 1 min, 56 ^°C for 30 sec, 72 ^°C for 1 min. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 °C. The 418 bp PCR product was ligated into pCR-Blunt to generate plP-tef. The BamHI/EcoRV fragment of plP-tef was subcloned into the BamHI/Smal sites of pY2 to generate pY4.

The XPR2 transcriptional terminator was amplified by PCR using plNA532 as template and XPR5' (SEQ ID NO:40) and XPR3' (SEQ ID NO:41) as primers. The PCR amplification was carried out in a 50 μl total volume, using the components and conditions described above. The 179 bp PCR product was digested with Sacll and then ligated into the Sacll site of pY4 to generate pY5. Thus, pY5 (shown in Figures 3 and 4) is useful as a Yarrowia-E. coli shuttle plasmid containing: 1.) a Yarrowia autonomous replication sequence (ARS18); 2.) a ColE1 plasmid origin of replication; 3.) an ampicillin-resistance gene (Amp^R), for selection in E. coli; 4.) a Yarrowia LEU2 gene (E.C. 4.2.1.33, encoding isopropylmalate isomerase), for selection in Yarrowia;

5.) the translation elongation promoter (TEF P), for expression of heterologous genes in Yarrowia; and 6.) the extracellular protease gene terminator (XPR2) for transcriptional termination of heterologous gene expression in Yarrowia. pY5-13 (Figure 4) was constructed as a derivative of pY5 to faciliate subcloning and heterologous gene expression in Yarrowia lipolytica. Specifically, pY5-13 was constructed by 6 rounds of site-directed mutagenesis using pY5 as template. Both Sail and Clal sites were eliminated from pY5 by site-directed mutagenesis using oligonucleotides YL5 and YL6 (SEQ ID NOs:106 and 107) to generate pY5-5. A Sail site was introduced into pY5-5 between the Leu2 gene and the TEF promoter by site-directed mutagenesis using oligouncleotides YL9 and YL10 (SEQ ID NOs:110 and 111) to generate pY5-6. A Pad site was introduced into , pY5-6 between the LEU2 gene and ARS18 using oligouncleotides YL7 and YL8 (SEQ ID NOs:108 and 109) to generate pY5-8. A Ncol site was introduced into pY5-8 around the translation start codon of theTEF promoter using oligonucleotides YL3 and YL4 (SEQ ID NOs:104 and 105) to generate pY5-9. The Ncol site inside the Leu2 gene of pY5-9 was eliminated using YL1 and YL2 oligonucleotides (SEQ ID NOs:102 and

103) to generate pY5-12. Finally, a BsiWI site was introduced into pY5-12 between the ColEI and XPR2 region using oligouncleotides YL61 and YL62 (SEQ ID NOs:88 and 89) to generate pY5-13.

A second derivative of plasmid pY5 was constructed to faciliate subcloning. Specifically, pY5-4 (Figure 4) was constructed by three rounds of site-directed mutagenesis using pY5 as template. A Ncol site located inside the Leu2 reporter gene was eliminated from pY5 using oligonucleotides YL1 and YL2 (SEQ ID NOs:102 and 103) to generate pY5-1. A Ncol site was introduced into pY5-1 between the TEF promoter and XPR2 transcriptional terminator by site-directed mutagenesis using oligonucleotides YL3 and YL4 (SEQ ID NOs:104 and 105) to generate pY5-2. A Pad site was then introduced into pY5-2 between the TEF promoter and XPR2 transcriptional terminator using oligonucleotides YL23 and YL24 (SEQ ID NOs:112 and 113) to generate pY5-4.

EXAMPLE 2 Selection Of Δ6 Desaturase, Δ5 Desaturase, Δ17 Desaturase And High Affinity PUFA Elongase Genes For Expression In Yarrowia lipolytica

Prior to the introduction of specific genes encoding an ω-3 and/or ω-6 biosynthetic pathway into oleaginous yeast, it was necessary to confirm the functionality of heterologous Δ6 desaturase, elongase, Δ5 desaturase and Δ17 desaturase genes expressed in Yarrowia. This was accomplished by measuring the conversion efficiency encoded by each wildtype gene in the alternate host. Specifically, four Δ5 desaturases, a Mortierella alpina Δ6 desaturase, a Saprolegnia diclina Δ17 desaturase and a M. alpina high affinity PUFA elongase were separately expressed and screened for activity in substrate-feeding trials. Based on these results, a M. alpina Δ5 desaturase gene was selected for use in conjunction with the Δ6 and Δ17 desaturase and high affinity PUFA elongase genes. Construction Of Expression Plasmids

In general, wildtype desaturase or elongase genes were either isolated by restriction digestion or amplified by PCR and inserted into appropriate vectors for expression. Each PCR amplification was carried out in a 50 μl total volume, comprising PCR buffer containing: 10 ng template, 10 mM KCI, 10 mM (NH₄)₂S0₄, 20 mM Tris-HCl (pH 8.75), 2 mM MgS0₄, 0.1% Triton X-100, 100 μg/mL BSA (final concentration), 200 μM each deoxyribonucleotide triphosphate, 10 pmole of each primer and 1 μl of PfuTurbo DNA polymerase (Stratagene, San Diego, CA). Amplification was carried out as follows (unless otherwise specified): initial denaturation at 95 °C for 3 min, followed by 35 cycles of the following: 95 ^°C for 1 min, 56 ^°C for 30 sec, 72 °C for 1 min. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 ^°C.

Wild Type Mortierella alpina (Accession #AF465281) Δ6 Desaturase

The 1384 bp Ncol/Notl fragment of pCGR5 (U.S. 5,968,809), which contains the M. alpina Δ6 desaturase gene (SEQ ID NO:1), was inserted into the Ncol/Notl sites of pY5-2 (Example 1) to generate pY54. Wild Type Mortierella alpina (Accession #AF067654) Δ5

Desaturase

The M. alpina Δ5 desaturase gene (SEQ ID NO:3) was amplified by PCR using oligonucleotides YL11 and YL12 (SEQ ID NOs:72 and 73) as primers and plasmid pCGR-4 (U.S. 6,075,183) as template. PCR amplification was carried out as described above, with the exception that the elongation step was extended to 1.5 min (for cycles 1-35). The 1357 bp PCR product was digested with Ncol/Notl and ligated to Λ/co//Λ/of/-digested pY5-13 (described in Example 1) to generate pYMAδpb (Figure 5).

Wild Type Saprolegnia diclina (ATCC #56851) Δ5 Desaturase

The S. diclina Δ5 desaturase gene (SEQ ID NO:114) was amplified by PCR using oligonucleotides YL13A and YL14 (SEQ ID NOs:116 and 117) as primers and plasmid pRSP3 (WO 02/081668) as template. PCR amplification was carried out as described above, with the exception that the elongation step was extended to 1.5 min (for cycles 1-35). The 1.4 kB PCR product was digested with Ncol/Pacl and ligated to Ncol/Pacl- digested pY5-4 (Figure 4; described in Example 1) to generate pYSD5.

Wild Type Isochrysis galbana CCMP1323 Δ5 Desaturase The /. galbana Δ5-desaturase gene (SEQ ID NO:118) was amplified by PCR using oligonucleotides YL19A and YL20 (SEQ ID NOs:120 and 121) as primers and plasmid pRIG-1 (WO 02/081668 A2) as template. PCR amplification was carried out as described above, with the exception that the elongation step was extended to 1.5 min (for cycles 1-35). The 1.4 kB PCR product was digested with BamHI/Pacll and ligated to BamHI/Pacll -digested pY5-4 (described in Example 1) to generate pYIG5.

Wild Type Thraustochytrium aureum (ATCC #34304) Δ5 Desaturase

The T aureum Δ5-desaturase gene (SEQ ID NO: 122) was amplified by PCR using oligonucleotides YL15 and YL16B (SEQ ID

NOs:124 and 125) as primers and plasmid pRTA4 (WO 02/081668 A2) as template. PCR amplification was carried out as described above, with the exception that the elongation step was extended to 1.5 min (for cycles 1- 35). The 1.4 kB PCR product was digested with Ncol/Notl and ligated to Ncol/Notl -digested pY5-2 (described in Example 1) to generate pYTA5. Wild Type Saprolegnia diclina (ATCC #56851) Δ17 Desaturase

The wild type Δ17 desaturase gene of S. diclina was amplified from plasmid pRSP19 (US 2003/0196217 A1) by PCR using oligonucleotides YL21A (SEQ ID NO:42) and YL22 (SEQ ID NO:43) as primers. The PCR products were digested with Ncol/Pacl and then ligated to Ncol/Pacl - digested pY5-4 (Figure 4; described in Example 1) to generate pYSD17.

Wild Type Mortierella alpina (Accession #AX464731) High Affinity Elongase

The 973 bp Notl fragment of pRPB2 (WO 00/12720), containing the coding region of a M. alpina high affinity PUFA elongase gene (SEQ ID NO:7), was inserted into the Notl s^'Ae of pY5 (described in Example 1 ; Figures 3 and 4) to generate pY58. Transformation Of Yarrowia lipolytica

The plasmids pY54, pYMAδpb, pYSD5, pYIG5, pYTA5, pYSD17 and pY58 were transformed separately into Y. lipolytica ATCC #76982 according to the method of Chen, D. C. et al. (Appl Microbiol Biotechnol. 48(2):232-235-(1997)).

Briefly, a leucine auxotroph of Yarrowia was streaked onto a YPD plate and grown at 30 ^°C for approximately 18 hr. Several large loopfuls of cells were scraped from the plate and resuspended in 1 mL of transformation buffer containing:

• 2.25 mL of 50% PEG, average MW 3350;

• 0.125 mL of 2 M Li acetate, pH 6.0;

• 0.125 mL of 2M DTT; and • 50 μg sheared salmon sperm DNA.

About 500 ng of plasmid DNA were incubated in 100 μl of resuspended cells, and maintained at 39 ^°C for 1 hr with vortex mixing at 15 min intervals. The cells were plated onto minimal media plates lacking leucine and maintained at 30 °C for 2 to 3 days. Determination Of Percent Substrate Conversion

Single colonies of transformant Y. lipolytica containing pY54, pYMAδpb, pYSD5, pYIG5, pYTA5, pYSD17 or pY58 were each grown in 3 mL minimal media (20 g/L glucose, 1.7 g/L yeast nitrogen base without amino acids, 1 g/L L-proline, 0.1 g/L L-adenine, 0.1 g/L L-lysine, pH 6.1) at 30 °C to an OD₆₀₀ ~ 1.0. For substrate feeding, 100 μl of cells were then subcultured in 3 mL minimal media containing 10 μg of substrate for about 24 hr at 30 ^°C. Cells were subsequently collected by centrifugation, the lipids were extracted, and fatty acid methyl esters were prepared by transesterification and subsequently analyzed by GC (as described in the General Methods). Percent substrate conversion was determined as: [product/(substrate + product)]*100. Percent Substrate Conversion By M. alpina Δ6 Desaturase

The M. alpina Δ6 desaturase converts LA to GLA and ALA to STA. Y. lipolytica strains containing pY54 were grown as described above (no substrate feeding required) and lipids were analyzed. The results showed that Yarrowia strains with pY54 converted about 30% LA to GLA. Percent Substrate Conversion By M. alpina, S. diclina, I. galbana and T. aureum Δ5 Desaturases

The Δ5 desaturases from M. alpina, S. diclina, I. galbana and T. aureum each convert DGLA to ARA and ETA to EPA. Y. lipolytica strains containing pYMAδpb, pYSD5, pYIG5 or pYTA5 were grown separately from single colonies, subcultured in minimal media containing 10 μg of DGLA, and then subjected to lipid analysis as described above. Yarrowia strains with pYMAδpb (M. alpina) converted about 30% of intracellular DGLA to ARA; the Yarrowia strains with pYSD5 (S. diclina) converted about 12%; the Yarrowia strains with pYlGδ (/. galbana) converted about 7%; and the Yarrowia strains with pYTAδ (T. aureum) converted about 23% of intracellular DGLA to ARA.

Percent Substrate Conversion By S. diclina Δ17 Desaturase The S. diclina Δ17 desaturase converts ARA to EPA and DGLA to ETA. Y. lipolytica strains containing pYSD17 were grown from single colonies, subcultured in minimal media containing 10 μg of ARA, and subjected to lipid analysis as described above. The results of the ARA feeding experiments showed that Yarrowia strains with pYSD17 converted about 23% of intracellular ARA to EPA.

Percent Substrate Conversion Of Wild Type M. alpina High Affinity Elongase

The M. alpina high affinity PUFA elongase converts GLA to DGLA, STA to ETA and EPA to DPA. Y. lipolytica strains containing pYδδ were grown from single colonies, subcultured in minimal media containing 10 μg of GLA, and subjected to lipid analysis as described above. The results of the GLA feeding experiments showed that Yarrowia strains with pYδ8 converted about 30% of intracellular GLA to DGLA.

δ7 EXAMPLE 3 Synthesis And Expression Of A Codon-Qptimized Δ17 Desaturase Gene

In Yarrowia lipolytica Based on the results of Example 2, genes encoding Δ6 desaturase, elongase and Δδ desaturase activies were available that each enabled -30% substrate conversion nYarrowia lipolytica. The Δ17 desaturase from S. diclina, however, had a maximum percent substrate conversion of only 23%. Thus, a codon-optimized Δ17 desaturase gene was designed, based on the Saprolegnia diclina DNA sequence (SEQ ID NO:δ), according to the Yarrowia codon usage pattern, the consensus sequence around the ATG translation initiation codon and the general rules of RNA stability (Guhaniyogi, G. and J. Brewer, Gene 26δ(1-2):11-23 (2001)).

In addition to modification to the translation initiation site, 127 bp of the 1077 bp coding region (comprising 117 codons) were codon- optimized. A comparison between this codon-optimized DNA sequence (SEQ ID NO:9) and the S. diclina Δ17 desaturase gene DNA sequence (SEQ ID NO:δ) is shown in Figure 6, wherein nucleotides in bold text correspond to nucleotides that were modified in the codon-optimized gene. None of the modifications in the codon-optimized gene changed the amino acid sequence of the encoded protein (SEQ ID NO:6).

Determining The Preferred Codon Usage In Yarrowia lipolytica

Approximately 100 genes of Y. lipolytica were found in the National Center for Biotechnology Information public database. The coding regions of these genes, comprising 121 ,167 bp, were translated by the Editseq program of DNAStar to the corresponding 40,389 amino acids and were tabulated to determine the Y. lipolytica codon usage profile shown in Table 3. The column titled "No." refers to the number of times a given codon encodes a particular amino acid in the sample of 40,389 amino acids. The column titled "%" refers to the frequency that a given codon encodes a particular amino acid. Entries shown in bold text represent the codons favored in Yarrowia lipolytica.

δ8 TABLE 3 Codon Usage In Yarrowia lipolytica

For further optimization of gene expression in Y. lipolytica, the consensus sequence around the 'ATG' initiation codon of 79 genes was examined. In Figure 7, the first 'A' of the underlined ATG translation codon is considered to be +1. Seventy seven percent of the genes analyzed had an 'A' in the -3 position, indicating a strong preference for 'A' at this position. There was also preference for 'A' or 'C at the -4, -2 and -1 positions, an 'A', 'C or T at position +5, and a 'G' or 'C at position +6. Thus, the preferred consensus sequence of the codon-optimized translation initiation site for optimal expression of genes in Y. lipolytica is 'MAMMATGNHS' (SEQ ID NO:126), wherein the nucleic acid degeneracy code used is as follows: M=A/C; S=C/G; H=A/C/T; and N=A/C/G/T. In Vitro Synthesis Of A Codon-Optimized Gene

The method used to synthesize the codon-optimized Δ17 desaturase gene is illustrated in Figure 8. First, eleven pairs of oligonucleotides were designed to extend the entire length of the codon- optimized coding region of the S. diclina Δ17 desaturase gene (e.g., D17- 1A, D17-1 B, D17-2A, D17-2B, D17-3A, D17-3B, D17-4A, D17-4B, D17-δA, D17-6B, D17-6A, D17-6B, D17-7A, D17-7B, D17-8A, D17-8B, D17-9A, D17-9B, D17-10A, D17-10B, D17-11A and D17-11 B, corresponding to SEQ ID NOs:10-31). Each pair of sense (A) and anti-sense (B) oligonucleotides were complementary, with the exception of a 4 bp overhang at each δ'-end. Additionally, primers D17-1A, D17-4B, D17-6A, D17-8A and D17-8B also introduced Ncol, Bgili and Sail restriction sites for subsequent subcloning, respectively.

100 ng of each oligonucleotide was phosphorylated at 37 °C for 1 hr in a volume of 20 μl containing δO mM Tris-HCl (pH 7.δ), 10 mM MgCI₂, 10 mM DTT, O.δ mM spermidine, O.δ mM ATP and 10 U of T4 polynucleotide kinase. Each pair of sense and antisense oligonucleotides was mixed and annealed in a thermocycler using the following parameters: 9δ °C (2 min), 8δ °C (2 min), 6δ °C (1δ min), 37 °C (1δ min), 24 °C (1δ min) and 4 °C (1δ min). Thus, D17-1A (SEQ ID NO:10) was annealed to D17-1 B (SEQ ID NO:11) to produce the double-stranded product "D17- 1AB". Similarly, D17-2A (SEQ ID NO:12) was annealed to D17-2B (SEQ ID NO:13) to produce the double-stranded product "D17-2AB", etc. Three separate pools of annealed, double-stranded oligonucleotides were then ligated together, as shown below:

• Pool 1: comprised D17-1AB, D17-2AB, D17-3AB and D17-4AB;

• Pool 2: comprised D17-δAB, D17-6AB, D17-7AB and D17-8AB; and

• Pool 3: comprised D17-9AB, D17-10AB and D17-11AB.

Each pool of annealed oligonucleotides was mixed in a volume of 20 μl with 10 U of T4 DNA ligase and the ligation reaction was incubated overnight at 16 °C.

The product of each ligation reaction was then amplified by PCR. Specifically, using the ligated "Pool 1" mixture (i.e., D17-1AB, D17-2AB, D17-3AB and D17-4AB) as template, and oligonucleotides D17-1 (SEQ ID NO:32) and D17-4R (SEQ ID NO:33) as primers, the first portion of the codon-optimized Δ17 desaturase gene was amplified by PCR. The PCR amplification was carried out in a δO μl total volume, comprising PCR buffer containing 10 mM KCI, 10 mM (NH₄) ₂S0₄, 20 mM Tris-HCl (pH 8.7δ), 2 mM MgS0₄, 0.1 % Triton X-100, 100 μg/mL BSA (final concentration), 200 μM each deoxyribonucleotide triphosphate, 10 pmole of each primer and 1 μl of PfuTurbo DNA polymerase (Stratagene, San Diego, CA). Amplification was carried out as follows: initial denaturation at 9δ °C for 3 min, followed by 3δ cycles of the following: 9δ °C for 1 min, δδ °C for 30 sec, 72 ^°C for 40 sec. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 °C. The 430 bp PCR fragment was subcloned into the pGEM-T easy vector (Promega) to generate pT17(1-4).

Using the ligated "Pool 2" mixture (i.e., D17-δAB, D17-6AB, D17- 7AB and D17-8AB) as template, and oligonucleotides D17-δ (SEQ ID NO:34) and D17-8D (SEQ ID NO:3δ) as primers, the second portion of the codon-optimized Δ17 desaturase gene was amplified similarly by PCR and cloned into pGEM-T-easy vector to generate pT17(δ-8). Finally, using the "Pool 3" ligation mixture (i.e., D17-9AB, D17-10AB and D17-11AB) as template, and oligonucleotides D17-8U (SEQ ID NO:36) and D17-11 (SEQ ID NO:37) as primers, the third portion of the codon-optimized Δ17 desaturase gene was amplified similarly by PCR and cloned into pGEM-T- easy vector to generate pT17(9-11). E. coli was transformed separately with pT17(1-4), pT17(δ-8) and pT17(9-11) and the plasmid DNA was isolated from ampicillin-resistant transformants. Plasmid DNA was purified and digested with the appropriate restriction endonucleases to liberate the 420 bp Ncol/Bglll fragment of pT17(1-4), the 400 bp Bglll/Sall fragment of pT17(5-8) and the 300 bp Sall/Notl fragment of pT17(9-11). These fragments were then combined, ligated together and used as template for amplification of the entire synthetic codon-optimized Δ17 desaturase gene using D17-1 (SEQ ID NO: 32) and D17-11 (SEQ ID NO:37) as primers. The PCR amplification was carried out in a 50 μl total volume, using the conditions described above for each portion of the Δ17 desaturase gene and the thermocycling program as follows: initial denaturation at 9δ ^°C for 3 min, followed by 3δ cycles of the following: 9δ ^°C for 1 min, δ6 ^°C for 30 sec, 72 °C for 1.1 min. A final extension cycle of 72 ^°C for 10 min was carried out, followed by reaction termination at 4 ^°C. This generated a 1.1 kB PCR product.

Construction Of Plasmid pYSD17s Containing The Codon-Optimized Δ17 Desaturase

The 1.1 kB PCR product comprising the entire synthetic Δ17 desaturase was digested with Ncol/Notl and subcloned into Ncol/Notl - digested pYδ-13 (Example 1) to generate pYSD17S (Figure 9A).

As an additional "control", to compare the efficiency of the wild type and synthetic genes in Yarrowia, the AT-rich Pad site in pYSD17 (comprising the wild-type gene; described in Example 2) was eliminated by site-directed mutagenesis using YLδ3 (SEQ ID NO:44) and YLδ4 (SEQ ID NO:4δ) as primers to generate pYSD17M (Figure 9B). Transformation Of Yarrowia lipolytica With The Codon-Optimized Δ17 Desaturase Gene

Plasmids containing the wildtype and codon-optimized Δ17 desaturase were transformed separately into Y. lipolytica ATCC #76982 according to the methods described above in Example 2. Using this technique, transformants were obtained that contained the following plasmids: Table 4 Summary Of Plasmids In Transformant Yarrowia

Percent Substrate Conversion With The Codon-Optimized Δ17 Desaturase Gene

Δ17 desaturase converts ARA to EPA (see Figure 2). The percent substrate conversion ([product]/[substrate + product]*100) of the wildtype and codon-optimized Δ17 desaturase genes was determined in Yarrowia lipolytica containing each alternate plasmid construct, using the methodology described in the General Methods.

The results of the ARA feeding experiments showed that Yarrowia strains with control plasmids pYSD17 or pYSD17M converted about 23% of intracellular ARA to EPA (Figure 10A) while those containing the codon- optimized Δ17 desaturase gene within pYSD17S converted about 45% of intracellular ARA to EPA (Figure 10B). Thus, Yarrowia containing the codon-optimized Δ17 desaturase converted about 2-fold more ARA than the strains containing the wild type S. diclina gene.

EXAMPLE 4 Construction Of Plasmids Suitable For The Coordinate Expression Of

Multiple Omega Fatty Acid Biosynthesis Genes In Yarrowia lipolytica The present Example describes the synthesis of a variety of expression plasmids that were required in order to construct: 1.) a DNA fragment suitable for integration into the Yarrowia genome for expression of the Δ6 desaturase, PUFA elongase and Δ5 desaturase (for ARA production); and 2.) a DNA fragment suitable for integration into the Yarrowia genome for expression of the Δ6 desaturase, PUFA elongase, Δδ desaturase and Δ17 desaturase (for EPA production). Construction Of Plasmid pY24 Plasmid pY24 (Figure 11) was a parent vector for construction of expression cassettes suitable for integration into the genome of Yarrowia lipolytica. pY24 was constructed as follows: Using oligonucleotides KU5 and KU3 (SEQ ID NOs:46 and 47) as primers and Yarrowia genomic DNA as template, a 1.7 kB DNA fragment (SEQ ID NO:48) containing the Yarrowia URA3 gene was PCR amplified. The PCR amplification was carried out in a 50 μl total volume containing: 100 ng Yarrowia genomic DNA, PCR buffer containing 10 mM KCI, 10 mM (NH₄)₂S0₄, 20 mM Tris-HCl (pH 8.75), 2 mM MgS0₄, 0.1% Triton X-100, 100 μg/mL BSA (final concentration), 200 μM each deoxyribonucleotide triphosphate, 10 pmole of each primer and 1 μl of PfuTurbo DNA polymerase (Stratagene, San Diego, CA). Amplification was carried out as follows: initial denaturation at 95 ^°C for 3 min, followed by 3δ cycles of the following: 9δ °C for 1 min, δ6 °C for 30 sec, 72 °C for 2 min. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 ^°C. The PCR product was inserted into pGEM-T easy vector (Promega, Madison, WI) to generate pGYUM. Using oligonucleotides Klδ and KI3 (SEQ ID NOs:δ0 and δ1), a

1.1 kB DNA fragment (SEQ ID NO:δ2) containing the conjugase gene (or "imp H8") of Impatients balsama (clone ids.pk0001.h8; E. I. du Pont de Nemours and Company, Inc., Wilmington, DE) was PCR amplified. The PCR amplification was carried out in a δO μl total volume using the components described above, with the exception that 10 ng plasmid DNA of ids.pk0001.h8 was used as template. Amplification was carried out as follows: initial denaturation at 9δ ^°C for 3 min, followed by 36 cycles of the following: 9δ °C for 1.δ min, δ6 ^°C for 30 sec, 72 ^°C for 1.2 min. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 °C. The PCR products were digested with Notl, and then inserted into the Notl site of pYδ (Figure 3) to generate pY9.

Using oligonucleotides KTIδ and KTI3 (SEQ ID NOs:δ4 and δδ), a 1.7 kB DNA fragment (SEQ ID NO:δ6) containing the TEF::IMP H8::XPR chimeric gene of pY9 was PCR amplified. The PCR amplification was carried out in a δO μl total volume as described above, with the exception that 10 ng plasmid DNA of pGYUM was used as template. Amplification was carried out as follows: initial denaturation at 9δ ^°C for 3 min, followed by 36 cycles of the following: 9δ °C for 1 min, δδ ^°C for 30 sec, 72 ^°C for 2 min. A final extension cycle of 72 °C for 10 min was carried out, followed by reaction termination at 4 ^°C. The PCR products were inserted into PCR-Script (Stratagene) to generate pY9R. The 1.7 kB Xho/EcoRV fragment of pY9R was exchanged with the Xhol/EcoRV fragment of pGYUM to generate pY21.

Using oligonucleotides KHδ and KH3 (SEQ ID NOs:δ8 and 69) as primers and genomic DNA of KS6δ as template, a 1 kB DNA fragment (SEQ ID NO:60) containing the E. coli hygromycin resistance gene ("HPT"; Kaster, K.R., et al., Nucleic Acids Res. 11 :6896-6911 (1983)) was PCR amplified. The PCR amplification was carried out in a δO μl total volume using the components described above, with the exception that 10 ng plasmid DNA of ids.pk0001.h8 was used as template. Amplification was carried out as follows: initial denaturation at 9δ °C for 3 min, followed by 3δ cycles of the following: 9δ ^°C for 1 min, δδ ^°C for 30 sec, 72 °C for 1.2 min. A final extension cycle of 72 ^°C for 10 min was carried out, followed by reaction termination at 4 ^°C. The PCR products were digested with Notl, and then inserted into the Notl site of pYδ (Figure 3) to generate pTHPT-1. Using oligonucleotides KTHδ and KTH3 (SEQ ID NOs:62 and 63) as primers and pTHPT-1 plasmid DNA as template, a 1.6 kB DNA fragment (SEQ ID NO:64) containing the TEF::HPT::XPR fusion gene was amplified as described above. The PCR products were digested with Bglll and then inserted into pY21 to generate pY24. Construction Of pY24-4

Plasmid pY24 (Figure 11) was used for construction of expression cassettes suitable for integration into the Yarrowia lipolytica genome. The 401 bp of δ'-sequence (SEQ ID NO:66) and the 668 bp of 3'-sequence (SEQ ID NO:67) from the Y. lipolytica URA3 gene in pY24 plasmid were used to direct integration of expression cassettes into the Ura loci of

VneYarrowia genome. Two chimeric genes (TEF::HPT::XPR and TEF::IMP H8::XPR) were first removed from pY24 by digestion with BamHI and self- ligation to generate pY24-1. Pad and BsiWI sites were introduced into pY24-1 by site-directed mutagenesis using YL63/YL64 (SEQ ID NOs:68 and 69) and YL66/YL66 (SEQ ID NOs:70 and 71) primer pairs, respectively, to generate pY24-4.

Construction Of An Integration Vector For Expression Of Δδ Desaturase The 4261 bp Pacl/BsiWI fragment of pYMAδpb (comprising the M. alpina Δδ desaturase gene; described in Example 2) was ligated into the Pacl/BsiWI sites of pY24-4 (Figure 11 ) to generate pYZMδ (Figure δ). Hindlll and Clal sites were introduced into pYZMδ by site-directed mutagenesis using primer pairs YL81 and YL82 (SEQ ID NOs:74 and 76)

6δ and YL83 and YL84 (SEQ ID NOs:76 and 77), respectively, to generate pYZMδCH. A Pmel site was introduced into pYZMδCH by site-directed mutagenesis using YL106 and YL106 (SEQ ID NOs:78 and 79) as primers to generate pYZMδCHPP. An Asd site was introduced into pYZMδCHPP by site-directed mutagenesis using YL119 and YL120 (SEQ ID NOs:80 and 81) as primers to generate pYZMδCHPPA (Figure δ).

To optimize the integration vector, 440 bp of δ'-non-coding DNA sequence upstream from the Yarrowia lipolytica URA3 gene (SEQ ID NO:84) was amplified by PCR using YL121 and YL122 (SEQ ID NOs:82 and 83) as primers. The PCR product was digested with Asd and BsiWI and then exchanged with the Ascl/BsiWl fragment of pYZMδCHPPA (Figure δ and 12) to generate pYZMδUPA (Figure 12). An Asd site was introduced into pYZMδUPA by site-directed mutagenesis using oligonucleotides YL114 and YL11δ (SEQ ID NOs:8δ and 86) to generate pYZVδ. In order to reduce the size of the 3'-non-coding region of the URA3 gene in pYZVδ, a second Pad site was introduced into the middle of this region by site-directed mutagenesis using oligonucleotides YL114 and YL116 (described above) to generate pYZVδP. The Pad fragment of pYZVδP was excised by digestion with Pad and religation to generate pYZV16 (Figure 12). Digestion of pYZV16 with Asd liberates a δ.2 kB DNA fragment (SEQ ID NO:87) suitable for integration and expression of the Δδ desaturase gene ("MADδ") in the Y. lipolytica genome. Construction Of An Integration Vector For Expression Of The High Affinity Elongase And Δδ Desaturase BsiWI and Hindlll sites were introduced into pYδδ (containing the coding region of the M. alpina high affinity PUFA elongase; described in Example 2) by site-directed mutagenesis using YL61/YL62 (SEQ ID NOs:88 and 89) and YL69/YL70 (SEQ ID NOs:90 and 91) primer pairs, respectively, to generate pYδδBH (Figure 13; elongase gene labeled as "EL"). The 1.7 kB BsiWI/Hindlll fragment of pYδδBH, which contains the TEF::EL::XPR chimeric gene, was ligated into the BsiWI/Hindlll site of pYZMδCHPP (construction described in Figure δ) to generate pYZMδEL (Figure 13). This plasmid is suitable for integration and coordinate expression of the M. alpina Δδ desaturase and high affinity PUFA elongase genes in Y. lipolytica. Construction Of An Integration Vector For Expression Of The Δ6 Desaturase, High Affinity Elongase And Δδ Desaturase

Pad and Clal sites were introduced into pYδ4 (containing the M. alpina Δ6 desaturase; described in Example 2) by site-directed mutagenesis using YL77/YL78 (SEQ ID NOs:92 and 93) and

YL79A/YL60A (SEQ ID NOs:94 and 96) primer pairs, respectively, to generate pY64PC (Figure 13; Δ6 desaturase gene labeled as "MAD6"). The 2 kB Clal/Pacl DNA fragment of pYδ4PC, which contains the TEF::MAD6::XPR chimeric gene, was ligated into the Clal/Pacl sites of pYZMδEL to generate pYZMδELδ (Figure 13). This plasmid is suitable for integration and coordinate expression of the M. alpina Δ6 desaturase, Δ5 desaturase and high affinity PUFA elongase genes in the Y. lipolytica genome. Construction Of A DNA Fragment Suitable For Integration Into The Yarrowia Genome, For Expression Of The Δ6 Desaturase, PUFA Elongase And Δδ Desaturase

The plasmid pYZV16 (construction described in Figure 12) was used for construction of plasmids containing multiple expression cassettes. First, the 3.δ kB BsiWI/Pacl fragment of pYZV16 was ligated to the 7.9 kB BsiWI/Pacl fragment of pYZMδEL6 (construction described in Figure 13) to generate pYZV6EL6 (Figure 14). Digestion of pYZVδELδ with As liberates a 8.9 kB DNA fragment (SEQ ID NO:96) suitable for integration and coordinate expression of the Δ6 desaturase, PUFA elongase and Δδ desaturase genes in the Y. lipolytica genome. Construction Of A DNA Fragment Suitable For Integration Into The Yarrowia Genome, For Expression Of The Δ6 Desaturase, PUFA Elongase, Δδ Desaturase And Δ17 Desaturase

As described in Example 3, the synthetic^'s, diclina Δ17 desaturase gene was inserted into the Ncol/Notl sites of pYδ-13 to generate pYSD17S (Figure 9A). Clal and Pmel sites were introduced into pYSD17S by site-directed mutagenesis using YL101/YL102 (SEQ ID NOs:97 and 98) and YL103/YL104 (SEQ ID NOs:99 and 100) primer pairs, respectively, to generate pYSD17SPC (Figure 14).

The 347 bp Clal/Pmel fragment of pYZV6EL6 (Figure 14) was exchanged with the 1760 bp Clal/Pmel fragment from pYSD17SPC containing the Δ17 desaturase expression cassette to generate pYZVδE6/17. Digestion of pYZV6E6/17 with Asd liberates a 10.3 kB DNA fragment (SEQ ID NO: 101) suitable for integration and coordinate expression of the Δ6 desaturase, PUFA elongase, Δ5 desaturase and Δ17 desaturase genes in the Y. lipolytica genome.

EXAMPLE 5 Biosynthesis Of ω-6 Fattv Acids In Yarrowia lipolytica Transformants pYZV6EL6 (from Example 4, containing the Δ6 desaturase, PUFA elongase and Δδ desaturase genes) was digested with the Asd restriction endonuclease and transformed into Yarrowia lipolytica according to the methodology described in Example 2. Of 62 transformants selected on minimal media lacking leucine, 34 could not grow on media also lacking uracil, suggesting that 66% of the transformants contained the 8.9 kB multi-gene expression cassette integrated into the targeted Yarrowia lipolytica URA3 locus. Transformants from single colonies were inoculated in minimal media lacking leucine and were incubated at 30 °C for up to 4δ hr.

The cells were collected by centrifugation, lipids were extracted, and fatty acid methyl esters were prepared by transesterification and subsequently analyzed with a Hewlett-Packard 6690 GC (according to the methodology described in the General Methods). GC analyses showed the presence of arachidonic acid (ARA) in the transformants containing the 3 chimeric genes (Figure 16), but not in the wild type Yarrowia control strain. These data confirmed that Yarrowia lipolytica was engineered to produce ARA, an ω-6 fatty acid.

EXAMPLE 6 Biosynthesis Of ω-3 Fattv Acids In Yarrowia lipolytica Transformants

In a manner similar to that in Example 5, pYZV6E6/17 (from Example 4, containing the Δ6 desaturase, PUFA elongase, Δδ desaturase and Δ17 desaturase) was digested with the Asd restriction endonuclease and transformed into Yarrowia lipolytica (ATCC #76962). Of 133 transformants selected on minimal media lacking leucine, 89 could not grow on media also lacking uracil, suggesting that 67% of the transformants contained the 10.3 kB multi-gene expression cassette integrated into the targeted Yarrowia lipolytica URA3 locus.

GC analyses (according to the methodology described in the General Methods) showed the presence of eicosapentaenoic acid (EPA) in the transformants containing the 4 chimeric genes (Figure 16), but not in the wild-type Yarrowia control strain. These data confirmed that Yarrowia lipolytica was engineered to produce EPA, an ω-3 fatty acid.

Claims

CLAIMS What is claimed is:

1. A method for producing omega-3 or omega-6 fatty acids comprising: a) providing an oleaginous yeast comprising the genes encoding a functional ω-3/ ω-6 fatty acid biosynthetic pathway; b) growing the yeast of step (a) in the presence of a fermentable carbon source whereby one or more ω-3 or ω-6 fatty acids are produced; and c) optionally recovering the ω-3 or ω-6 fatty acid.

2. A method according to Claim 1 wherein the genes encoding a functional ω-3/ω-6 fatty acid biosynthetic pathway are selected from the group consisting of: Δ12 desaturase, Δ6 desaturase, elongase, Δ5 desaturase, Δ17 desaturase, Δ1δ desaturase, Δ9 desaturase and Δ4 desaturase.

3. A method according to Claim 1 wherein the fermentable carbon source is selected from the group consisting of monosaccharides, oligosaccharides, polysaccharides, monoglycerides, diglycerides, triglycerides, methanol and carbon-containing amines.

4. A method according to Claim 1 wherein the oleaginous yeast is selected from the group consisting of Yarrowia, Candida, Rhodotorula, Rhodosporidium, Cryptococcus, Trichosporon and Lipomyces. δ. A method according to Claim 4 wherein the oleaginous yeast is Yarrowia lipolytica.

6. A method for producing linoleic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ12 desaturase polypeptide; and (ii) a source of desaturase substrate consisting of oleic acid; b) growing the yeast of step (a) in the presence of a suitable carbon source wherein the gene encoding a Δ12 desaturase polypeptide is expressed and the oleic acid is converted to linoleic acid; and c) optionally recovering the linoleic acid of step (b).

7. A method for producing γ-linolenic acid comprising: a) providing an oleaginous yeast comprising: (i) a gene encoding a Δ6 desaturase polypeptide; and (ii) a source of desaturase substrate consisting of linoleic acid; b) growing the yeast of step (a) in the presence of a suitable carbon source wherein the gene encoding a Δ6 desaturase polypeptide is expressed and the linoleic acid is converted to γ-linolenic acid; and c) optionally recovering the γ-linolenic acid of step (b).

8. A method for producing stearidonic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ6 desaturase polypeptide; and (ii) a source of desaturase substrate consisting of α- linoleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon substrate wherein the gene encoding a Δ6 desaturase polypeptide is expressed and the α- linoleic acid is converted to stearidonic acid; and c) optionally recovering the stearidonic acid of step (b).

9. A method for producing α-linoleic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ1δ desaturase polypeptide; and (ii) a source of desaturase substrate consisting of linoleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a

Δ1δ desaturase polypeptide is expressed and the linoleic acid is converted to α-linoleic acid; and c) optionally recovering the α-linoleic acid of step (b). 10. A method for producing dihomo-γ- linoleic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a C_18/20 elongase polypeptide; and (ii) a source of elongase substrate consisting of γ- linolenic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding an elongase polypeptide is expressed and the γ-linolenic acid is converted to dihomo-γ- linoleic acid; and c) optionally recovering the dihomo-γ- linoleic acid of step (b).

11. A method for producing eicosatetraenoic acid comprising: a) providing an oleaginous yeast comprising: (i) a gene encoding a C_18/20 elongase polypeptide; and

(ii) a source of elongase substrate consisting of stearidonic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding an elongase polypeptide is expressed and the stearidonic acid is converted to eicosatetraenoic acid; and c) optionally recovering the eicosatetraenoic acid of step (b).

12. A method for producing docosapentaenoic acid comprising: a) providing an oleaginous yeast comprising: (i) a gene encoding an a C_20/22 elongase polypeptide; and (ii) a source of elongase substrate consisting of eicosapentaenoic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding an elongase polypeptide is expressed and the eicosapentaenoic acid is converted to docosapentaenoic acid; and c) optionally recovering the docosapentaenoic acid of step (b).

13. A method for producing arachidonic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δδ desaturase polypeptide; and (ii) a source of desaturase substrate consisting of dihomo-γ- linoleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a Δδ desaturase polypeptide is expressed and the dihomo- γ- linoleic acid is converted to arachidonic acid; and c) optionally recovering the arachidonic acid of step (b).

14. A method for producing eicosapentaenoic acid comprising: a) providing an oleaginous yeast comprising: (i) a gene encoding a Δδ desaturase polypeptide; and (ii) a source of desaturase substrate consisting of eicosatetraenoic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a

Δδ desaturase polypeptide is expressed and the eicosatetraenoic acid is converted to eicosapentaenoic acid; and c) optionally recovering the eicosapentaenoic acid of step (b) 15. A method for producing docosahexaenoic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ4 desaturase polypeptide; and (ii) a source of desaturase substrate consisting of docosapentaenoic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a Δ4 desaturase polypeptide is expressed and the docosapentaenoic acid is converted to docosahexaenoic acid; and c) optionally recovering the docosahexaenoic acid of step

(b).

16. A method for producing eicosatetraenoic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ17 desaturase polypeptide; and (ii) a source of desaturase substrate consisting of dihomo-γ- linoleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a Δ17 desaturase polypeptide is expressed and dihomo-γ- linoleic acid is converted to eicosatetraenoic acid; and c) optionally recovering the eicosatetraenoic acid of step (b).

17. A method for producing eicosapentaenoic acid comprising: a) providing an oleaginous yeast comprising:

(i) a gene encoding a Δ17 desaturase polypeptide; and (ii) a source desaturase substrate consisting of arachidonic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the gene encoding a Δ17 desaturase polypeptide is expressed and arachidonic acid is converted to eicosapentaenoic acid; and c) optionally recovering the eicosapentaenoic acid of step

(b). 18. A method according to any one of Claims 6-17 wherein the source of desaturase or elongase substrate is endogenous to the oleaginous yeast. 19. A method according to any one of Claims 6-17 wherein the source of desaturase or elongase substrate is exogenous to the oleaginous yeast.

20. A method according to any one of Claims 6-17 wherein the oleaginous yeast is selected from the group consisting of Yarrowia, Candida, Rhodotorula, Rhodosporidium, Cryptococcus, Trichosporon and Lipomyces.

21. A method according to Claim 20 wherein the oleaginous yeast is Yarrowia lipolytica.

22. A method according to either of Claims 7 or 8, wherein the gene encoding a Δ6 desaturase polypeptide is selected from the group consisting of:

(a) an isolated nucleic acid molecule encoding the amino acid sequence as set forth in SEQ ID NO:2; and

(b) an isolated nucleic acid molecule that hybridizes with (a) under the following hybridization conditions: 0.1X SSC,

0.1% SDS, 66 °C and washed with 2X SSC, 0.1% SDS followed by 0.1 X SSC, 0.1 % SDS.

23. A method according to any of Claims 10, 11 and 12, wherein the gene encoding an elongase polypeptide is selected from the group consisting of:

(a) an isolated nucleic acid molecule encoding the amino acid sequence as set forth in SEQ ID NO:8; and

(b) an isolated nucleic acid molecule that hybridizes with (a) under the following hybridization conditions: 0.1X SSC, 0.1 % SDS, 65 °C and washed with 2X SSC, 0.1 % SDS followed by 0.1 X SSC, 0.1% SDS.

24. A method according to any of Claims 13 or 14, wherein the gene encoding a Δδ desaturase polypeptide is selected from the group consisting of:

(a) an isolated nucleic acid molecule encoding the amino acid sequence selected from the group consisting of SEQ ID

NO:4, SEQ ID NO:11δ, SEQ ID NO:119 and SEQ ID NO:123; and

(b) an isolated nucleic acid molecule that hybridizes with (a) under the following hybridization conditions: 0.1X SSC, 0.1% SDS, 6δ °C and washed with 2X SSC, 0.1% SDS followed by 0.1X SSC, 0.1% SDS. 26. A method according to any of Claims 16 or 17, wherein the gene encoding a Δ17 desaturase polypeptide is selected from the group consisting of: (a) an isolated nucleic acid molecule encoding the amino acid sequence as set forth in SEQ ID NO:6; and (b) an isolated nucleic acid molecule that hybridizes with (a) under the following hybridization conditions: 0.1X SSC, 0.1% SDS, 66 °C and washed with 2X SSC, 0.1% SDS followed by 0.1X SSC, 0.1% SDS.

26. A method according to any of Claims 5 or 21 , wherein the host cell is selected from the group consisting of Yarrowia lipolytica ATCC #20362, Yarrowia lipolytica ATCC #6862, Yarrowia lipolytica ATCC #18944, Yarrowia lipolytica ATCC #76982 and Yarrowia lipolytica LGAM S(7)1.

27. A method for producing eicosapentaenoic acid comprising: a) providing an oleaginous yeast comprising:

(i) genes encoding enzymes in the ω-3/ ω-6 fatty acid biosynthetic pathway; and (ii) an endogenous source of oleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the genes encoding the enzymes in the ω-3/ ω-6 fatty acid biosynthetic pathway are expressed and the oleic acid is converted to eicosapentaenoic acid; and c) optionally recovering the eicosapentaenoic acid of step (b).

76

28. A method for producing docosahexaenoic acid comprising: a) providing an oleaginous yeast comprising:

(i) genes encoding enzymes in the ω-3/ ω-6 fatty acid biosynthetic pathway; and (ii) an endogenous source of oleic acid; b) growing the yeast of step (a) in the presence of a suitable fermentable carbon source wherein the genes encoding the enzymes in the ω-3/ ω-6 fatty acid biosynthetic pathway are expressed and the oleic acid is converted to docosahexaenoic acid; and c) optionally recovering the docosahexaenoic acid of step (b).

29. A transformed oleaginous yeast comprising genes encoding enzymes of the ω-3/ ω-6 fatty acid biosynthetic pathway. 30. An oleaginous yeast according to Claim 29 selected from the group consisting of Yarrowia, Candida, Rhodotorula, Rhodosporidium,

Cryptococcus, Trichosporon and Lipomyces.

31. An oleaginous yeast according to Claim 30 wherein the yeast is

Yarrowia lipolytica. 32. A microbial oil produced by the method of any one of

Claims 1-28.