AU2005235627B2 - Synthesis of long-chain polyunsaturated fatty acids by recombinant cells - Google Patents

Abstract

The present invention relates to methods of synthesizing long-chain polyunsaturated fatty acids, especially eicosapentaenoic acid, docosapentaenoic acid and docosahexaenoic acid, in recombinant cells such as yeast or plant cells. Also provided are recombinant cells or plants which produce long-chain polyunsaturated fatty acids. Furthermore, the present invention relates to a group of new enzymes which possess desatorase or elongase activity that can be used in methods of synthesizing long-chain polyunsaturated fatty acids.

Description

WO 2005/103253 PCT/AU2005/000571 SYNTHESIS OF LONG-CHAIN POLYUNSATURATED FATTY ACIDS BY RECOMBINANT CELLS FIELD OF THE INVENTION The present invention relates to methods of synthesizing long-chain polyunsaturated fatty acids, especially eicosapentaenoic acid, docosapentaenoic acid and docosahexaenoic acid, in recombinant cells such as yeast or plant cells. Also provided are recombinant cells or plants which produce long-chain polyunsaturated fatty acids. Furthermore, the present invention relates to a group of new enzymes which possess desaturase or elongase activity that can be used in methods of synthesizing long-chain polyunsaturated fatty acids. BACKGROUND OF THE INVENTION Omega-3 long-chain polyunsaturated fatty acid(s) (LC-PUFA) are now widely recognized as important compounds for human and animal health. These fatty acids may be obtained from dietary sources or by conversion of linoleic (LA, omega-6) or aX linolenic (ALA, omega-3) fatty acids, both of which are regarded as essential fatty acids in the human diet. While humans and many other vertebrate animals are able to convert LA or ALA, obtained from plant sources, to LC-PUFA, they carry out this conversion at a very low rate. Moreover, most modem societies have imbalanced diets in which at least 90% of polyunsaturated fatty acid(s) (PUFA) consist of omega-6 fatty acids, instead of the 4:1 ratio or less for omega-6:omega-3 fatty acids that is regatded as ideal (Trautwein, 2001). The immediate dietary source of LC-PUFA such as eicosapentaenoic acid (EPA, 20:5) and docosahexaenoic acid (DHA, 22:6) for humans is mostly from fish or fish oil. Health professionals have therefore recommended the regular inclusion of fish containing significant levels of LC-PUFA into the human diet. Increasingly, fish-derived LC-PUFA oils are being incorporated into food products and in infant formula. However, due to a decline in global and national fisheries, alternative sources of these beneficial health enhancing oils are needed. Inclusion of omega-3 LC-PUFA such as EPA and DHA in the human diet has been linked with numerous health-related benefits. These include prevention or reduction of coronary heart disease, hypertension, type-2 diabetes, renal disease, rheumatoid arthritis, ulcerative colitis and chronic obstructive pulmonary disease, and aiding brain development and growth (Simopoulos, 2000). More recently, a number of studies have also indicated that omega-3 PUFA-may be beneficial in infant nutrition and development WO 2005/103253 PCT/AU2005/000571 2 and against various mental disorders such as schizophrenia, attention deficit hyperactive disorder and Alzheimer's disease. Higher plants, in contrast to animals, lack the capacity to synthesise polyunsaturated fatty acids with chain lengths longer than 18 carbons. In particular, crop and horticultural plants along with other angiosperms do not have the enzymes needed to synthesize the longer chain omega-3 fatty acids such as EPA, DPA and DHA that are derived from ALA. An important goal in plant biotechnology is therefore the engineering of crop plants, particularly oilseed crops, that produce substantial quantities of LC-PUFA, thus providing an alternative source of these compounds. Pathways of LC-PUFA synthesis Biosynthesis of LC-PUFA from linoleic and ca-linolenic fatty acids in organisms such as microalgae, mosses and fungi may occur by a series of alternating oxygen dependent desaturations and elongation -reactions as shown schematically in Figure 1. In one pathway (Figure 1, II), the desaturation reactions are catalysed by A6, A5, and A4 desaturases, each of which adds an additional double bond into the fatty acid carbon chain, while each of a A6 and a A5 elongase reaction adds a two-carbon unit to lengthen the chain. The conversion of ALA to DHA in these organisms therefore requires three desaturations and two elongations. Genes encoding the enzymes required for the production of DHA in this aerobic pathway have been cloned from various microorganisms and lower plants including microalgae, mosses, fungi. Genes encoding some of the enzymes including one that catalyses the fifth step, the A5 elongase, have been isolated from vertebrate animals including mammals (reviewed in Sayanova and Napier, 2004). However, the A5 elongase isolated from human cells is not specific for the EPA to DPA reaction, having a wide specificity for fatty acid substrates (Leonard et al., 2002). Alternative routes have been shown to exist for two sections of the ALA to DHA pathway in some groups of organisms. The conversion of ALA to ETA may be carried out by a combination of a A9 elongase and a A8 desaturase (the so-called A8 desaturation route, see Figure 1, IV) in certain protists and thraustochytrids, as evidenced by the isolated of genes encoding such enzymes (Wallis and Browse, 1999; Qi et al., 2002). In mammals, the so-called "Sprecher" pathway converts DPA to DHA by three reactions, independent of a A4 desaturase (Sprecher et al., 1995). Besides these desaturase/elongase systems, EPA and DHA can also be synthesized through an anaerobic pathway in a number of organisms such as Shewanella, Mortiella and Schizhochytriumn (Abbadi et al., 2001). The operons encoding these polyketide WO 2005/103253 PCT/AU2005/000571 3 synthase (PKS) enzyme complexes have been cloned from some bacteria (Morita et al., 2000; Metz et al., 2001; Tanaka et al., 1999; Yazawa, 1996; Yu et al., 2000; WO 00/42195). The EPA PKS operon isolated from Shewanella spp has been expressed in Synechococcus allowing it to synthesize EPA (Takeyama et al., 1997). The genes encoding these enzymes are arranged in relatively large operons, and their expression in transgenic plants has not been reported. Therefore it remains to be seen if the anaerobic PKS-like system is a possible alternative to the more classic aerobic desaturase/elongase for the transgenic synthesis of LC-PUFA. Desaturases The desaturase enzymes that have been shown to participate in LC-PUFA biosynthesis all belong to the group of so-called "front-end" desaturases which are characterised by the presence of a cytochrome b 5 domain at the N-terminus of each protein. The cyt b 5 domain presumably acts as a receptor of electrons required for desaturation (Napier et al., 1999; Sperling and Heinz, 2001). The enzyme A5 desaturase catalyses the further desaturation of C20 LC-PUFA leading to arachidonic acid (ARA, 20:4co6) and EPA (20:5o3). Genes encoding this enzyme have been isolated from a number of organisms, including algae (Thraustochytrium sp. Qiu et al., 2001), fungi (M. alpine, Pythium irregulare, Michaelson et al., 1998; Hong et al., 2002), Caenorhabditis elegans and mammals. A gene encoding a bifunctional A5-/A6- desaturase has also been identified from zebrafish (Hasting et al., 2001). The gene encoding this enzyme might represent an ancestral form of the "front-end desaturase" which later duplicated and evolved distinct functions. The last desaturation step to produce DHA is catalysed by a A4 desaturase and a gene encoding this enzyme has been isolated from the freshwater protist species Euglena gracilis and the marine species Thraustochytrium sp. (Qiu et al., 2001; Meyer et al., 2003). Elongases Several genes encoding PUFA-elongation enzymes have also been isolated (Sayanova and Napier, 2004). The members of this gene family were unrelated to the elongase genes present in higher plants, such as FAEl of Arabidopsis, that are involved in the extension of saturated and monounsaturated fatty acids. An example of the latter is erucic acid (22:1) in Brassicas. In some protist species, LC-PUFA are synthesized by elongation of linoleic or a-linolenic acid with a C2 unit, before desaturation with A8 desaturase (Figure 1 part IV; "A8-desaturation" pathway). A6 desaturase and A6 WO 2005/103253 PCT/AU2005/000571 4 elongase activities were not detected in these species. Instead, a A9-elongase activity would be expected in such organisms, and in support of this, a C18 A9-elongase gene has recently been isolated from Isochrysis galbana (Qi et al., 2002). Engineered production ofLC-PUFA Transgenic oilseed crops that are engineered to produce major LC-PUFA by the insertion of these genes have been suggested as a sustainable source of nutritionally important fatty acids. However, the requirement for coordinate expression and activity of five new enzymes encoded by genes from possibly diverse sources has made this goal difficult to achieve and the proposal remained speculative until now. The LC-PUFA oxygen-dependent biosynthetic pathway to form EPA (Figure 1) has been successfully constituted in yeast by the co-expression of a A6-elongase with A6 and A5 fatty acid desaturases, resulting in small but significant accumulation of ARA and EPA from exogenously supplied linoleic and a-linolenic acids (Beaudoin et al., 2000; Zank et al., 2000). This demonstrated the ability of the genes belonging to the LC-PUFA synthesis pathway to function in heterologous organisms. However, the efficiency of producing EPA was very low. For example, three genes obtained from C. elegans, Borago officinalis and Mortierella alpina were expressed in yeast (Beaudoin et al., 2000). When the transformed yeast were supplied with 18:2o-3 (LA) or 18:3o-3 (ALA), there was slight production of 20:4co-6 or 20:5o-3, at conversion efficiencies of 0.65% and 0.3%, respectively. Other workers similarly obtained very low efficiency production of EPA by using genes expressing two desaturases and one elongase in yeast (Domergue et al., 2003a; Zank et al., 2002). There remains, therefore, a need to improve the efficiency of production of EPA in organisms such as yeast, let alone the production of the C22 PUFA which requires the provision of additional enzymatic steps. Some progress has been made in the quest for introducing the aerobic LC-PUFA biosynthetic pathway into higher plants including oilseed crops (reviewed by Sayanova and Napier, 2004; Drexler et al., 2003; Abbadi et al., 2001). A gene encoding a A6-fatty acid desaturase isolated from borage (Borago officinalis) was expressed in transgenic tobacco and Arabidopsis, resulting in the production of GLA (18:3(06) and SDA (18:4o3), the direct precursors for LC-PUFA, in the transgenic plants (Sayanova et al., 1997; 1999). However, this provides only a single, first step. Domergue et al. (2003a) used a combination of three genes, encoding A6- and A5 fatty acid desaturases and a A6-elongase in both yeast and transgenic linseed. The desaturase genes were obtained from the diatom Phaeodactylum tricornutum and the elongase gene from the moss Physcomitrella patens. Low elongation yields were WO 2005/103253 PCT/AU2005/000571 5 obtained for endogenously produced A6-fatty acids in yeast cells (i.e. combining the first and second enzymatic steps), and the main C20 PUFA product formed was 2 0: 2 Aru, representing an unwanted side reaction. Domergue et al. (2003a) also state, without presenting data, that the combination of the three genes were expressed in transgenic linseed which consequently produced ARA and EPA, but that production was inefficient. They commented that the same problem as had been observed in yeast existed in the seeds of higher plants and that the "bottleneck" needed to be circumvented for production of LC-PUFA in oil seed crops. WO 2004/071467 (DuPont) reported the expression of various desaturases and elongases in soybean cells but did not show the synthesis of DHA in regenerated plants or in seeds. Abbadi et al. (2004) described attempts to express combinations of desaturases and elongases in transgenic linseed, but achieved only low levels of synthesis of EPA. Abbadi et al. (2004) indicated that their low levels of EPA production were also due to an unknown "bottleneck". Qi et al. (2004) achieved synthesis in leaves but did not report results in seeds. This is an important issue as the nature of LC-PUFA synthesis can vary between leaves and seeds. In particular, oilseeds store lipid in seeds mostly as TAG while leaves synthesize the lipid mostly as phosphatidyl lipids. Furthermore, Qi et al. (2004) only produced AA and EPA. As a result, there is a need for further methods of producing long-chain polyunsaturated, particularly EPA, DPA and DHA, in recombinant cells. SUMMARY OF THE INVENTION In a first aspect, the present invention provides a recombinant cell which is capable of synthesising a long chain polyunsaturated fatty acid(s) (LC-PUFA), comprising one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, wherein said recombinant cell is derived from a cell that is not capable of synthesising said LC-PUFA. In a second aspect, the present invention provides a recombinant cell with an enhanced capacity to synthesize a LC-PUFA relative to an isogenic non-recombinant cell, comprising one or more polynucleotides which encode at least two enzymes each of 6 which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of expressing said polynucleotides in said recombinant cell. 5 In one embodiment, at least one of the enzymes is a A5 elongase. In another aspect, the present invention provides a recombinant plant cell which is capable of synthesising a long chain polyunsaturated fatty acid(s) (LC-PUFA) derived from a cell that is not capable of synthesising said LC-PUFA, or a recombinant cell with an enhanced capacity to synthesize an LC-PUFA relative to an isogenic non 10 recombinant cell, the cell comprising one or more exogenous polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of 15 said polynucleotides in the cell, wherein at least one of the enzymes is a A5 elongase which catalyses the conversion of EPA to DPA in the cell, and wherein the plant cell is homozygous for at least one of the exogenous polynucleotides, and wherein the A5 desaturase catalyses the conversion of ETA to EPA in the cell, and wherein the plant cell comprises DPA which is synthesized from EPA in the cell, and wherein the 20 efficiency of conversion of EPA to DPA in the cell is at least 5%. The present inventors are the first to identify an enzyme which has greater A5 elongase activity than A6 elongase activity. As a result, this enzyme provides an efficient means of producing DPA in a recombinant cell as the A5 elongation of EPA is favoured over the A6 elongation of SDA. Thus, in an embodiment, the A5 elongase is 25 relatively specific, that is, where the A5 elongase also has A6 elongase activity the elongase is more efficient at synthesizing DPA from EPA than it is at synthesizing ETA from SDA. In another embodiment, the AS elongase comprises i) an amino acid sequence as provided in SEQ ID NO:2, 30 ii) an amino acid sequence which is at least 50%, more preferably at least 80%, even more preferably at least 90%, identical to SEQ ID NO:2, or iii) a biologically active fragment of i) or ii). In another embodiment, the A5 elongase can be purified from algae. In another embodiment, at least one of the enzymes is a A9 elongase. 35 The present inventors are the first to identify an enzyme which has both A9 elongase activity and A6 elongase activity. When expressed in a cell with a A6 desaturase and a A8 desaturase this enzyme can use the two available pathways to produce ETA from ALA, DGLA from LA, or both (see Figure 1), thus increasing the 6A efficiency of ETA and/or DGLA production. Thus, in an embodiment, the A9 elongase also has A6 elongase activity. Preferably, the A9 elongase is more efficient at synthesizing ETrA from ALA than it is at synthesizing ETA from SDA. Furthermore, in another embodiment the A9 elongase is able to elongate SDA to ETA, GLA to DGLA, or both, in a yeast cell. In a further embodiment, the A9 elongase comprises i) an amino acid sequence as provided in SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, ii) an amino acid sequence which is at least 50%, more preferably at least 80%, even more preferably at least 90%, identical to SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, or iii) a biologically active fragment of i) or ii). Preferably, the A9 elongase can be purified from algae or fungi.

WO 2005/103253 PCT/AU2005/000571 7 It is well known in the art that the greater the number of transgenes in an organism, the greater the likelihood that at least one fitness parameter of the organism, such as expression level of at least one of the transgenes, growth rate, oil production, reproductive capacity etc, will be compromised. Accordingly, it is desirable to minimize the number of transgenes in a recombinant cell. To this end, the present inventors have devised numerous strategies for producing LC-PUFA's in a cell which avoid the need for a gene to each step in the relevant pathway. Thus, in another embodiment, at least one of the enzymes is a A5/A6 bifunctional desaturase or a A5/A6 bifunctional elongase. The A5/A6 bifunctional desaturase may be naturally produced by a freshwater species of fish. In a particular embodiment, the A5/A6 bifunctional desaturase comprises i) an amino acid sequence as provided in SEQ ID NO:15, ii) an amino acid sequence which is at least 50%, more preferably at least 80%, even more preferably at least 90%, identical to SEQ ID NO: 15, or iii) a biologically active fragment of i) or ii). Preferably, the A5/A6 bifunctional desaturase is naturally produced by a freshwater species of fish. Preferably, the A5/A6 bifunctional elongase comprises i) an amino acid sequence as provided in SEQ ID NO:2 or SEQ ID NO: 14, ii) an amino acid sequence which is at least 50%, more preferably at least 80%, even more preferably at least 90%, identical to SEQ ID NO:2 or SEQ ID NO: 14, or iii) a biologically active fragment of i) or ii). In another embodiment, at least one of the enzymes is a A5 desaturase. In a further embodiment, at least one of the enzymes is a A8 desaturase. In another embodiment, the LC-PUFA is docosahexaenoic acid (DHA). Preferably, the introduced polynucleotide(s) encode three or four enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, or A4 desaturase. More preferably, the enzymes are any one of the following combinations; i) a A5/A6 bifunctional desaturase, a A5/A6 bifunctional elongase, and a A4 desaturase, ii) a A5/A6 bifunctional desaturase, a A5 elongase, a A6 elongase, and a A4 desaturase, or iii) a A5 desaturase, a A6 desaturase, a A5/A6 bifunctional elongase, and a A4 desaturase.

WO 2005/103253 PCT/AU2005/000571 8 In another embodiment, the LC-PUFA is DHA and the introduced polynucleotide(s) encode five enzymes wherein the enzymes are any one of the following combinations; i) a A4 desaturase, a A5 desaturase, a A6 desaturase, a A5 elongase and a A6 elongase, or ii) a A4 desaturase, a A5 desaturase, a A8 desaturase, a A5 elongase and a A9 elongase. In a further embodiment, the cell is of an organism suitable for fermentation, and the enzymes are at least a A5/A6 bifunctional desaturase, a A5 elongase, a A6 elongase, and a A4 desaturase. In another embodiment, the LC-PUFA is docosapentaenoic acid (DPA). Preferably, the introduced polynucleotide(s) encode two or three enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, or A6 elongase. More preferably, the enzymes are any one of the following combinations; i) a A5/A6 bifunctional desaturase and a A5/A6 bifunctional elongase, ii) a A5/A6 bifunctional desaturase, a A5 elongase,- and a A6 elongase, or iii) a A5 desaturase, a A6 desaturase, and a A5/A6 bifunctional elongase. In a further embodiment, the LC-PUFA is DPA and the introduced polynucleotide(s) encode four enzymes wherein the enzymes are any one of the following combinations; i) a A5 desaturase, a A6 desaturase, a A5 elongase and a A6 elongase, or ii) a A5 desaturase, a A8 desaturase, a A5 elongase and a A9 elongase. In another embodiment, the cell is of an organism suitable for fermentation, and the enzymes are at least a A5/A6 bifunctional desaturase, a A5 elongase, and a A6 elongase. In a further embodiment, the LC-PUFA is eicosapentaenoic acid (EPA). Preferably, the introduced polynucleotide(s) encode a A5/A6 bifunctional desaturase and a A5/A6 bifunctional elongase. In another embodiment, the introduced polynucleotide(s) encode three enzymes wherein the enzymes are any one of the following combinations; i) a A5 desaturase, a A6 desaturase, and a A6 elongase, or ii) a A5 desaturase, a A8 desaturase, and a A9 elongase. Evidence to date suggests that desaturases expressed in at least some recombinant cells, particularly yeast, have relatively low activity. However, the present inventors have identified that this may be a function of the capacity of the desaturase to use acyl- WO 2005/103253 PCT/AU2005/000571 9 CoA as a substrate in LC-PUFA synthesis. In this regard, it has also been determined that desaturase of vertebrate origin are particularly useful for the production of LC-PUFA in recombinant cells, for example, plant cells, seeds, or yeast. Thus, in another preferred embodiment, the recombinant cell comprises either i) at least one A5 elongase catalyses the conversion of EPA to DPA in the cell, ii) at least one desaturase which is able to act on an acyl-CoA substrate, iii) at least one desaturase from a vertebrate or a variant desaturase thereof, or iv) any combination of i), ii) or iii). In a particular embodiment, the A5 elongase comprises i) an amino acid sequence as provided in SEQ ID NO:2, ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:2, or iii) a biologically active fragment of i) or ii). The desaturase able to act on an acyl-CoA substrate or from a vertebrate may be a A5 desaturase, a A6 desaturase, or both. In a particular embodiment, the desaturase comprises i) an amino acid sequence as provided in SEQ ID NO: 16, SEQ ID NO:21 or SEQ ID NO:22, ii) an amino acid sequence which is at least 50% identical to SEQ ID NO: 16, SEQ ID NO:21 or SEQ ID NO:22, or iii) a biologically active fragment of i) or ii). Preferably, the at least one desaturase is naturally produced by a vertebrate. Alternatively, when the cell is a yeast cell, the LC-PUFA is DHA, and the enzymes are at least a A5/A6 bifunctional desaturase, a A5 elongase, a A6 elongase, and a A4 desaturase. In a further alternative, when the cell is a yeast cell, the LC-PUFA is DPA, and the enzymes are at least a A5/A6 bifunctional desaturase, a A5 elongase, and a A6 elongase. Although the cell may be any cell type, preferably, said cell is capable of producing said LC-PUFA from endogenously produced linoleic acid (LA), a-linolenic acid (ALA), or both. More preferably, the ratio of the endogenously produced ALA to LA is at least 1:1 or at least 2:1. In one embodiment, the cell is a plant cell, a plant cell from an angiosperm, an oilseed plant cell, or a cell in a seed. Preferably, at least one promoter is a seed specific promoter. In another embodiment, the cell is of a unicellular microorganism. Preferably, the unicellular microorganism is suitable for fermentation. Preferably, the microorganism is a yeast.

WO 2005/103253 PCT/AU2005/000571 10 In a further embodiment, the cell is a non-human animal cell or a human cell in vitro. In a further embodiment, the recombinant cell produces a LC-PUFA which is incorporated into triacylglycerols in said cell. More preferably, at least 50% of the LC PUFA that is produced in said cell is incorporated into triacylglycerols. In another embodiment, at least the protein coding region of one, two or more of the polynucleotides is obtained from an algal gene. Preferably, the algal gene is from the genus Pavlova such as from the species Pavlova salina. In another aspect, the present invention provides a recombinant cell that is capable of producing DHA from a fatty acid which is ALA, LA, GLA, ARA, SDA, ETA, EPA, or any combination or mixture of these, wherein said recombinant cell is derived from a cell that is not capable of synthesising DHA. In a further aspect, the present invention provides a recombinant cell that is capable of producing DPA from a fatty acid which is ALA, LA, GLA, ARA, SDA, ETA, EPA, or any combination or mixture of these, wherein said recombinant cell is derived from a cell that is not capable of synthesising DPA. In yet a further aspect, the present invention provides a recombinant cell that is capable of producing EPA from a fatty acid which is ALA, LA, GLA, SDA, ETA or any combination or mixture of these, wherein said recombinant cell is derived from a cell that is not capable of synthesising EPA. In another aspect, the present invention provides a recombinant cell that is capable of producing both ETrA from ALA and ETA from SDA, and which produces EPA from a fatty acid which is ALA, LA, GLA, SDA, ETA, or any combination or mixture of these, wherein said recombinant cell is derived from a cell that is not capable of synthesising ETrA, ETA or both. In a further aspect, the present invention provides a recombinant cell of an organism useful in fermentation processes, wherein the cell is capable of producing DPA from LA, ALA, arachidonic acid (ARA), eicosatetraenoic acid (ETA), or any combination or mixture of these, wherein said recombinant cell is derived from a cell that is not capable of synthesising DPA. In another aspect, the present invention provides a recombinant plant cell capable of producing DPA from LA, ALA, EPA, or any combination or mixture of these, wherein the plant cell is from an angiosperm. In an embodiment, the plant cell is also capable of producing DHA. In yet another aspect, the present invention provides a recombinant cell which is capable of synthesising DGLA, comprising a polynucleotide(s) encoding one or both of: WO 2005/103253 PCT/AU2005/000571 11 a) a polypeptide which is an A9 elongase, wherein the A9 elongase is selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, ii) a polypeptide comprising an amino acid sequence which is at least 40% identical to SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, and iii) a biologically active fragment of i) or ii), and/or b) a polypeptide which is an A8 desaturase, wherein the A8 desaturase is selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:1, ii) a polypeptide comprising an amino acid sequence which is at least 40% identical to SEQ ID NO: 1, and iii) a biologically active fragment of i) or ii), wherein the polynucleotide(s) is operably linked to one or more promoters that are capable of directing expression of said polynucleotide(s) in the cell, and wherein said recombinant cell is derived from a cell that is not capable of synthesising DGLA. In an embodiment, the cell is capable of converting DGLA to ARA. In another embodiment, the cell further comprises a polynucleotide which encodes a A5 desaturase, wherein the polynucleotide encoding the A5 desaturase is operably linked to one or more promoters that are capable of directing expression of said polynucleotide in the cell, and wherein the cell is capable of producing ARA. In a particular embodiment, the cell lacks 3 desaturase activity and is not capable of producing ALA. Such cells may be naturally occurring, or produced by reducing the o3 desaturase activity of the cell using techniques well known in the art. Preferably, the cell is a plant cell or a cell of an organism suitable for fermentation. In a further embodiment, a recombinant cell of the invention also possesses the enzyme required to perform the "Sprecher" pathway of converting EPA to DHA. These enzymes may be native to the cell or produced recombinantly. Such enzymes at least include a A7 elongase, A6 desaturase and enzymes required for the peroxisomal p oxidation of tetracosahexaenoic acid to produce DHA. The present inventors have also identified a group of new desaturases and elongases. As a result, further aspects of the invention relate to these enzymes, as well as homologs/variants/derivatives thereof.

12 The polypeptide may be a fusion protein further comprising at least one other polypeptide sequence. The at least one other polypeptide may be a polypeptide that enhances the stability of a polypeptide of the present invention, or a polypeptide that assists in the purification of the fusion protein. Also provided are isolated polynucleotides which, inter alia, encode polypeptides of the invention. In a further aspect, the present invention provides a vector comprising or encoding a polynucleotide according to the invention. Preferably, the polynucleotide is operably linked to a seed specific promoter. In another aspect, the present invention provides a recombinant cell comprising an isolated polynucleotide according to the invention, In a further aspect, the present invention provides a method of producing a cell capable of synthesising one or more LC-PUFA, the method comprising introducing into the cell one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase. A4 desaturase, A9 elongase, or A8 desaturase. wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, in another aspect, the present invention provides a method of producing a recombinant cell with an enhanced capacity to synthesize one or more LC-PUFA, the method comprising introducing into a first cell one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, AS elongase, A6 elongase. A4 desaturase, A9 elongase, or AS desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the recombinant cell, and wherein said recombinant cell has an enhanced capacity to synthesize said one or more LC-PUFA relative to said first cell. In another aspect, the present invention provides a method of producing a plant cell of the invention, the method comprising introducing into the cell one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, AS desaturase, A6 desaturase, A5/A6 bifunctional elongase, AS clongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, wherein at least one of the enzymes is a AS elongase which catalyses the conversion of EPA to DPA in the cell, and wherein 12A the plant cell is homozygous for at least one of the exogenous polynucleotides, and wherein the A5 desaturase catalyses the conversion of ETA to EPA in the cell, and optionally wherein said recombinant cell has an enhanced capacity to synthesize said one or more LC-PUFA relative to said first cell. Naturally, it will be appreciated that each of the embodiments described herein in relation to the recombinant cells of the invention will equally apply to methods for the production of said cells. In a further aspect. the present invention provides a cell produced by a method of the invention. In another aspect, the present invention provides a transgenic plant comprising at least one recombinant cell according to the invention.

WO 2005/103253 PCT/AU2005/000571 13 Preferably, the plant is an angiosperm. More preferably, the plant is an oilseed plant. In a further embodiment, the transgenic plant, or part thereof including a transgenic seed, does not comprise a transgene which encodes an enzyme which preferentially converts an o6 LC-PUFA into an o3 LC-PUFA. In yet a further embodiment, the transgenic plant, or part thereof including a transgenic seed, comprises a transgene encoding a A8 desaturase and/or a A9 elongase. In a further aspect, the present invention provides a method of producing an oilseed, the method comprising i) growing a transgenic oilseed plant according to the invention under suitable conditions, and ii) harvesting the seed of the plant. In a further aspect, the invention provides a part of the transgenic plant of the invention, wherein said part comprises an increased level of LC-PUFA in its fatty acid relative to the corresponding part from an isogenic non-transformed plant. Preferably, said plant part is selected from, but not limited to, the group consisting of: a seed, leaf, stem, flower, pollen, roots or specialised storage organ (such as a tuber). Previously, it has not been shown that LC-PUFA can be produced in plant seeds, nor that these LC-PUFA can be incorporated into plant oils such as triacylglycerol. Thus, in another aspect the present invention provides a transgenic seed comprising a LC-PUFA. Preferably, the LC-PUFA is selected from the group consisting of: i) EPA, ii) DPA, iii) DHA, iv) EPA and DPA, and v) EPA, DHA, and DPA. More preferably, the LC-PUFA is selected from the group consisting of: i) DPA, ii) DHA, or iii) DHA and DPA. Even more preferably, the LC-PUFA is EPA, DHA, and DPA. Preferably, the seed is derived from an isogenic non-transgenic seed which produces LA and/or ALA. More preferably, the isogenic non-transgenic seed comprises a higher concentration of ALA than LA in its fatty acids. Even more preferably, the WO 2005/103253 PCT/AU2005/000571 14 isogenic non-transgenic seed comprises at least about 13% ALA or at least about 27% ALA or at least about 50% ALA in its fatty acid. Preferably, the total fatty acid in the oil of the seed comprises at least 9% C20 fatty acids. Preferably, the seed is derived from an oilseed plant. More preferably, the oilseed plant is oilseed rape (Brassica napus), maize (Zea mays), sunflower (Helianthus annuus), soybean (Glycine max), sorghum (Sorghum bicolor), flax (Linum usitatissimum), sugar (Saccharum officinarum), beet (Beta vulgaris), cotton (Gossypium hirsutum), peanut (Arachis hypogaea), poppy (Papaver somnferum), mustard (Sinapis alba), castor bean (Ricinus communis), sesame (Sesamum indicum), or safflower (Carthamus tinctorius). It is preferred that the seed has a germination rate which is substantially the same as that of the isogenic non-transgenic seed. It is further preferred that the timing of germination of the seed is substantially the same as that of the isogenic non-transgenic seed. Preferably, at least 25%, or at least 50%, or at least 75% of the LC-PUFA in the seed form part of triacylglycerols. Surprisingly, the present inventors have found that transgenic seeds produced using the methods of the invention have levels of ALA and LA which are substantially the same as those of an isogenic non-transgenic seed. As a result, it is preferred that the transgenic seed has levels of ALA and LA which are substantially the same as those of an isogenic non-transgenic seed. Furthermore, it was surprising to note that the levels of monounsaturated fatty acids were decreased in transgenic seeds produced using the methods of the invention. Accordingly, in a further preferred embodiment, the transgenic seed has decreased levels of monounsaturated fatty acids when compared to an isogenic non-transgenic seed. In another aspect, the present invention provides a method of producing a transgenic seed according to the invention, the method comprising i) introducing into a progenitor cell of a seed one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, thereby producing a recombinant progenitor cell, ii) culturing said recombinant progenitor cell to produce a plant which comprises said transgenic seed, and iii) recovering the seed from the plant so produced.

WO 2005/103253 PCT/AU2005/000571 15 In yet a further aspect, the present invention provides a method of producing a transgenic seed comprising cultivating a transgenic plant which produces the transgenic seed of the invention, and harvesting said transgenic seed from the plant. In a further aspect, the invention provides an extract from the transgenic plant of the invention, or a plant part of the invention, or a seed of the invention, wherein said extract comprises an increased level of LC-PUFA in its fatty acid relative to a corresponding extract from an isogenic non-transformed plant. Preferably, the extract is substantially purified oil comprising at least 50% triacylglycerols. In a further aspect, the present invention provides a non-human transgenic animal comprising at least one recombinant cell according to the invention. Also provided is a method of producing a LC-PUFA, the method comprising culturing, under suitable conditions, a recombinant cell according to the invention. In one embodiment, the cell is of an organism suitable for fermentation and the method further comprises exposing the cell to at least one LC-PUFA precursor. Preferably, the LC-PUFA precursor is at least one of linoleic acid or c-linolenic acid. In a particular embodiment, the LC-PUFA precursor is provided in a vegetable oil. In another embodiment, the cell is an algal cell and the method further comprises growing the algal cell under suitable conditions for production of said LC-PUFA. In a further aspect, the present invention provides a method of producing one or more LC-PUFA, the method comprising cultivating, under suitable conditions, a transgenic plant of the invention. In another aspect, the present invention provides a method of producing oil comprising at least one LC-PUFA, comprising obtaining the transgenic plant of the invention, or the plant part of the invention, or the seed of the invention, and extracting oil from said plant, plant part or seed. Preferably, said oil is extracted from the seed by crushing said seed. In another aspect, the present invention provides a method of producing DPA from EPA, the method comprising exposing EPA to a polypeptide of the invention and a fatty acid precursor, under suitable conditions. In an embodiment, the method occurs in a cell which uses the polyketide-like system to produce EPA. In yet another aspect, the present invention provides a fermentation process comprising the steps of: i) providing a vessel containing a liquid composition comprising a cell of the invention and constituents required for fermentation and fatty acid biosynthesis; and WO 2005/103253 PCT/AU2005/000571 16 ii) providing conditions conducive to the fermentation of the liquid composition contained in said vessel. Preferably, a constituent required for fermentation and fatty acid biosynthesis is LA. Preferably, the cell is a yeast cell. In another aspect, the present invention provides a composition comprising a cell of the invention, or an extract or portion thereof comprising LC-PUFA, and a suitable carrier. In another aspect, the present invention provides a composition comprising the transgenic plant of the invention, or the plant part of the invention, or the seed of the invention, or an extract or portion thereof comprising LC-PUFA, and a suitable carrier. In yet another aspect, the present invention provides a feedstuff comprising a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition of the invention. Preferably, the feedstuff at least comprises DPA, wherein at least one enzymatic reaction in the production of DPA was performed by a recombinant enzyme in a cell. Furthermore, it is preferred that the feedstuff comprises at least comprises DHA, wherein at least one enzymatic reaction in the production of DHA was performed by a recombinant enzyme in a cell. In a further aspect, the present invention provides a method of preparing a feedstuff, the method comprising admixing a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition of the invention, with a suitable carrier. Preferably, the feedstuff is for consumption by a mammal or a fish. In a further aspect, the present invention provides a method of increasing the levels of a LC-PUFA in an organism, the method comprising administering to the organism a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition of the invention, or a feedstuff of the invention. Preferably, the administration route is oral. Preferably, the organism is a vertebrate. More preferably, the vertebrate is a human, fish, companion animal or livestock animal.

WO 2005/103253 PCT/AU2005/000571 17 In a further aspect, the present invention provides a method of treating or preventing a condition which would benefit from a LC-PUFA, the method comprising administering to a subject a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition of the invention, or a feedstuff of the invention. Preferably, the condition is arrhythmia's, angioplasty, inflammation, asthma, psoriasis, osteoporosis, kidney stones, AIDS, multiple sclerosis, rheumatoid arthritis, Crohn's disease, schizophrenia, cancer, fetal alcohol syndrome, attention deficient hyperactivity disorder, cystic fibrosis, phenylketonuria, unipolar depression, aggressive hostility, adrenoleukodystophy, coronary heart disease, hypertension, diabetes, obesity, Alzheimer's disease, chronic obstructive pulmonary disease, ulcerative colitis, restenosis after angioplasty, eczema, high blood pressure, platelet aggregation, gastrointestinal bleeding, endometriosis, premenstrual syndrome, myalgic encephalomyelitis, chronic fatigue after viral infections or ocular disease. Whilst providing the subject with any amount of LC-PUFA will be beneficial to the subject, it is preferred that an effective amount to treat the condition is administered. In another aspect, the present invention provides for the use of a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition of the invention, or a feedstuff of the invention, for the manufacture of a medicament for treating or preventing a condition which would benefit from a LC-PUFA. The Caenorhabditis elegans A6 elongase has previously been expressed in yeast and been shown to convert octadecatetraenoic acid to eicosatetraenoic acid. However, the present inventors have surprisingly found that this enzyme also possesses A5 elongase activity, being able to convert eicosapentaenoic acid to docosapentaenoic acid. In a further aspect, the present invention provides a method of producing an unbranched LC-PUFA comprising 22 carbon atoms, the method comprising incubating an unbranched 20 carbon atom LC-PUFA with a polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:2 or SEQ ID NO: 14, ii) a polypeptide comprising an amino acid sequence which is at least 50% identical to SEQ ID NO:2 or SEQ ID NO: 14, and iii) a biologically active fragment of i) or ii), WO 2005/103253 PCT/AU2005/000571 18 wherein the polypeptide also has A6 elongase activity. Preferably, the unbranched LC-PUFA comprising 22 carbon atoms is DPA, and the unbranched 20 carbon atom LC-PUFA is EPA. Preferably, the method is performed within a recombinant cell which produces the polypeptide and EPA. In yet a further aspect, the present invention provides a substantially purified antibody, or fragment thereof, that specifically binds a polypeptide of the invention. In another aspect, the present invention provides a method of identifying a recombinant cell, tissue or organism capable of synthesising one or more LC-PUFA, the method comprising detecting the presence in said cell, tissue or organism of one or more polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, tissue or organism. Preferably, the method comprises a nucleic acid amplification step, a nucleic acid hybridisation step, a step of detecting the presence of a transgene in the cell, tissue or organism, or a step of determining the fatty acid content or composition of the cell, tissue or organism. Preferably, the organism is an animal, plant, angiosperm plant or microorganism. In another aspect, the present invention provides a method of producing DPA from EPA, the method comprising exposing EPA to a A5 elongase of the invention and a fatty acid precursor, under suitable conditions. Preferably, the method occurs in a cell which uses the polyketide-like system to produce EPA. Naturally, recombinant (transgenic) cells, plants, non-human animals comprising a new polynucleotide provided herein may also produce other elongase and/or desaturases such as those defined herein. As will be apparent, preferred features and characteristics of one aspect of the invention are applicable to many other aspects of the invention. Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps. The invention is hereinafter described by way of the following non-limiting Examples and with reference to the accompanying figures.

WO 2005/103253 PCT/AU2005/000571 19 BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS Figure 1. Possible pathways of 03 and o)6 LC-PUFA synthesis. The sectors labelled I, II, III, and IV correspond to the o6 (A6), 03 (A6), o)6 (A8), and 03 (A) pathways, respectively. Compounds in sectors I and III are 06 compounds, while those in sectors II and IV are 03 compounds. "Des" refers to desaturase steps in the pathway catalysed by desaturases as indicated, while "Elo" refers to elongase steps catalysed by elongases as indicated. The thickened arrow indicates the A5 elongase step. The dashed arrows indicate the steps in the "Sprecher" pathway that operates in mammalian cells for the production of DHA from DPA. Figure 2. Distribution of LC-PUFA in microalgal classes. Chlorophyceae and Prasinophyceae are described as "green algae", Eustigmatophyceae as "yellow-green algae", Rhodophyceae as "red algae", and Bacillariophyceae and Prymnesiophyceae as diatoms and golden brown algae. Figure 3. Genetic construct for expression of LC-PUFA biosynthesis genes in plant cells. Figure 4. PILEUP of desaturase enzymes. d8-atg - Pavlova salina A8 desaturase; euglena - AAD45877 (A8 desaturase, Euglena gracilis); rhizopus - AAP83964 (A6 desaturase, Rhizopus sp. NK030037); mucor - BAB69055 (A6 desaturase, Mucor circinelloides); mortierella - AAL73948 (A6 desaturase, Mortierella isabellina); malpina - BAA85588 (A6 desaturase, Mortierella alpina); physcomitrella - CAA1 1032 (A6 acyl lipid desaturase, Physcomitrella patens); ceratadon - CAB94992 (A6 fatty acid acetylenase, Ceratodon purpureus). Figure 5. Southern blot of PCR products, hybridized to Elo1 or Elo2 probes. Figure 6. PILEUP of elongase enzymes. Figure 7. Transgene constructs used to express genes encoding LC-PUFA biosynthetic enzymes in Arabidopsis. The "EPA construct" . pSSP-5/6D.6E (also called pZebdesatCeloPWvec8 in Example 5) (Figure 7A) contained the zebra-fish dual function A5/A6-desaturase (D5/D6Des) and the nematode A6-elongase (D6Elo) both driven by the truncated napin promoter (Fpl), and the hygromycin resistance selectable marker gene (hph) driven by the CaMV-35S (35SP) promoter. The "DHA construct" pXZP355 (Figure 7B) comprised the Pavlova salina A4-desaturase (D4Des) and A5-elongase (D5Elo) genes both driven by the truncated napin promoter (Fpl), and the kanamycin resistance selectable marker gene (nptIl) driven by the nopaline synthase promoter (NosP). All genes were flanked at the 3' end by the nopaline synthase terminator (NosT).

WO 2005/103253 PCT/AU2005/000571 20 Figure 8. A. Gas chromatogram (GLC) showing fatty acid profile for Arabidopsis thaliana line DO11 carrying EPA and DHA gene constructs. B. Mass spectra for EPA and DHA obtained from Arabidopsis thaliana line DO11. Figure 9. Autoradiograms of dot-blot hybridisations carried out under low stringency or high stringency conditions, as described in Example 12, to DNA from various microalgal species indicated at the top, using radiolabelled probes consisting of P. salina LC-PUFA gene coding regions as indicated on the right. Figure 10. Amino acid sequence alignment of A6- and A8-desaturases from higher plants. The amino acid sequences of A6-desaturases from E. plantagineum (EplD6Des) (SEQ ID NO:64), E. gentianoides (EgeD6Des, accession number AY055117) (SEQ ID NO:65), E. pitardii (EpiD6Des, AY055118) (SEQ ID NO:66), Borago officinalis (BofD6Des, U79010) (SEQ ID NO:67) and A8-desaturases from B. officinalis (BofD8Des, AF133728) (SEQ ID NO:68), Helianthus annus (HanD8Des, S68358) (SEQ ID NO:69), and Arabidopsis thaliana (AtD8DesA, AAC62885.1; and AtD8DesB, CAB71088.1) (SEQ ID NO:70 and SEQ ID NO:71 respectively) were aligned by PILEUP (GCG, Wisconsin, USA). HBI, HBII1, HBIII are three conserved histidine boxes. Fl and R1 are the corresponding regions for the degenerate primers EpD6Des-F1 and EpD6Des-R1 used to amplify the cDNA. The N-terminal cytochrome b 5 domain with conserved HPGG motif is also indicated. Figure 11. Variant EplD6Des enzymes isolated and representative enzyme activities. EplD6Des with cytochrome b5, histidine boxes I, II, and III are shown as b5, HBI, HBII, HBIII respectively. Variants isolated are shown in panel A in the format: wild-type amino acid - position number - variant amino acid. Empty diamonds indicate mutants with significant reduction of enzyme activity, while solid diamonds indicate the variants with no significant effect on enzyme activity. Panel B shows the comparison of GLA and SDA production in transgenic tobacco leaves from two variants with that of wild-type enzyme. Figure 12. Alternative pathways for synthesis of the o3 LC-PUFA SDA (18:4), EPA (20:5) and DHA (22:6) from ALA (18:3). Desaturases, elongases and acyltransferases are shown as solid, open and dashed arrows respectively. Chain elongation occurs only on acyl-CoA substrates, whereas desaturation can occur on either acyl-PC [A&B] or acyl CoA substrates [C]. The acyl-PC or acyl-CoA substrate preference of the final A4 desaturase step has not yet been determined. Pathways involving acyl-PC desaturases require acyltransferase-mediated shuttling of acyl groups between the PC and CoA substrates. Panels A and B show the "A6 pathway" and "A8 pathway" variants of the acyl-PC desaturase pathway respectively. Panel C shows the pathway expressed in the current study in which the acyl-CoA A6 and A5 desaturase activities were encoded by the WO 2005/103253 PCT/AU2005/000571 21 zebra-fish A6/A5 dual-function desaturase. Synthesis of o6 LC-PUFA such as ARA (20:4) occurs by the same set of reactions but commencing with LA (18:2) as the initial substrate. Figure 13. Growth rates of Synechococcus 7002 at 22*C, 250C, 300C. Figure 14. Synechococcus 7002 linoleic and linolenic acid levels at various growth temperatures. KEY TO THE SEQUENCE LISTING SEQ ID NO:1 - A8 desaturase from Pavlova salina. SEQ ID NO:2 - A5 elongase from Pavlova salina. SEQ ID NO:3 - A9 elongase from Pavlova salina. SEQ ID NO:4 - A4 desaturase from Pavlova salina. SEQ ID NO:5 - cDNA encoding open reading frame of A8 desaturase from Pavlova salina. SEQ ID NO:6 - Full length cDNA encoding of A8 desaturase from Pavlova salina. SEQ ID NO:7 - cDNA encoding open reading frame of A5 elongase from Pavlova salina. SEQ ID NO:8 - Full length cDNA encoding of A5 elongase from Pavlova salina. SEQ ID NO:9 - cDNA encoding open reading frame of A9 elongase from Pavlova salina. SEQ ID NO:10 - Full length cDNA encoding of A9 elongase from Pavlova salina. SEQ ID NO: 11 - Partial cDNA encoding N-terminal portion of A4 desaturase from Pavlova salina. SEQ ID NO: 12 - cDNA encoding open reading frame of A4 desaturase from Pavlova salina. SEQ D NO:13 - Full length cDNA encoding A4 desaturase from Pavlova salina. SEQ ID NO: 14 - A5/A6 bifunctional elongase from Caenorhabditis elegans. SEQ ID NO: 15 - A5/A6 bifunctional desaturase from Danio rerio (zebrafish). SEQ ID NO: 16 - A5 desaturase from humans (Genbank Accession No: AAF29378). SEQ ID NO: 17 - A5 desaturase from Pythium irregulare (Genbank Accession No: AAL1331 1). SEQ ID NO:18 - A5 desaturase from Thraustochytrium sp. (Genbank Accession No: AAM09687). SEQ ID NO: 19 - A5 desaturase from Mortierella alpina (Genbank Accession No: 074212). SEQ ID NO:20 - A5 desaturase from Caenorhabditis elegans (Genbank Accession No: T43319). SEQ ID NO:21 - A6 desaturase from humans (Genbank Accession No: AAD20018).

WO 2005/103253 PCT/AU2005/000571 22 SEQ ID NO:22 - A6 desaturase from mouse (Genbank Accession No: NP 062673). SEQ ID NO:23 - A6 desaturase from Pythium irregulare (Genbank Accession No: AAL13310). SEQ ID NO:24 - A6 desaturase from Borago officinalis (Genbank Accession No: AAD01410). SEQ ID NO:25 - A6 desaturase from Anemone leveillei (Genbank Accession No: AAQ10731). SEQ ID NO:26 - A6 desaturase from Ceratodon purpureus (Genbank Accession No: CAB94993). SEQ ID NO:27 - A6 desaturase from Physcomitrella patens (Genbank Accession No: CAA1 1033). SEQ ID NO:28 - A6 desaturase from Mortierella alpina (Genbank Accession No: BAC82361). SEQ ID NO:29 - A6 desaturase from Caenorhabditis elegans (Genbank Accession No: AAC15586). SEQ ID NO:30 - A5 elongase from humans (Genbank Accession No: NP 068586). SEQ ID NO:31 - A6 elongase from Physcomitrella patens (Genbank Accession No: AAL84174). SEQ ID NO:32 - A6 elongase from Mortierella alpina (Genbank Accession No: AAF70417). SEQ ID NO:33 - A4 desaturase from Thraustochytrium sp. (Genbank Accession No: AAM09688). SEQ ID NO:34 - A4 desaturase from Euglena gracilis (Genbank Accession No: AAQ19605). SEQ ID NO:35 - A9 elongase from Isochrysis galbana (Genbank Accession No: AAL37626). SEQ ID NO:36 - A8 desaturase from Euglena gracilis (Genbank Accession No: AAD45877). SEQ ID NO:37 - cDNA encoding A5/A6 bifunctional elongase from Caenorhabditis elegans. SEQ ID NO:38 - cDNA encoding A5/A6 bifunctional desaturase from Danio rerio (zebrafish). SEQ ID NO's:39 to 42, 46, 47, 50, 51, 53, 54, 56, 57, 81, 82, 83, 84 and 87 Oligonucleotide primers. SEQ ID NO's:43 to 45, 48, 49 and 52 - Conserved motifs of various desaturases/elongases.

WO 2005/103253 PCT/AU2005/000571 23 SEQ ID NO:55 - Partial cDNA encoding Pavlova salina FAE-like elongase. SEQ ID NO:58 - Full length cDNA encoding A5 desaturase from Pavlova salina. SEQ ID NO:59 - cDNA encoding open reading frame of A5 desaturase from Pavlova salina. SEQ ID NO:60 - A5 desaturase from Pavlova salina. SEQ ID NO's 61 and 62 - Fragments of Echium pitardii A6 desaturase. SEQ ID NO:63 - cDNA encoding open reading frame of A6 desaturase from Echium plantagineum. SEQ ID NO:64 - A6 desaturase from Echium plantagineum. SEQ ID NO:65 - A6 desaturase from Echium gentianoides (Genbank Accession No: AY055117). SEQ ID NO:66 - A6 desaturase from Echium pitardii (Genbank Accession No: AY055118). SEQ ID NO:67 - A6 desaturase from Borago officinalis (Genbank Accession No: U79010). SEQ ID NO:68 - A8 desaturase from Borago officinalis (Genbank Accession No: AF133728). SEQ ID NO:69 - A8 desaturase from Helianthus annus (Genbank Accession No: S68358). SEQ ID NO:70 - A8 desaturaseA from Arabiposis thaliana (Genbank Accession No: AAC62885.1). SEQ ID NO:71 - A8 desaturaseB from Arabiposis thaliana (Genbank Accession No: CAB71088.1). SEQ ID NO:72 and 73 - Conserved motifs of A6- and A8-desaturases. SEQ ID NO:74 - A6 elongase from Thraustochytrium sp. (Genbank Accession No: AX951565). SEQ ID NO:75 - A9 elongase from Danio rerio (Genbank Accession No: NM_199532). SEQ ID NO:76 - A9 elongase from Pavlova lutheri. SEQ ID NO:77 - A5 elongase from Danio rerio (Genbank Accession No: AF532782). SEQ ID NO:78 - A5 elongase from Pavlova lutheri. SEQ ID NO:79 - Partial gene sequence from Heterocapsa niei encoding an elongase. SEQ ID NO:80 - Protein encoded by SEQ ID NO:79, presence of stop codon suggests an intron in SEQ ID NO:79. SEQ ID NO:85 - A9 elongase from Pavlova salina, encoded by alternate start codon at position 31 of SEQ ID NO:9.

WO 2005/103253 PCT/AU2005/000571 24 SEQ ID NO:86 - A9 elongase from Pavlova salina, encoded by alternate start codon at position 85 of SEQ ID NO:9. SEQ ID NO:88 - Partial elongase amino acid sequence from Melosira sp. SEQ ID NO:89 - cDNA sequence encoding partial elongase from Melosira sp. DETAILED DESCRIPTION OF THE INVENTION General Techniques and Definitions Unless specifically defined otherwise, all technical and scientific terms used herein shall be taken to have the same meaning as commonly understood by one of ordinary skill in the art (e.g., in cell culture, plant biology, molecular genetics, immunology, immunohistochemistry, protein chemistry, fatty acid synthesis, and biochemistry). Unless otherwise indicated, the recombinant nucleic acid, recombinant protein, cell culture, and immunological techniques utilized in the present invention are standard procedures, well known to those skilled in the art. Such techniques are described and explained throughout the literature in sources such as, J. Perbal, A Practical Guide to Molecular Cloning, John Wiley and Sons (1984), J. Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbour Laboratory Press (1989), T.A. Brown (editor), Essential Molecular Biology: A Practical Approach, Volumes 1 and 2, IRL Press (1991), D.M. Glover and B.D. Hames (editors), DNA Cloning: A Practical Approach, Volumes 1-4, IRL Press (1995 and 1996), and F.M. Ausubel et al. (editors), Current Protocols in Molecular Biology, Greene Pub. Associates and Wiley-Interscience (1988, including all updates until present), Ed Harlow and David Lane (editors) Antibodies: A Laboratory Manual, Cold Spring Harbour Laboratory, (1988), and J.E. Coligan et al. (editors) Current Protocols in Immunology, John Wiley & Sons (including all updates until present), and are incorporated herein by reference. As used herein, the terms "long-chain polyunsaturated fatty acid", "LC-PUFA" or "C20+ polyunsaturated fatty acid" refer to a fatty acid which comprises at least 20 carbon atoms in its carbon chain and at least three carbon-carbon double bonds. As used herein, the term "very long-chain polyunsaturated fatty acid", "VLC-PUFA" or "C22+ polyunsaturated fatty acid" refers to a fatty acid which comprises at least 22 carbon atoms in its carbon chain and at least three carbon-carbon double bonds. Ordinarily, the number of carbon atoms in the carbon chain of the fatty acids refers to an unbranched carbon chain. If the carbon chain is branched, the number of carbon atoms excludes those in sidegroups. In one embodiment, the long-chain polyunsaturated fatty acid is an co3 fatty acid, that is, having a desaturation (carbon-carbon double bond) in the third carbon- WO 2005/103253 PCT/AU2005/000571 25 carbon bond from the methyl end of the fatty acid. In another embodiment, the long chain polyunsaturated fatty acid is an o6 fatty acid, that is, having a desaturation (carbon carbon double bond) in the sixth carbon-carbon bond from the methyl end of the fatty acid. In a further embodiment, the long-chain polyunsaturated fatty acid is selected from the group consisting of; arachidonic acid (ARA, 20:4A5,8,11,14; co6), eicosatetraenoic acid (ETA, 20:4A8,11,14,17, 03) eicosapentaenoic acid (EPA, 20:5A5,8,11,14,17; 03), docosapentaenoic acid (DPA, 22:5A7,10,13,16,19, o3), or docosahexaenoic acid (DHA, 22:6A4,7,10,13,16,19, o3). The LC-PUFA may also be dihomo-y-linoleic acid (DGLA) or eicosatrienoic acid (ETrA, 20:3A11,14,17, 03). It would readily be apparent that the LC-PUFA that is produced according to the invention may be a mixture of any or all of the above and may include other LC-PUFA or derivatives of any of these LC-PUFA. In a preferred embodiment, the 03 fatty acid is EPA, DPA, or DHA, or even more preferably DPA or DHA. Furthermore, as used herein the terms "long-chain polyunsaturated fatty acid" or "very long-chain polyunsaturated fatty acid" refer to the fatty acid being in a free state (non-esterified) or in an esterified form such as part of a triglyceride, diacylglyceride, monoacylglyceride, acyl-CoA bound or other bound form. The fatty acid may be esterified as a phospholipid such as a phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylglycerol, phosphatidylinositol or diphosphatidylglycerol forms. Thus, the LC-PUFA may be present as a mixture of forms in the lipid of a cell or a purified oil or lipid extracted from cells, tissues or organisms. In preferred embodiments, the invention provides oil comprising at least 75% or 85% triacylglycerols, with the remainder present as other forms of lipid such as those mentioned, with at least said triacylglycerols comprising the LC-PUFA. The oil may be further purified or treated, for example by hydrolysis with a strong base to release the free fatty acid, or by fractionation, distillation or the like. As used herein, the abbreviations "LC-PUFA" and "VLC-PUFA" can refer to a single type of fatty acid, or to multiple types of fatty acids. For example, a transgenic plant of the invention which produces LC-PUFA may produce EPA, DPA and DHA. The desaturase and elongase proteins and genes. encoding them that may be used in the invention are any of those known in the art or homologues or derivatives thereof. Examples of such genes and the encoded protein sizes are listed in Table 1. The desaturase enzymes that have been shown to participate in LC-PUFA biosynthesis all belong to the group of so-called "front-end" desaturases which are characterised by the presence of a cytochrome b 5 -like domain at the N-terminus of each protein. The WO 2005/103253 PCT/AU2005/000571 26 TABLE 1. Cloned genes involved in LC-PUFA biosynthesis. Enzyme Type of Species Accession Protein References organism Nos. size _ (aa's) A4- Algae Euglena gracilis AY278558 541 Meyer et al., desaturase 2003 Pavlova lutherii AY332747 445 Tonon et al., 2003 Thraustochytrium AF489589 519 Qiu et al., 2001 sp. Thraustochytrium AF391543- 515 (NCBI) aureum 5 A5- Mammals Homo sapiens AF199596 444 Cho et al., 1999b desaturase Leonard et al., 2000b Nematode Caenorhabditis AF1 1440, 447 Michaelson et elegans NM_06935 al., 1998b; Watts 0 and Browse, 1999b Fungi Mortierella AF067654 446 Michaelson et alpina al., 1998a; Knutzon et al., 1998 Pythium AF419297 456 Hong et al., irregulare 2002a Dictyostelium AB022097 467 Saito et al., 2000 discoideum Saprolegnia 470 W002081668 diclina Diatom Phaeodactylum AY082392 469 Domergue et al., tricornutum 2002 Algae Thraustochytrium AF489588 439 Qiu et al., 2001 sp Thraustochytrium 439 W002081668 aureum Isochrysis 442 W002081668 galbana Moss Marchantia AY583465 484 Kajikawa et al., polymorpha 2004 WO 2005/103253 PCT/AU2005/000571 27 Enzyme Type of Species Accession Protein References organism Nos. size (aa's) A6- Mammals Homo sapiens NM_013402 444 Cho et al., 1999a; desaturase Leonard et al., 2000 Mus musculus NM 019699 444 Cho et al., 1999a Nematode Caenorhabditis Z70271 443 Napier et al., 1998 elegans Plants Borago officinales U79010 448 Sayanova et al., 1997 Echium AY055117 Garcia-Maroto et AY055118 al., 2002 Primula vialii AY234127 453 Sayanova et al., 2003 Anemone leveillei AF536525 446 Whitney et al., 2003 Mosses Ceratodon AJ250735 520 Sperling et al., 2000 purpureus Marchantia AY583463 481 Kajikawa et al., polymorpha 2004 Physcomitrella Girke et al., 1998 patens Fungi Mortierella alpina AF110510 457 Huang et al., 1999; AB020032 Sakuradani et al., 1999 Pythium AF419296 459 Hong et al., 2002a irregulare Mucor AB052086 467 NCBI* circinelloides Rhizopus sp. AY320288 458 Zhang et al., 2004 Saprolegnia 453 W002081668 diclina Diatom Phaeodactylum AY082393 477 Domergue et al., tricornutum 2002 Bacteria Synechocystis L 11421 359 Reddy et al., 1993 Algae Thraustochytrium 456 W002081668 aureum Bifunction Fish Danio rerio AF309556 444 Hastings et al., al A5/A6 2001 desaturase C20 A8- Algae Euglena gracilis AF139720 419 Wallis and Browse, desaturase 1999 Plants Borago officinales AF133728 WO 2005/103253 PCT/AU2005/000571 28 Enzyme Type of Species Accession Protein References organism Nos. size (aa's) A6-elongase Nematode NM_069288 288 . Beaudoin et al., Caenorhabditis 2000 elegans Mosses Physcomitrella AF428243 290 Zank et al., 2002 patens Marchantia AY583464 290 Kajikawa et al., polymorpha 2004 Fungi Mortierella AF206662 318 Parker-Barnes et alpina al., 2000 Algae Pavlova lutheri** 501 WO 03078639 Thraustochytrium AX951565 271 WO 03093482 Thraustochytrium AX214454 271 WO 0159128 sp** PUFA- Mammals Homo sapiens AF231981 299 Leonard et al., elongase 2000b; Leonard et al., 2002 Rattus norvegicus AB071985 299 Inagaki et al., 2002 Rattus AB071986 267 Inagaki et al., norvegicus** 2002 Mus musculus AF170907 279 Tvrdik et al., 2000 Mus musculus AF170908 292 Tvrdik et al., 2000 Fish Danio rerio AF532782 291 Agaba et al., 2004 (282) Danio rerio** NM 199532 266 Lo et al., 2003 Worm Caenorhabditis Z68749 309 Abbott et al 1998 elegans Beaudoin et al 2000 Algae Thraustochytrium AX464802 272 WO 0208401-A2 aureum * * Pavlova lutheri** WO 03078639 A9-elongase Algae Isochrysis AF390174 263 Qi et al., 2002 galbana I I * http://www.ncbi.nlm.nih.gov/ ** Function not proven/not demonstrated WO 2005/103253 PCT/AU2005/000571 29 cytochrome bs-like domain presumably acts as a receptor of electrons required for desaturation (Napier et al., 1999; Sperling and Heinz, 2001). Activity of any of the elongases or desaturases for use in the invention may be tested by expressing a gene encoding the enzyme in a cell such as, for example, a yeast cell or a plant cell, and determining whether the cell has an increased capacity to produce LC-PUFA compared to a comparable cell in which the enzyme is not expressed. Unless stated to the contrary, embodiments of the present invention which relate to cells, plants, seeds, etc, and methods for the production thereof, and that refer to at least "two enzymes" (or at least "three enzymes" etc) of the list that is provided means that the polynucleotides encode at least two "different" enzymes from the list provided and not two identical (or very similar with only a few differences as to not substantially alter the activity of the encoded enzyme) open reading frames encoding essentially the same enzyme. As used herein, unless stated to the contrary, the term "substantially the same", or variations thereof, means that two samples being analysed, for example two seeds from different sources, are substantially the same if they only vary about +/-10% in the trait being investigated. As used herein, the term "an enzyme which preferentially converts an o6 LC PUFA into an 03 LC-PUFA" means that the enzyme is more efficient at performing said conversion than it is at performing a desaturation reaction outlined in pathways II or III of Figure 1. Whilst certain enzymes are specifically described herein as "bifunctional", the absence of such a term does not necessarily imply that a particular enzyme does not possess an activity other than that specifically defined. Desaturases As used herein, a "A5/A6 bifunctional desaturase" or "A5/A6 desaturase" is at least capable of i) converting c-linolenic acid to octadecatetraenoic acid, and ii) converting eicosatetraenoic acid to eicosapentaenoic acid. That is, a A5/A6 bifunctional desaturase is both a A5 desaturase and a A6 desaturase, and A5/A6 bifunctional desaturases may be considered a sub-class of each of these. A gene encoding a bifunctional A5-/A6 desaturase has been identified from zebrafish (Hasting et al., 2001). The gene encoding this enzyme might represent an ancestral form of the "front-end desaturase" which later duplicated and the copies evolved distinct A5- and A6-desaturase functions. In one embodiment, the A5/A6 bifunctional desaturase is naturally produced by a freshwater species of fish. In a particular embodiment, the A5/A6 bifunctional desaturase comprises WO 2005/103253 PCT/AU2005/000571 30 i) an amino acid sequence as provided in SEQ ID NO:15, ii) an amino acid sequence which is at least 50% identical to SEQ ID NO: 15, or iii) a biologically active fragment of i) or ii). As used herein, a "A5 desaturase" is at least capable of converting eicosatetraenoic acid to eicosapentaenoic acid. In one embodiment, the enzyme A5 desaturase catalyses the desaturation of C20 LC-PUFA, converting DGLA to arachidonic acid (ARA, 20:4o6) and ETA to EPA (20:503). Genes encoding this enzyme have been isolated from a number of organisms, including algae (Thraustochytrium sp. Qiu et al., 2001), fungi (M alpine, Pythium irregulare, P. tricornutum, Dictyostelium), Caenorhabditis elegans and mammals (Table 1). In another embodiment, the A5 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO: 19, SEQ ID NO:20 or SEQ ID NO:60, (ii) an amino acid sequence which is at least 50% identical to any one of SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20 or SEQ ID NO:60, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A5 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:20 or SEQ ID NO:60, (ii) an amino acid sequence which is at least 90% identical to any one of SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:20 or SEQ ID NO:60, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A5 desaturase is encoded by the protein coding region of one of the A5 desaturase genes listed in Table 1 or gene substantially identical thereto. As used herein, a "A6 desaturase" is at least capable of converting u-linolenic acid to octadecatetraenoic acid. In one embodiment, the enzyme A6 desaturase catalyses the desaturation of C18 LC-PUFA, converting LA to GLA and ALA to SDA. In another embodiment, the A6 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:27, SEQ ID NO:28, SEQ ID NO:29, SEQ ID NO:64 SEQ ID NO:65, SEQ ID NO:66 or SEQ ID NO:67, (ii) an amino acid sequence which is at least 50% identical to any one of SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:27, SEQ ID NO:28, SEQ ID NO:29, SEQ ID NO:64 SEQ ID NO:65, SEQ ID NO:66 or SEQ ID NO:67, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A6 desaturase comprises an amino acid sequence which is at least 90% identical to any one of SEQ ID NO:21, SEQ ID NO:22, SEQ ID NO:23, SEQ ID NO:24, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:27, SEQ ID NO:28, SEQ ID NO:29, SEQ ID NO:64 SEQ ID NO:65, SEQ ID NO:66 or SEQ ID NO:67. In a further embodiment, the A6 desaturase is WO 2005/103253 PCT/AU2005/000571 31 encoded by the protein coding region of one of the A6 desaturase genes listed in Table 1 or gene substantially identical thereto As used herein, a "A4 desaturase" is at least capable of converting docosapentaenoic acid to docosahexaenoic acid. The desaturation step to produce DHA from DPA is catalysed by a A4 desaturase in organisms other than mammals, and a gene encoding this enzyme has been isolated from the freshwater protist species Euglena gracilis and the marine species Thraustochytrium sp. (Qiu et al., 2001; Meyer et al., 2003). In one embodiment, the A4 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:4, SEQ ID NO:33 or SEQ ID NO:34, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:4, SEQ ID NO:33 or SEQ ID NO:34, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A4 desaturase is encoded by the protein coding region of one of the A4 desaturase genes listed in Table 1 or gene substantially identical thereto. As used herein, a "A8 desaturase" is at least capable of converting 2 0: 3 All1,4,17o 3 to eicosatetraenoic acid. In one embodiment, the A8 desaturase is relatively specific for A8 substrates. That is, it has greater activity in desaturating A8 substrates than other substrates, in particular A6 desaturated substrates. In a preferred embodiment, the A8 desaturase has little or no A6 desaturase activity when expressed in yeast cells. In another embodiment, the A8 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:1, SEQ ID NO:36, SEQ ID NO:68, SEQ ID NO:69, SEQ ID NO:70 or SEQ ID NO:71, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:1, SEQ ID NO:36, SEQ ID NO:68, SEQ ID NO:69, SEQ ID NO:70 or SEQ ID NO:71, or (iii) a biologically active fragment of i) or ii). In further embodiment, the A8 desaturase comprises (i) an amino acid sequence as provided in SEQ ID NO:1, (ii) an amino acid sequence which is at least 90% identical to SEQ ID NO: 1, or (iii) a biologically active fragment of i) or ii). As used herein, an "o3 desaturase" is at least capable of converting LA to ALA and/or GLA to SDA and/or ARA to EPA. Examples of 3 desaturase include those described by Pereira et al. (2004), Horiguchi et al. (1998), Berberich et al. (1998) and Spychalla et al. (1997). In one embodiment, a cell of the invention is a plant cell which lacks o3 desaturase activity. Such cells can be produced using gene knockout technology well known in the art. These cells can be used to specifically produce large quantities of o6 LC-PUFA such as DGLA.

WO 2005/103253 PCT/AU2005/000571 32 Elongases Biochemical evidence suggests that the fatty acid elongation consists of 4 steps: condensation, reduction, dehydration and a second reduction. In the context of this invention, an "elongase" refers to the polypeptide that catalyses the condensing step in the presence of the other members of the elongation complex, under suitable physiological conditions. It has been shown that heterologous or homologous expression in a cell of only the condensing component ("elongase") of the elongation protein complex is required for the elongation of the respective acyl chain. Thus the introduced elongase is able to successfully recruit the reduction and dehydration activities from the transgenic host to carry out successful acyl elongations. The specificity of the elongation reaction with respect to chain length and the degree of desaturation of fatty acid substrates is thought to reside in the condensing component. This component is also thought to be rate limiting in the elongation reaction. Two groups of condensing enzymes have been identified so far. The first are involved in the extension of saturated and monounsaturated fatty acids (C 18-22) such as, for example, the FAE1 gene of Arabidopsis. An example of a product formed is erucic acid (22:1) in Brassicas. This group are designated the FAE-like enzymes and do not appear to have a role in LC-PUFA biosynthesis. The other identified class of fatty acid elongases, designated the ELO family of elongases, are named after the ELO genes whose activities are required for the synthesis of the very long-chain fatty acids of sphingolipids in yeast. Apparent paralogs of the ELO-type elongases isolated from LC PUFA synthesizing organisms like algae, mosses, fungi and nematodes have been shown to be involved in the elongation and synthesis of LC-PUFA. Several genes encoding such PUFA-elongation enzymes have also been isolated (Table 1). Such genes are unrelated in nucleotide or amino acid sequence to the FAE-like elongase genes present in higher plants. As used herein, a "A5/A6 bifunctional elongase" or "A5/A6 elongase" is at least capable of i) converting octadecatetraenoic acid to eicosatetraenoic acid, and ii) converting eicosapentaenoic acid to docosapentaenoic acid. That is, a A5/A6 bifunctional elongase is both a A5 elongase and a A6 elongase, and A5/A6 bifunctional elongases may be considered a sub-class of each of these. In one embodiment, the A5/A6 bifunctional elongase is able to catalyse the elongation of EPA to form DPA in a plant cell such as, for example, a higher plant cell, when that cell is provided with a source of EPA. The EPA may be provided exogenously or preferably endogenously. A gene encoding such an elongase has been isolated from an invertebrate, C. elegans (Beaudoin et al., 2000) although it was not previously known to catalyse the A5-elongation step. In one WO 2005/103253 PCT/AU2005/000571 33 embodiment, the A5/A6 bifunctional elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:2 or SEQ ID NO: 14, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:2 or SEQ ID NO:14, or (iii) a biologically active fragment of i) or ii). As used herein, a "A5 elongase" is at least capable of converting eicosapentaenoic acid to docosapentaenoic acid. In one embodiment, the A5 elongase is from a non vertebrate source such as, for example, an algal or fungal source. Such elongases can have advantages in terms of the specificity of the elongation reactions carried out (for example the A5 elongase provided as SEQ ID NO:2). In a preferred embodiment, the A5 elongase is relatively specific for C20 substrates over C22 substrates. For example, it may have at least 10-fold lower activity toward C22 substrates (elongated to C24 fatty acids) relative to the activity toward a corresponding C20 substrate when expressed in yeast cells. It is preferred that the activity when using C20 A5 desaturated substrates is high, such as for example, providing an efficiency for the conversion of 20:5(o3 into 22:5o3 of at least 7% when expressed in yeast cells. In another embodiment, the A5 elongase is relatively specific for A5 desaturated substrates over A6 desaturated substrates. For example, it may have at least 10-fold lower activity toward A6 desaturated C18 substrates relative to A5 desaturated C20 substrates when expressed in yeast cells. In a further embodiment, the A5 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:2, SEQ ID NO:30, SEQ ID NO:77 or SEQ ID NO:78, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:2, SEQ ID NO:30, SEQ ID NO:77 or SEQ ID NO:78, or (iii) a biologically active fragment of i) or ii). In another embodiment, the A5 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:2, (ii) an amino acid sequence which is at least 90% identical to SEQ ID NO:2, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A5 elongase is encoded by the protein coding region of one of the A5 elongase genes listed in Table 1 or gene substantially identical thereto. As used herein, a "A6 elongase" is at least capable of converting octadecatetraenoic acid to eicosatetraenoic acid. In one embodiment, the A6 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:31, SEQ ID NO:32, SEQ ID NO:74, SEQ ID NO:85, SEQ ID NO:86 or SEQ ID NO:88, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:31, SEQ ID NO:32, SEQ ID NO:74, SEQ ID NO:85, SEQ ID NO:86 or SEQ ID NO:88, or (iii) a biologically active fragment of i) or ii). In. another embodiment, the A6 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:2, SEQ ID NO:3 or SEQ ID NO:32, SEQ ID NO:85, SEQ ID NO:86 or SEQ ID WO 2005/103253 PCT/AU2005/000571 34 NO:88, (ii) an amino acid sequence which is at least 90% identical to SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:32, SEQ ID NO:85, SEQ ID NO:86 or SEQ ID NO:88, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A6 elongase is encoded by the protein coding region of one of the A6 elongase genes listed in Table 1 or gene substantially identical thereto. In some protist species, LC-PUFA are synthesized by elongation of linoleic or a linolenic acid with a C2 unit, before desaturation with A8 desaturase (Figure 1 part IV; "A8-desaturation" pathway). A6 desaturase and A6 elongase activities were not detected in these species. Instead, a A9-elongase activity would be expected in such organisms, and in support of this, a C18 A9-elongase gene has recently been isolated from Isochrysis galbana (Qi et al., 2002). As used herein, a "A9 elongase" is at least capable of converting a-linolenic acid to 20:3ll, 1 4 ,17o3. In one embodiment, the A9 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:3, SEQ ID NO:35, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:85 or SEQ ID NO:86, (ii) an amino acid sequence which is at least 50% identical to SEQ ID NO:3, SEQ ID NO:35, SEQ ID NO:75, SEQ ID NO:76, SEQ ID NO:85 or SEQ ID NO:86, or (iii) a biologically active fragment of i) or ii). In another embodiment, the A9 elongase comprises (i) an amino acid sequence as provided in SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, (ii) an amino acid sequence which is at least 90% identical to SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO: 86, or (iii) a biologically active fragment of i) or ii). In a further embodiment, the A9 elongase is encoded by the protein coding region of the A9 elongase gene listed in Table 1 or gene substantially identical thereto. In another embodiment, the A9 elongase also has A6 elongase activity. The elongase in this embodiment is able to convert SDA to ETA and/or GLA to DGLA (A6 elongase activity) in addition to converting ALA to ETrA (A9 elongase). In a preferred embodiment, such an elongase is from an algal or fungal source such as, for example, the genus Pavlova. As used herein, a "A4 elongase" is at least capable of converting docosahexaenoic acid to 24:6A 6

,

9

,

12 ,1 5 ,1 8

,

21 o3 Cells Suitable cells of the invention include any cell that can be transformed with a polynucleotide encoding a polypeptide/enzyme described herein, and which is thereby capable of being used for producing LC-PUFA. Host cells into which the polynucleotide(s) are introduced can be either untransformed cells or cells that are already transformed with at least one nucleic acid molecule. Such nucleic acid molecule may related to LC-PUFA synthesis, or unrelated. Host cells of the present invention WO 2005/103253 PCT/AU2005/000571 35 either can be endogenously (i.e., naturally) capable of producing proteins of the present invention or can be capable of producing such proteins only after being transformed with at least one nucleic acid molecule. As used herein, the term "cell with an enhanced capacity to synthesize a long chain polyunsaturated fatty acid" is a relative term where the recombinant cell of the invention is compared to the native cell, with the recombinant cell producing more long chain polyunsaturated fatty acids, or a greater concentration of LC-PUFA such as EPA, DPA or DHA (relative to other fatty acids), than the native cell. The cells may be prokaryotic or eukaryotic. Host cells of the present invention can be any cell capable of producing at least one protein described herein, and include bacterial, fungal (including yeast), parasite, arthropod, animal and plant cells. Preferred host cells are yeast and plant cells. In a preferred embodiment, the plant cells are seed cells. In one embodiment, the cell is an animal cell or an algal cell. The animal cell may be of any type of animal such as, for example, a non-human animal cell, a non-human vertebrate cell, a non-human mammalian cell, or cells of aquatic animals such as fish or crustacea, invertebrates, insects, etc. An example of a bacterial cell useful as a host cell of the present invention is Synechococcus spp. (also known as Synechocystis spp.), for example Synechococcus elongatus. The cells may be of an organism suitable for fermentation. As used herein, the term the "fermentation process" refers to any fermentation process or any process comprising a fermentation step. A fermentation process includes, without limitation, fermentation processes used to produce alcohols (e.g., ethanol, methanol, butanol); organic acids (e.g., citric acid, acetic acid, itaconic acid, lactic acid, gluconic acid); ketones (e.g., acetone); amino acids (e.g., glutamic acid); gases (e.g., H 2 and C0 2 ); antibiotics (e.g., penicillin and tetracycline); enzymes; vitamins (e.g., riboflavin, beta carotene); and hormones. Fermentation processes also include fermentation processes used in the consumable alcohol industry (e.g., beer and wine), dairy industry (e.g., fermented dairy products), leather industry and tobacco industry. Preferred fermentation processes include alcohol fermentation processes, as are well known in the art. Preferred fermentation processes are anaerobic fermentation processes, as are well known in the art. Suitable fermenting cells, typically microorganisms are able to ferment, i.e., convert, sugars, such as glucose or maltose, directly or indirectly into the desired fermentation product. Examples of fermenting microorganisms include fungal organisms, WO 2005/103253 PCT/AU2005/000571 36 such as yeast. As used herein, "yeast" includes Saccharomyces spp., Saccharomyces cerevisiae, Saccharomyces carlbergensis, Candida spp., Kluveromyces spp., Pichia spp., Hansenula spp., Trichoderma spp., Lipomyces starkey, and Yarrowia lipolytica. Preferred yeast include strains of the Saccharomyces spp., and in particular, Saccharomyces cerevisiae. Commercially available yeast include, e.g., Red Star/Lesaffre Ethanol Red (available from Red Star/Lesaffre, USA) FALI (available from Fleischmann's Yeast, a division of Burns Philp Food Inc., USA), SUPERSTART (available from Alltech), GERT STRAND (available from Gert Strand AB, Sweden) and FERMIOL (available from DSM Specialties). Evidence to date suggests that some desaturases expressed heterologously in yeast have relatively low activity in combination with some elongases. However, the present inventors have identified that this may be alleviated by providing a desaturase with the capacity of to use an acyl-CoA form of the fatty acid as a substrate in LC-PUFA synthesis, and this is thought to be advantageous in recombinant cells other than yeast as well. In this regard, it has also been determined that desaturases of vertebrate origin are particularly useful for the production of LC-PUFA. Thus in embodiments of the invention, either (i) at least one of the enzymes is a A5 elongase that catalyses the conversion of EPA to DPA in the cell, (ii) at least one of the desaturases is able to act on an acyl-CoA substrate, (iii) at least one desaturase is from vertebrate or is a variant thereof, or (iv) a combination of ii) and iii). In a particularly preferred embodiment, the host cell is a plant cell, such as those described in further detail herein. As used herein, a "progenitor cell of a seed" is a cell that divides and/or differentiates into a cell of a transgenic seed of the invention, and/or a cell that divides and/or differentiates into a transgenic plant that produces a transgenic seed of the invention. Levels of LC-PUFA produced The levels of the LC-PUFA that are produced in the recombinant cell are of importance. The levels may be expressed as a composition (in percent) of the total fatty acid that is a particular LC-PUFA or group of related LC-PUFA, for example the o3 LC PUFA or the o6 LC-PUFA, or the C22+ PUFA, or other which may be determined by methods known in the art. The level may also be expressed as a LC-PUFA content, such as for example the percentage of LC-PUFA in the dry weight of material comprising the recombinant cells, for example the percentage of the dry weight of seed that is LC-PUFA. It will be appreciated that the LC-PUFA that is produced in an oilseed may be WO 2005/103253 PCT/AU2005/000571 37 considerably higher in terms of LC-PUFA content than in a vegetable or a grain that is not grown for oil production, yet both may have similar LC-PUFA compositions, and both may be used as sources of LC-PUFA for human or animal consumption. The levels of LC-PUFA may be determined by any of the methods known in the art. For example, total lipid may be extracted from the cells, tissues or organisms and the fatty acid converted to methyl esters before analysis by gas chromatography (GC). Such techniques are described in Example 1. The peak position in the chromatogram may be used to identify each particular fatty acid, and the area under each peak integrated to determine the amount. As used herein, unless stated to the contrary, the percentage of particular fatty acid in a sample is determined as the area under the peak for that fatty acid as a percentage of the total area for fatty acids in the chromatogram. This corresponds essentially to a weight percentage (w/w). The identity of fatty acids may be confirmed by GC-MS, as described in Example 1. In certain embodiments, where the recombinant cell is useful in a fermentation process such as, for example, a yeast cell, the level of EPA that is produced may be at least 0.21% of the total fatty acid in the cell, preferably at least 0.82% or at least 2% and even more preferably at least 5%. In other embodiments, the total fatty acid of the recombinant cell may comprise at least 1.5% EPA, preferably at least 2.1% EPA, and more preferably at least 2.5%, at least 3.1%, at least 4% or at least 5.1% EPA. In further embodiments, where the recombinant cell is useful in a fermentation process or is a plant cell and DPA is produced, the total fatty acid in the cell may comprise at least 0.1% DPA, preferably at least 0.13% or at least 0.15% and more preferably at least 0.5% or at least 1% DPA. In further embodiments, the total fatty acid of the cell may comprise at least 2% C20 LC-PUFA, preferably at least 3% or at least 4% C20 LC-PUFA, more preferably at least 4.7% or at least 7.9% C20 LC-PUFA and most preferably at least 10.2% C20 LC PUFA. In further embodiments, the total fatty acid of the cell may comprise at least 2.5% C20 o3 LC-PUFA, preferably at least 4.1% or more preferably at least 5% C20 o3 LC PUFA. In other embodiments, where both EPA and DPA are synthesized in a cell, the level of EPA reached is at least 1.5%, at least 2.1% or at least 2.5% and the level of DPA at least 0.13%, at least 0.5% or at least 1.0%. In each of these embodiments, the recombinant cell may be a cell of an organism that is suitable for fermentation such as, for example, a unicellular microorganism which WO 2005/103253 PCT/AU2005/000571 38 may be a prokaryote or a eukaryote such as yeast, or a plant cell. In a preferred embodiment, the cell is a cell of an angiosperm (higher plant). In a further preferred embodiment, the cell is a cell in a seed such as, for example, an oilseed or a grain or cereal. The level of production of LC-PUFA in the recombinant cell may also be expressed as a conversion ratio, i.e. the amount of the LC-PUFA formed as a percentage of one or more substrate PUFA or LC-PUFA. With regard to EPA, for example, this may be expressed as the ratio of the level of EPA (as a percentage in the total fatty acid) to the level of a substrate fatty acid (ALA, SDA, ETA or ETrA). In a preferred embodiment, the conversion efficiency is for ALA to EPA. In particular embodiments, the conversion ratio for production of EPA in a recombinant cell may be at least 0.5%, at least 1%, or at least 2%. In another embodiment, the conversion efficiency for ALA to EPA is at least 14.6%. In further embodiments, the conversion ratio for production of DPA from EPA in a recombinant cell is at least 5%, at least 7%, or at least 10%. In other embodiments, the total o3 fatty acids produced that are products of A6 desaturation (i.e. downstream of 18:303 (ALA), calculated as the sum of the percentages for 18:403 (SDA), 20:403 (ETA), 20:5o3 (EPA) and 22:503 (DPA)) is at least 4.2%. In a particular embodiment, the conversion efficiency of ALA to o3 products through a A6 desaturation step and/or an A9 elongation step in a recombinant cell, preferably a plant cell, more preferably a seed cell, is at least 22% or at least 24%. Stated otherwise, in this embodiment the ratio of products derived from ALA to ALA (products:ALA) in the cell is at least 1:3.6. The content of the LC-PUFA in the recombinant cell may be maximized if the parental cell used for introduction of the genes is chosen such that the level of fatty acid substrate that is produced or provided exogenously is optimal. In particular embodiments, the cell produces ALA endogenously at levels of at least 30%, at least 50%, or at least 66% of the total fatty acid. The level of LC-PUFA may also be maximized by growing or incubating the cells under optimal conditions, for example at a slightly lower temperature than the standard temperature for that cell, which is thought to favour accumulation of polyunsaturated fatty acid. There are advantages to maximizing production of a desired LC-PUFA while minimizing the extent of side-reactions. In a particular embodiment, there is little or no ETrA detected (less than 0.1%) while the level of EPA is at least 2.1%. Turning to transgenic plants of the invention, in one embodiment, at least one plant part synthesizes EPA, wherein the total fatty acid of the plant part comprises at least 1.5%, at least 2.1%, or at least 2.5% EPA.

WO 2005/103253 PCT/AU2005/000571 39 In another embodiment, at least one plant part synthesizes DPA, wherein the total fatty acid of the plant part comprises at least 0.1%, at least 0.13%, or at least 0.5% DPA. In a further embodiment, at least one plant part synthesizes DHA. In another embodiment, at least one plant part synthesizes DHA, wherein the total fatty acid of the plant part comprises at least 0.1%, at least 0.2%, or at least 0.5% DHA. In another embodiment, at least one plant part synthesizes at least one o3 C20 LC PUFA, wherein the total fatty acid of the plant part comprises at least 2.5%, or at least 4.1% 03 C20 LC-PUFA. In yet another embodiment, at least one plant part synthesizes EPA, wherein the efficiency of conversion of ALA to EPA in the plant part is at least 2% or at least 14.6%. In a further embodiment, at least one plant part synthesizes 03 polyunsaturated fatty acids that are the products of A6-desaturation of ALA and/or the products of A9 elongation of ALA, wherein the efficiency of conversion of ALA to said products in the plant part is at least 22% or at least 24%. In yet another embodiment, at least one plant part synthesizes DPA from EPA, wherein the efficiency of conversion of EPA to DPA in the plant part is at least 5% or at least 7%. With regard to transgenic seeds of the invention, in one embodiment EPA is synthesized in the seed and the total fatty acid of the seed comprises at least 1.5%, at least 2.1%, or at least 2.5% EPA. In another embodiment, DPA is synthesized in the seed and the total fatty acid of the seed comprises at least 0.1%, at least 0.13%, or at least 0.5% DPA. In a further embodiment, DHA is synthesized in the seed. In another embodiment, DHA is synthesized in the seed and the total fatty acid of the seed comprises at least 0.1%, at least 0.2%, or at least 0.5% DHA. In yet a further embodiment, at least one 03 C20 LC-PUFA is synthesized in the seed and the total fatty acid of the seed comprises at least 2.5%, or at least 4.1% 03 C20 LC-PUFA. In a further embodiment, EPA is synthesized in the seed and the efficiency of conversion of ALA to EPA in the seed is at least 2% or at least 14.6%. In another embodiment, 03 polyunsaturated fatty acids that are the products of A6-desaturation of ALA and/or the products of A9 elongation of ALA, are synthesized in the seed, and the efficiency of conversion of ALA to said products in the seed is at least 22% or at least 24%. In a further embodiment, DPA is synthesized from EPA in the seed and the efficiency of conversion of EPA to DPA in the seed is at least 5% or at least 7%.

WO 2005/103253 PCT/AU2005/000571 40 Referring to extracts of the invention, in one embodiment, the total fatty acid content of the extract comprises at least 1.5%, at least 2.1%, or at least 2.5% EPA. In another embodiment, the total fatty acid content of the extract comprises at least 0.1%, at least 0.13%, or at least 0.5% DPA. In a further embodiment, the extract comprises DHA. In another embodiment, the total fatty acid content of the extract comprises at least 0.1%, at least 0.2%, or at least 0.5% DHA. In another embodiment, the total fatty acid content of the extract comprises at least 2.5%, or at least 4.1% o3 C20 LC-PUFA. In yet a further embodiment, the extract comprises ARA, EPA, DPA, DHA, or any mixture of these in the triacylglycerols. With regard to methods of the invention for producing a LC-PUFA, in on embodiment, the cell comprises at least one C20 LC-PUFA, and the total fatty acid of the cell comprises at least 2%, at least 4.7%, or at least 7.9% C20 LC-PUFA. In another embodiment, the cell comprises at least one o3 C20 LC-PUFA, and the total fatty acid of the cell comprises at least 2.5%, or at least 4.1% Co3 C20 LC-PUFA. In a further embodiment, the cell comprises o3 polyunsaturated fatty acids that are the products of A6-desaturation of ALA and/or the products of A9 elongation of ALA, and the efficiency of conversion of ALA to said products in the cell is at least 22% or at least 24%. In yet another embodiment, the cell comprises DPA, and the total fatty acid of the cell comprises at least 0.1%, at least 0.13%, or at least 0.5% DPA. In a further embodiment, the cell comprises DPA, and the efficiency of conversion of EPA to DPA in the cell is at least 5% or at least 7%. In another embodiment, the cell comprises EPA, and wherein the total fatty acid of the cell comprises at least 1.5%, at least 2.1%, or at least 2.5% EPA. In a further embodiment, the cell comprises EPA, and the efficiency of conversion of ALA to EPA in the cell is at least 2% or at least 14.6%. Polypeptides In one aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO: 1, ii) a polypeptide comprising an amino acid sequence which is at least 40% identical to SEQ ID NO: 1, and iii) a biologically active fragment of i) or ii), WO 2005/103253 PCT/AU2005/000571 41 wherein the polypeptide has A8 desaturase activity. Preferably, the A8 desaturase does not also have A6 desaturase activity. In another aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:2, ii) a polypeptide comprising an amino acid sequence which is at least 60% identical to SEQ ID NO:2, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A5 elongase and/or A6 elongase activity. Preferably, the polypeptide has A5 elongase and A6 elongase activity, and wherein the polypeptide is more efficient at synthesizing DPA from EPA than it is at synthesizing ETA from SDA. More preferably, the polypeptide can be purified from algae. Furthermore, when expressed in yeast cells, is more efficient at elongating C20 LC PUFA than C22 LC-PUFA. In another aspect, the invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, ii) a polypeptide comprising an amino acid sequence which is at least 40% identical to SEQ ID NO:3, SEQ ID NO:85 or SEQ ID NO:86, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A9 elongase and/or A6 elongase activity. Preferably, the polypeptide has A9 elongase and A6 elongase activity. Preferably, the polypeptide is more efficient at synthesizing ETrA from ALA than it is at synthesizing ETA from SDA. Further, it is preferred that the polypeptide can be purified from algae or fungi. In yet another aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:4, ii) a polypeptide comprising an amino acid sequence which is at least 70% identical to SEQ ID NO:4, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A4 desaturase activity. In a further aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: WO 2005/103253 PCT/AU2005/000571 42 i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:60, ii) a polypeptide comprising an amino acid sequence which is at least 55% identical to SEQ ID NO:60, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A5 desaturase activity. In yet another aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:64, ii) a polypeptide comprising an amino acid sequence which is at least 90% identical to SEQ ID NO:64, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A6 desaturase activity. In yet another aspect, the present invention provides a substantially purified polypeptide selected from the group consisting of: i) a polypeptide comprising an amino acid sequence as provided in SEQ ID NO:88, ii) a polypeptide comprising an amino acid sequence which is at least 76% identical to SEQ ID NO:88, and iii) a biologically active fragment of i) or ii), wherein the polypeptide has A6 elongase activity. Preferably, in relation to any one of the above aspects, it is preferred that the polypeptide can be isolated from a species selected from the group consisting of Pavlova and Melosira. By "substantially purified polypeptide" we mean a polypeptide that has been at least partially separated from the lipids, nucleic acids, other polypeptides, and other contaminating molecules with which it is associated in its native state. Preferably, the substantially purified polypeptide is at least 60% free, preferably at least 75% free, and most preferably at least 90% free from other components with which they are naturally associated. Furthermore, the term "polypeptide" is used interchangeably herein with the term "protein". The % identity of a polypeptide is determined by GAP (Needleman and Wunsch, 1970) analysis (GCG program) with a gap creation penalty=5, and a gap extension penalty=0.3. Unless stated otherwise, the query sequence is at least 15 amino acids in length, and the GAP analysis aligns the two sequences over a region of at least 15 amino WO 2005/103253 PCT/AU2005/000571 43 acids. More preferably, the query sequence is at least 50 amino acids in length, and the GAP analysis aligns the two sequences over a region of at least 50 amino acids. Even more preferably, the query sequence is at least 100 amino acids in length and the GAP analysis aligns the two sequences over a region of at least 100 amino acids. With regard to the defined polypeptides/enzymes, it will be appreciated that % identity figures higher than those provided above will encompass preferred embodiments. Thus, where applicable, in light of the minimum % identity figures, it is preferred that the polypeptide comprises an amino acid sequence which is at least 60%, more preferably at least 65%, more preferably at least 70%, more preferably at least 75%, more preferably at least 76%, more preferably at least 80%, more preferably at least 85%, more preferably at least 90%, more preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, more preferably at least 99%, more preferably at least 99.1%, more preferably at least 99.2%, more preferably at least 99.3%, more preferably at least 99.4%, more preferably at least 99.5%, more preferably at least 99.6%, more preferably at least 99.7%, more preferably at least 99.8%, and even more preferably at least 99.9% identical to the relevant nominated SEQ IDNO. In a further embodiment, the present invention relates to polypeptides which are substantially identical to those specifically described herein. As used herein, with reference to a polypeptide the term "substantially identical" means the deletion, insertion and/or substitution of one or a few (for example 2, 3, or 4) amino acids whilst maintaining at least one activity of the native protein. As used herein, the term "biologically active fragment" refers to a portion of the defined polypeptide/enzyme which still maintains desaturase or elongase activity (whichever is relevant). Such biologically active fragments can readily be determined by serial deletions of the full length protein, and testing the activity of the resulting fragment. Amino acid sequence mutants/variants of the polypeptides/enzymes defined herein can be prepared by introducing appropriate nucleotide changes into a nucleic acid encoding the polypeptide, or by in vitro synthesis of the desired polypeptide. Such mutants include, for example, deletions, insertions or substitutions of residues within the amino acid sequence. A combination of deletion, insertion and substitution can be made to arrive at the final construct, provided that the final protein product possesses the desired characteristics.

WO 2005/103253 PCT/AU2005/000571 44 In designing amino acid sequence mutants, the location of the mutation site and the nature of the mutation will depend on characteristic(s) to be modified. The sites for mutation can be modified individually or in series, e.g., by (1) substituting first with conservative amino acid choices and then with more radical selections depending upon the results achieved, (2) deleting the target residue, or (3) inserting other residues adjacent to the located site. Amino acid sequence deletions generally range from about 1 to 30 residues, more preferably about 1 to 10 residues and typically about 1 to 5 contiguous residues. Substitution mutants have at least one amino acid residue in the polypeptide molecule removed and a different residue inserted in its place. The sites of greatest interest for substitutional mutagenesis include sites identified as the -active or binding site(s). Other sites of interest are those in which particular residues obtained from various strains or species are identical. These positions may be important for biological activity. These sites, especially those falling within a sequence of at least three other identically conserved sites, are preferably substituted in a relatively conservative manner. Such conservative substitutions are shown in Table 2. Furthermore, if desired, unnatural amino acids or chemical amino acid analogues can be introduced as a substitution or addition into the polypeptides of the present invention. Such amino acids include, but are not limited to, the D-isomers of the common amino acids, 2,4-diaminobutyric acid, a-amino isobutyric acid, 4-aminobutyric acid, 2-aminobutyric acid, 6-amino hexanoic acid, 2-amino isobutyric acid, 3-amino propionic acid, ornithine, norleucine, norvaline, hydroxyproline, sarcosine, citrulline, homocitrulline, cysteic acid, t-butylglycine, t-butylalanine, phenylglycine, cyclohexylalanine, p-alanine, fluoro-amino acids, designer amino acids such as p-methyl amino acids, Ca-methyl amino acids, Na-methyl amino acids, and amino acid analogues in general. Also included within the scope of the invention are polypeptides of the present invention which are differentially modified during or after synthesis, e.g., by biotinylation, benzylation, glycosylation, acetylation, phosphorylation, amidation, derivatization by known protecting/blocking groups, proteolytic cleavage, linkage to an antibody molecule or other cellular ligand, etc. These modifications may serve to increase the stability and/or bioactivity of the polypeptide of the invention.

WO 2005/103253 PCT/AU2005/000571 45 TABLE 2. Exemplary substitutions. Original Exemplary Residue Substitutions Ala (A) val; leu; ile; gly Arg (R) lys Asn (N) gln; his Asp (D) glu Cys (C) ser Gln (Q) asn; his Glu (E) asp Gly (G) pro, ala His (H) asn; gln Ile (I) leu; val; ala Leu (L) ile; val; met; ala; phe Lys (K) arg Met (M) leu; phe Phe (F) leu; val; ala Pro (P) gly Ser (S) thr Thr (T) ser Trp(W) tyr Tyr (Y) trp; phe Val (V) ile; leu; met; phe, ala Polypeptides of the present invention can be produced in a variety of ways, including production and recovery of natural proteins, production and recovery of recombinant proteins, and chemical synthesis of the proteins. In one embodiment, an isolated polypeptide of the present invention is produced by culturing a cell capable of expressing the polypeptide under conditions effective to produce the polypeptide, and recovering the polypeptide. A preferred cell to culture is a recombinant cell of the present invention. Effective culture conditions include, but are not limited to, effective media, bioreactor, temperature, pH and oxygen conditions that permit protein production. An effective medium refers to any medium in which a cell is cultured to produce a WO 2005/103253 PCT/AU2005/000571 46 polypeptide of the present invention. Such medium typically comprises an aqueous medium having assimilable carbon, nitrogen and phosphate sources, and appropriate salts, minerals, metals and other nutrients, such as vitamins. Cells of the present invention can be cultured in conventional fermentation bioreactors, shake flasks, test tubes, microtiter dishes, and petri plates. Culturing can be carried out at a temperature, pH and oxygen content appropriate for a recombinant cell, Such culturing conditions are within the expertise of one of ordinary skill in the art. Polynucleotides In one aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:5 or SEQ ID NO:6; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 50% identical to SEQ ID NO:5 or SEQ ID NO:6; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In another aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:7 or SEQ ID NO:8; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 51% identical to SEQ ID NO:7 or SEQ ID NO:8; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In yet another aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:9 or SEQ ID NO:10; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 51% identical to SEQ ID NO:9 or SEQ ID NO:10; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In a preferred embodiment, the sequence encoding a polypeptide of the invention is nucleotides 31 to 915 or SEQ ID NO:9 or nucleotides 85 to 915 of SEQ ID NO:9.

WO 2005/103253 PCT/AU2005/000571 47 In a further aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:11, SEQ ID NO:12 or SEQ ID NO:13; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 70% identical to SEQ ID NO: 11, SEQ ID NO:12 or SEQ ID NO:13; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In another aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:58 or SEQ ID NO:59; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 55% identical to SEQ ID NO:58 or SEQ ID NO:59; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In another aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:63; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 90% identical to SEQ ID NO:63; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. In another aspect, the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:89; ii) a sequence encoding a polypeptide of the invention; iii) a sequence of nucleotides which is at least 76% identical to SEQ ID NO:89; and iv) a sequence which hybridizes to any one of i) to iii) under high stringency conditions. The present inventors are also the first to isolate polynucleotide encoding a keto acyl synthase-like fatty acid elongase from a non-higher plant.

WO 2005/103253 PCT/AU2005/000571 48 Accordingly, in a further aspect the present invention provides an isolated polynucleotide comprising a sequence of nucleotides selected from the group consisting of: i) a sequence of nucleotides as provided in SEQ ID NO:55; ii) a sequence of nucleotides which is at least 40% identical to SEQ ID NO:55; and iii) a sequence which hybridizes to i) or ii) under high stringency conditions. By an "isolated polynucleotide", including DNA, RNA, or a combination of these, single or double stranded, in the sense or antisense orientation or a combination of both, dsRNA or otherwise, we mean a polynucleotide which is at least partially separated from the polynucleotide sequences with which it is associated or linked in its native state. Preferably, the isolated polynucleotide is at least 60% free, preferably at least 75% free, and most preferably at least 90% free from other components with which they are naturally associated. Furthermore, the term "polynucleotide" is used interchangeably herein with the term "nucleic acid molecule". The % identity of a polynucleotide is determined by GAP (Needleman and Wunsch, 1970) analysis (GCG program) with a gap creation penalty=5, and a gap extension penalty=0.3. Unless stated otherwise, the query sequence is at least 45 nucleotides in length, and the GAP analysis aligns the two sequences over a region of at least 45 nucleotides. Preferably, the query sequence is at least 150 nucleotides in length, and the GAP analysis aligns the two sequences over a region of at least 150 nucleotides. More preferably, the query sequence is at least 300 nucleotides in length and the GAP analysis aligns the two sequences over a region of at least 300 nucleotides. With regard to the defined polynucleotides, it will be appreciated that % identity figures higher than those provided above will encompass preferred embodiments. Thus, where applicable, in light of the minimum % identity figures, it is preferred that the polynucleotide comprises a nucleotide sequence which is at least 60%, more preferably at least 65%, more preferably at least 70%, more preferably at least 75%, more preferably at least 76%, more preferably at least 80%, more preferably at least 85%, more preferably at least 90%, more preferably at least 91%, more preferably at least 92%, more preferably at least 93%, more preferably at least 94%, more preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, more preferably at least 99%, more preferably at least 99.1%, more preferably at least 99.2%, more preferably at least 99.3%, more preferably at least 99.4%, more preferably at least 99.5%, more preferably at least 99.6%, more preferably at least 99.7%, more preferably at least WO 2005/103253 PCT/AU2005/000571 49 99.8%, and even more preferably at least 99.9% identical to the relevant nominated SEQ ID NO. In a further embodiment, the present invention relates to polynucleotides which are substantially identical to those specifically described herein. As used herein, with reference to a polynucleotide the term "substantially identical" means the substitution of one or a few (for example 2, 3, or 4) nucleotides whilst maintaining at least one activity of the native protein encoded by the polynucleotide. In addition, this term includes the addition or deletion of nucleotides which results in the increase or decrease in size of the encoded native protein by one or a few (for example 2, 3, or 4) amino acids whilst maintaining at least one activity of the native protein encoded by the polynucleotide. Oligonucleotides of the present invention can be RNA, DNA, or derivatives of either. The minimum size of such oligonucleotides is the size required for the formation of a stable hybrid between an oligonucleotide and a complementary sequence on a nucleic acid molecule of the present invention. Preferably, the oligonucleotides are at least 15 nucleotides, more preferably at least 18 nucleotides, more preferably at least 19 nucleotides, more preferably at least 20 nucleotides, even more preferably at least 25 nucleotides in length. The present invention includes oligonucleotides that can be used as, for example, probes to identify nucleic acid molecules, or primers to produce nucleic acid molecules. Oligonucleotide of the present invention used as a probe are typically conjugated with a label such as a radioisotope, an enzyme, biotin, a fluorescent molecule or a chemiluminescent molecule. Polynucleotides and oligonucleotides of the present invention include those which hybridize under stringent conditions to a sequence provided as SEQ ID NO's: 5 to 13. As used herein, stringent conditions are those that (1) employ low ionic strength and high temperature for washing, for example, 0.015 M NaCI/0.0015 M sodium citrate/0.1% NaDodSO 4 at 50 0 C; (2) employ during hybridisation a denaturing agent such as formamide, for example, 50% (vol/vol) formamide with 0.1% bovine serum albumin, 0.1% Ficoll, 0.1% polyvinylpyrrolidone, 50 mM sodium phosphate buffer at pH 6.5 with 750 mM NaCl, 75 mM sodium citrate at 42 0 C; or (3) employ 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt's solution, sonicated salmon sperm DNA (50 g/ml), 0.1% SDS and 10% dextran sulfate at 42 0 C in 0.2 x SSC and 0.1% SDS. Polynucleotides of the present invention may possess, when compared to naturally occurring molecules, one or more mutations which are deletions, insertions, or substitutions of nucleotide residues. Mutants can be either naturally occurring (that is to WO 2005/103253 PCT/AU2005/000571 50 say, isolated from a natural source) or synthetic (for example, by performing site-directed mutagenesis on the nucleic acid). Also provided are antisense and/or catalytic nucleic acids (such as ribozymes) which hybridize to a polynucleotide of the invention, and hence inhibit the production of an encoded protein. Furthermore, provided are dsRNA molecules, particularly small dsRNA molecules with a double stranded region of about 21 nucleotides, which can be used in RNA interference to inhibit the production of a polypeptide of the invention in a cell. Such inhibitory molecules can be used to alter the types of fatty acids produced by a cell, such an animal cell, moss, or algael cell. The production of such antisense, catalytic nucleic acids and dsRNA molecules is well with the capacity of the skilled person (see for example, G. Hartmann and S. Endres, Manual of Antisense Methodology, Kluwer (1999); Haseloff and Gerlach, 1988; Perriman et a1., 1992; Shippy et al., 1999; Waterhouse et al. (1998); Smith et al. (2000); WO 99/32619, WO 99/53050, WO 99/49029, and WO 01/34815). Gene Constructs and Vectors One embodiment of the present invention includes a recombinant vector, which includes at least one isolated polynucleotide molecule encoding a polypeptide/enzyme defined herein, inserted into any vector capable of delivering the nucleic acid molecule into a host cell. Such a vector contains heterologous nucleic acid sequences, that is nucleic acid sequences that are not naturally found adjacent to nucleic acid molecules of the present invention and that preferably are derived from a species other than the species from which the nucleic acid molecule(s) are derived. The vector can be either RNA or DNA, either prokaryotic or eukaryotic, and typically is a virus or a plasmid. One type of recombinant vector comprises a nucleic acid molecule of the present invention operatively linked to an expression vector. As indicated above, the phrase operatively linked refers to insertion of a nucleic acid molecule into an expression vector in a manner such that the molecule is able to be expressed when transformed into a host cell. As used herein, an expression vector is a DNA or RNA vector that is capable of transforming a host cell and effecting expression of a specified nucleic acid molecule. Preferably, the expression vector is also capable of replicating within the host cell. Expression vectors can be either prokaryotic or eukaryotic, and are typically viruses or plasmids. Expression vectors of the present invention include any vectors that function (i.e., direct gene expression) in recombinant cells of the present invention, including in bacterial, fungal, endoparasite, arthropod, other animal, and plant cells. Preferred WO 2005/103253 PCT/AU2005/000571 51 expression vectors of the present invention can direct gene expression in yeast, animal or plant cells. In particular, expression vectors of the present invention contain regulatory sequences such as transcription control sequences, translation control sequences, origins of replication, and other regulatory sequences that are compatible with the recombinant cell and that control the expression of nucleic acid molecules of the present invention. In particular, recombinant molecules of the present invention include transcription control sequences. Transcription control sequences are sequences which control the initiation, elongation, and termination of transcription. Particularly important transcription control sequences are those which control transcription initiation, such as promoter, enhancer, operator and repressor sequences. Suitable transcription control sequences include any transcription control sequence that can function in at least one of the recombinant cells of the present invention. A variety of such transcription control sequences are known to those skilled in the art. Another embodiment of the present invention includes a recombinant cell comprising a host cell transformed with one or more recombinant molecules of the present invention. Transformation of a nucleic acid molecule into a cell can be accomplished by any method by which a nucleic acid molecule can be inserted into the cell. Transformation techniques include, but are not limited to, transfection, electroporation, microinjection, lipofection, adsorption, and protoplast fusion. A recombinant cell may remain unicellular or may grow into a tissue, organ or a multicellular organism. Transformed nucleic acid molecules can remain extrachromosomal or can integrate into one or more sites within a chromosome of the transformed (i.e., recombinant) cell in such a manner that their ability to be expressed is retained. Transgenic Plants and Parts Thereof The term "plant" as used herein as a noun refers to whole plants, but as used as an adjective refers to any substance which is present in, obtained from, derived from, or related to a plant, such as for example, plant organs (e.g. leaves, stems, roots, flowers), single cells (e.g. pollen), seeds, plant cells and the like. Plants provided by or contemplated for use in the practice of the present invention include both monocotyledons and dicotyledons. In preferred embodiments, the plants of the present invention are crop plants (for example, cereals and pulses, maize, wheat, potatoes, tapioca, rice, sorghum, millet, cassava, barley, or pea), or other legumes. The plants may be grown for production of edible roots, tubers, leaves, stems, flowers or fruit. The plants WO 2005/103253 PCT/AU2005/000571 52 may be vegetables or ornamental plants. The plants of the invention may be: corn (Zea mays), canola (Brassica napus, Brassica rapa ssp.), flax (Linum usitatissimum), alfalfa (Medicago sativa), rice (Oryza sativa), rye (Secale cerale), sorghum (Sorghum bicolour, Sorghum vulgare), sunflower (Helianthus annus), wheat (Tritium aestivum), soybean (Glycine max), tobacco (Nicotiana tabacum), potato. (Solanum tuberosum), peanuts (Arachis hypogaea), cotton (Gossypium hirsutum), sweet potato (Lopmoea batatus), cassava (Manihot esculenta), coffee (Cofea spp.), coconut (Cocos nucifera), pineapple (Anana comosus), citris tree (Citrus spp.), cocoa (Theobroma cacao), tea (Camellia senensis), banana (Musa spp.), avocado (Persea americana), fig (Ficus casica), guava (Psidium guajava), mango (Mangifer indica), olive (Olea europaea), papaya (Carica papaya), cashew (Anacardium occidentale), macadamia (Macadamia intergrifolia), almond (Prunus amygdalus), sugar beets (Beta vulgaris), oats, or barley. In one embodiment, the plant is an oilseed plant, preferably an oilseed crop plant. As used herein, an "oilseed plant" is a plant species used for the commercial production of oils from the seeds of the plant. The oilseed plant may be oil-seed rape (such as canola), maize, sunflower, soybean, sorghum, flax (linseed) or sugar beet. Furthermore, the oilseed plant may be other Brassicas, cotton, peanut, poppy, mustard, castor bean, sesame, safflower, or nut producing plants. The plant may produce high levels of oil in its fruit, such as olive, oil palm or coconut. Horticultural plants to which the present invention may be applied are lettuce, endive, or vegetable brassicas including cabbage, broccoli, or cauliflower. The present invention may be applied in tobacco, cucurbits, carrot, strawberry, tomato, or pepper. When the production of o3 LC-PUFA is desired it is preferable that the plant species which is to be transformed has an endogenous ratio of ALA to LA which is at least 1:1, more preferably at least 2:1. Examples include most, if not all, oilseeds such as linseed. This maximizes the amount of ALA substrate available for the production of SDA, ETA, ETrA, EPA, DPA and DHA. The plants produced using the methods of the invention may already be transgenic, and/or transformed with additional genes to those described in detail herein. In one embodiment, the transgenic plants of the invention also produce a recombinant o3 desaturase. The presence of a recombinant o3 desaturase increases the ratio of ALA to LA in the plants which, as outlined in the previous paragraph, maximizes the production of LC-PUFA such as SDA, ETA, ETrA, EPA, DPA and DHA. Grain plants that provide seeds of interest include oil-seed plants and leguminous plants. Seeds of interest include grain seeds, such as corn, wheat, barley, rice, sorghum, rye, etc. Leguminous plants include beans and peas. Beans include guar, locust bean, WO 2005/103253 PCT/AU2005/000571 53 fenugreek, soybean, garden beans, cowpea, mungbean, lima bean, fava bean, lentils, chickpea, etc. The term "extract or portion thereof' refers to any part of the plant. "Portion" generally refers to a specific tissue or organ such as a seed'or root, whereas an "extract" typically involves the disruption of cell walls and possibly the partial purification of the resulting material. Naturally, the "extract or portion thereof' will comprise at least one LC-PUFA. Extracts can be prepared using standard techniques of the art. Transgenic plants, as defined in the context of the present invention include plants and their progeny which have been genetically modified using recombinant techniques. This would generally be to cause or enhance production of at least one protein/enzyme defined herein in the desired plant or plant organ. Transgenic plant parts include all parts and cells of said plants such as, for example, cultured tissues, callus, protoplasts. Transformed plants contain genetic material that they did not contain prior to the transformation. The genetic material is preferably stably integrated into the genome of the plant. The introduced genetic material may comprise sequences that naturally occur in the same species but in a rearranged order or in a different arrangement of elements, for example an antisense sequence. Such plants are included herein in "transgenic plants". A "non-transgenic plant" is one which has not been genetically modified with the introduction of genetic material by recombinant DNA techniques. In a preferred embodiment, the transgenic plants are homozygous for each and every gene that has been introduced (transgene) so that their progeny do not segregate for the desired phenotype. Several techniques exist for introducing foreign genetic material into a plant cell. Such techniques include acceleration of genetic material coated onto microparticles directly into cells (see, for example, US 4,945,050 and US 5,141,131). Plants may be transformed using Agrobacterium technology (see, for example, US 5,177,010, US 5,104,310, US 5,004,863, US 5,159,135). Electroporation technology has also been used to transform plants (see, for example, WO 87/06614, US 5,472,869, 5,384,253, WO 92/09696 and WO 93/21335). In addition to numerous technologies for transforming plants, the type of tissue which is contacted with the foreign genes may vary as well. Such tissue would include but would not be limited to embryogenic tissue, callus tissue type I and II, hypocotyl, meristem, and the like. Almost all plant tissues may be transformed during development and/or differentiation using appropriate techniques described herein. A number of vectors suitable for stable transfection of plant cells or for the establishment of transgenic plants have been described in, e.g., Pouwels et al., Cloning Vectors: A Laboratory Manual, 1985, supp. 1987; Weissbach and Weissbach, Methods WO 2005/103253 PCT/AU2005/000571 54 for Plant Molecular Biology, Academic Press, 1989; and Gelvin et al., Plant Molecular Biology Manual, Kluwer Academic Publishers, 1990. Typically, plant expression vectors include, for example, one or more cloned plant genes under the transcriptional control of 5' and 3' regulatory sequences and a dominant selectable marker. Such plant expression vectors also can contain a promoter regulatory region (e.g., a regulatory region controlling inducible or constitutive, environmentally- or developmentally-regulated, or cell- or tissue-specific expression), a transcription initiation start site, a ribosome binding site, an RNA processing signal, a transcription termination site, and/or a polyadenylation signal. Examples of plant promoters include, but are not limited to ribulose-1,6 bisphosphate carboxylase small subunit, beta-conglycinin promoter, phaseolin promoter, high molecular weight glutenin (HMW-GS) promoters, starch biosynthetic gene promoters, ADH promoter, heat-shock promoters and tissue specific promoters. Promoters may also contain certain enhancer sequence elements that may improve the transcription efficiency. Typical enhancers include but are not limited to Adh-intron 1 and Adh-intron 6. Constitutive promoters direct continuous gene expression in all cells types and at all times (e.g., actin, ubiquitin, CaMV 35S). Tissue specific promoters are responsible for gene expression in specific cell or tissue types, such as the leaves or seeds (e.g., zein, oleosin, napin, ACP, globulin and the like) and these promoters may also be used. Promoters may also be active during a certain stage of the plants' development as well as active in plant tissues and organs. Examples of such promoters include but are not limited to pollen-specific, embryo specific, corn silk specific, cotton fibre specific, root specific, seed endosperm specific promoters and the like. In a particularly preferred embodiment, the promoter directs expression in tissues and organs in which lipid and oil biosynthesis take place, particularly in seed cells such as endosperm cells and cells of the developing embryo. Promoters which are suitable are the oilseed rape napin gene promoter (US 5,608,152), the Vicia faba USP promoter (Baumlein et al., 1991), the Arabidopsis oleosin promoter (WO 98/45461), the Phaseolus vulgaris phaseolin promoter (US 5,504,200), the Brassica Bce4 promoter (WO 91/13980) or the legumin B4 promoter (Baumlein et al., 1992), and promoters which lead to the seed-specific expression in monocots such as maize, barley, wheat, rye, rice and the like. Notable promoters which are suitable are the barley lpt2 or lptl gene promoter (WO 95/15389 and WO 95/23230) or the promoters described in WO 99/16890 (promoters from the barley hordein gene, the rice glutelin gene, the rice oryzin gene, the rice prolamin gene, the wheat gliadin gene, the wheat glutelin gene, the maize zein gene, WO 2005/103253 PCT/AU2005/000571 55 the oat glutelin gene, the sorghum kasirin gene, the rye secalin gene). Other promoters include those described by Broun et al. (1998) and US 20030159173. Under certain circumstances it may be desirable to use an inducible promoter. An inducible promoter is responsible for expression of genes in response to a specific signal, such as: physical stimulus (heat shock genes); light (RUBP carboxylase); hormone (Em); metabolites; and stress. Other desirable transcription and translation elements that function in plants may be used. In addition to plant promoters, promoters from a variety of sources can be used efficiently in plant cells to express foreign genes. For example, promoters of bacterial origin, such as the octopine synthase promoter, the nopaline synthase promoter, the mannopine synthase promoter; promoters of viral origin, such as the cauliflower mosaic virus (35S and 19S) and the like may be used. It will be apparent that transgenic plants adapted for the production of LC-PUFA as described herein, in particular DHA, can either be eaten directly or used as a source for the extraction of essential fatty acids, of which DHA would be a constituent. As used herein, "germination" refers to the emergence of the root tip from the seed coat after imbibition. "Germination rate" refers to the percentage of seeds in a population which have germinated over a period of time, for example 7 or 10 days, after imbibition. A population of seeds can be assessed daily over several days to determine the germination percentage over time. With regard to seeds of the present invention, as used herein the term "germination rate which is substantially the same" means that the germination rate of the transgenic seeds is at least 60%, more preferably at least 80%, and even more preferably at least 90%, that of isogenic non-transgenic seeds. Gemination rates can be calculated using techniques known in the art. With further regard to seeds of the present invention, as used herein the term "timing of germination of the seed is substantially the same" means that the timing of germination of the transgenic seeds is at least 60%, more preferably at least 80%, and even more preferably at least 90%, that of isogenic non-transgenic seeds. Timing of gemination can be calculated using techniques known in the art. The present inventors have found that at least in some circumstances the production of LC-PUFA in recombinant plant cells is enhanced when the cells are homozygous for the transgene. As a result, it is preferred that the recombinant plant cell, preferably the transgenic plant, is homozygous for at least one desaturase and/or elongase gene. In one embodiment, the cells/plant are homozygous for the zebrafish A6/A5 desaturase and/or the C. elegans elongase.

WO 2005/103253 PCT/AU2005/000571 56 Trans genic non-human animals Techniques for producing transgenic animals are well known in the art. A useful general textbook on this subject is Houdebine, Transgenic animals - Generation and Use (Harwood Academic, 1997). Heterologous DNA can be introduced, for example, into fertilized mammalian ova. For instance, totipotent or pluripotent stem cells can be transformed by microinjection, calcium phosphate mediated precipitation, liposome fusion, retroviral infection or other means, the transformed cells are then introduced into the embryo, and the embryo then develops into a transgenic animal. In a highly preferred method, developing embryos are infected with a retrovirus containing the desired DNA, and transgenic animals produced from the infected embryo. In a most preferred method, however, the appropriate DNAs are coinjected into the pronucleus or cytoplasm of embryos, preferably at the single cell stage, and the embryos allowed to develop into mature transgenic animals. Another method used to produce a transgenic animal involves microinjecting a nucleic acid into pro-nuclear stage eggs by standard methods. Injected eggs are then cultured before transfer into the oviducts of pseudopregnant recipients. Transgenic animals may also be produced by nuclear transfer technology. Using this method, fibroblasts from donor animals are stably transfected with a plasmid incorporating the coding sequences for a binding domain or binding partner of interest under the control of regulatory. Stable transfectants are then fused to enucleated oocytes, cultured and transferred into female recipients. Feedstuffs The present invention includes compositions which can be used as feedstuffs. For purposes of the present invention, "feedstuffs" include any food or preparation for human or animal consumption (including for enteral and/or parenteral consumption) which when taken into the body (a) serve to nourish or build up tissues or supply energy; and/or (b) maintain, restore or support adequate nutritional status or metabolic function. Feedstuffs of the invention include nutritional compositions for babies and/or young children. Feedstuffs of the invention comprise, for example, a cell of the invention, a plant of the invention, the plant part of the invention, the seed of the invention, an extract of the invention, the product of the method of the invention, the product of the fermentation process of the invention, or a composition along with a suitable carrier(s). The term "carrier" is used in its broadest sense to encompass any component which may or may WO 2005/103253 PCT/AU2005/000571 57 not have nutritional value. As the skilled addressee will appreciate, the carrier must be suitable for use (or used in a sufficiently low concentration) in a feedstuff such that it does not have deleterious effect on an organism which consumes the feedstuff. The feedstuff of the present invention comprises an oil, fatty acid ester, or fatty acid produced directly or indirectly by use of the methods, cells or plants disclosed herein. The composition may either be in a solid or liquid form. Additionally, the composition may include edible macronutrients, vitamins, and/or minerals in amounts desired for a particular use. The amounts of these ingredients will vary depending on whether the composition is intended for use with normal individuals or for use with individuals having specialized needs, such as individuals suffering from metabolic disorders and the like. Examples of suitable carriers with nutritional value include, but are not limited to, macronutrients such as edible fats, carbohydrates and proteins. Examples of such edible fats include, but are not limited to, coconut oil, borage oil, fungal oil, black current oil, soy oil, and mono- and diglycerides. Examples of such carbohydrates include (but are not limited to): glucose, edible lactose, and hydrolyzed search. Additionally, examples of proteins which may be utilized in the nutritional composition of the invention include (but are not limited to) soy proteins, electrodialysed whey, electrodialysed skim milk, milk whey, or the hydrolysates of these proteins. With respect to vitamins and minerals, the following may be added to the feedstuff compositions of the present invention: calcium, phosphorus, potassium, sodium, chloride, magnesium, manganese, iron, copper, zinc, selenium, iodine, and Vitamins A, E, D, C, and the B complex. Other such vitamins and minerals may also be added. The components utilized in the feedstuff compositions of the present invention can be of semi-purified or purified origin. By semi-purified or purified is meant a material which has been prepared by purification of a natural material or by de novo synthesis. A feedstuff composition of the present invention may also be added to food even when supplementation of the diet is not required. For example, the composition may be added to food of any type, including (but not limited to): margarine, modified butter, cheeses, milk, yogurt, chocolate, candy, snacks, salad oils, cooking oils, cooking fats, meats, fish and beverages. The genus Saccharomyces spp is used in both brewing of beer and wine making and also as an agent in baking, particularly bread. Yeast is a major constituent of vegetable extracts. Yeast is also used as an additive in animal feed. It will be apparent that genetically engineered yeast strains can be provided which are adapted to synthesise LC-PUFA as described herein. These yeast strains can then be used in food stuffs and in WO 2005/103253 PCT/AU2005/000571 58 wine and beer making to provide products which have enhanced fatty acid content and in particular DHA content. Additionally, LC-PUFA produced in accordance with the present invention or host cells transformed to contain and express the subject genes may also be used as animal food supplements to alter an animal's tissue or milk fatty acid composition to one more desirable for human or animal consumption. Examples of such animals include sheep, cattle, horses and the like. Furthermore, feedstuffs of the invention can be used in aquaculture to increase the levels of LC-PUFA in fish for human or animal consumption. In mammals, the so-called "Sprecher" pathway converts DPA to DHA by three reactions, independent of a A7 elongase, A4 desaturase, and a beta-oxidation step (Sprecher et al., 1995) (Figure 1). Thus, in feedstuffs for mammal consumption, for example formulations for the consumption by human infants, it may only be necessary to provide DPA produced using the methods of the invention as the mammalian subject should be able to fulfill its nutritional needs for DHA by using the "Sprecher" pathway to convert DPA to DHA. As a result, in an embodiment of the present invention, a feedstuff described herein for mammalian consumption at least comprises DPA, wherein at least one enzymatic reaction in the production of DPA was performed by a recombinant enzyme in a cell. Compositions The present invention also encompasses compositions, particularly pharmaceutical compositions, comprising one or more of the fatty acids and/or resulting oils produced using the methods of the invention. A pharmaceutical composition may comprise one or more of the LC-PUFA and/or oils, in combination with a standard, well-known, non-toxic pharmaceutically-acceptable carrier, adjuvant or vehicle such as phosphate-buffered saline, water, ethanol, polyols, vegetable oils, a wetting agent or an emulsion such as a water/oil emulsion. The composition may be in either a liquid or solid form. For example, the composition may be in the form of a tablet, capsule, ingestible liquid or powder, injectible, or topical ointment or cream. Proper fluidity can be maintained, for example, by the maintenance of the required particle size in the case of dispersions and by the use of surfactants. It may also be desirable to include isotonic agents, for example, sugars, sodium chloride, and the like. Besides such inert diluents, the composition can also include adjuvants, such as wetting agents, emulsifying and suspending agents, sweetening agents, flavoring agents and perfuming agents.

WO 2005/103253 PCT/AU2005/000571 59 Suspensions, in addition to the active compounds, may comprise suspending agents such as ethoxylated isostearyl alcohols, polyoxyethylene sorbitol and sorbitan esters, microcrystalline cellulose, aluminum metahydroxide, bentonite, agar-agar, and tragacanth or mixtures of these substances. Solid dosage forms such as tablets and capsules can be prepared using techniques well known in the art. For example, LC-PUFA produced in accordance with the present invention can be tableted with conventional tablet bases such as lactose, sucrose, and cornstarch in combination with binders such as acacia, cornstarch or gelatin, disintegrating agents such as potato starch or alginic acid, and a lubricant such as stearic acid or magnesium stearate. Capsules can be prepared by incorporating these excipients into a gelatin capsule along with antioxidants and the relevant LC-PUFA(s). For intravenous administration, the PUFA produced in accordance with the present invention or derivatives thereof may be incorporated into commercial formulations. A typical dosage of a particular fatty acid is from 0.1 mg to 20 g, taken from one to five times per day (up to 100 g daily) and is preferably in the range of from about 10 mg to about 1, 2, 5, or 10 g daily (taken in one or multiple doses). As known in the art, a minimum of about 300 mg/day of LC-PUFA is desirable. However, it will be appreciated that any amount of LC-PUFA will be beneficial to the subject. Possible routes of administration of the pharmaceutical compositions of the present invention include, for example, enteral (e.g., oral and rectal) and parenteral. For example, a liquid preparation may be administered orally or rectally. Additionally, a homogenous mixture can be completely dispersed in water, admixed under sterile conditions with physiologically acceptable diluents, preservatives, buffers or propellants to form a spray or inhalant. The dosage of the composition to be administered to the patient may be determined by one of ordinary skill in the art and depends upon various factors such as weight of the patient, age of the patient, overall health of the patient, past history of the patient, immune status of the patient, etc. Additionally, the compositions of the present invention may be utilized for cosmetic purposes. It may be added to pre-existing cosmetic compositions such that a mixture is formed or a LC-PUFA produced according to the subject invention may be used as the sole "active" ingredient in a cosmetic composition.

WO 2005/103253 PCT/AU2005/000571 60 Medical, Veterinary, Agricultural and Aquacultural Uses The present invention also includes the treatment of various disorders by use of the pharmaceutical and/or feedstuff compositions described herein. In particular, the compositions of the present invention may be used to treat restenosis after angioplasty. Furthermore, symptoms of inflammation, rheumatoid arthritis, asthma and psoriasis may also be treated with the compositions (including feedstuffs) of the invention. Evidence also indicates that LC-PUFA may be involved in calcium metabolism; thus, the compositions of the present invention may be utilized in the treatment or prevention of osteoporosis and of kidney or urinary tract stones. Additionally, the compositions of the present invention may also be used in the treatment of cancer. Malignant cells have been shown to have altered fatty acid compositions. Addition of fatty acids has been shown to slow their growth, cause cell death and increase their susceptibility to chemotherapeutic agents. Moreover, the compositions of the present invention may also be useful for treating cachexia associated with cancer. The compositions of the present invention may also be used to treat diabetes as altered fatty acid metabolism and composition have been demonstrated in diabetic animals. Furthermore, the compositions of the present invention, comprising LC-PUFA produced either directly or indirectly through the use of the cells of the invention, may also be used in the treatment of eczema and in the reduction of blood pressure. Additionally, the compositions of the present invention may be used to inhibit platelet aggregation, to induce vasodilation, to reduce cholesterol levels, to inhibit proliferation of vessel wall smooth muscle and fibrous tissue, to reduce or to prevent gastrointestinal bleeding and other side effects of non-steroidal anti-inflammatory drugs (US 4,666,701), to prevent or to treat endometriosis and premenstrual syndrome (US 4,758,592), and to treat myalgic encephalomyelitis and chronic fatigue after viral infections (US 5,116,871). Further uses of the compositions of the present invention include, but are not limited to, use in the treatment or prevention of cardiac arrhythmia's, angioplasty, AIDS, multiple sclerosis, Crohn's disease, schizophrenia, foetal alcohol syndrome, attention deficient hyperactivity disorder, cystic fibrosis, phenylketonuria, unipolar depression, aggressive hostility, adrenoleukodystophy, coronary heart disease, hypertension, obesity, Alzheimer's disease, chronic obstructive pulmonary disease, ulcerative colitis or an ocular disease, as well as for maintenance of general health. Furthermore, the above-described pharmaceutical and nutritional compositions may be utilized in connection with animals (i.e., domestic or non-domestic, including WO 2005/103253 PCT/AU2005/000571 61 mammals, birds, reptiles, lizards, etc.), as well as humans, as animals experience many of the same needs and conditions as humans. For example, the oil or fatty acids of the present invention may be utilized in animal feed supplements, animal feed substitutes, animal vitamins or in animal topical ointments. Compositions such as feedstuffs of the invention can also be used in aquaculture to increase the levels of LC-PUFA in fish for human or animal consumption. Any amount of LC-PUFA will be beneficial to the subject. However, it is preferred that an "amount effective to treat" the condition of interest is administered to the subject. Such dosages to effectively treat a condition which would benefit from administration of a LC-PUFA are known those skilled in the art. As an example, a dose of at least 300 mg/day of LC-PUFA for at least a few weeks, more preferably longer would be suitable in many circumstances. Antibodies The invention also provides monoclonal and/or polyclonal antibodies which bind specifically to at least one polypeptide of the invention or a fragment thereof. Thus, the present invention further provides a process for the production of monoclonal or polyclonal antibodies to polypeptides of the invention. The term "binds specifically" refers to the ability of the antibody to bind to at least one protein of the present invention but not other proteins present in a recombinant cell, particularly a recombinant plant cell, of the invention. As used herein, the term "epitope" refers to a region of a protein of the invention which is bound by the antibody. An epitope can be administered to an animal to generate antibodies against the epitope, however, antibodies of the present invention preferably specifically bind the epitope region in the context of the entire protein. If polyclonal antibodies are desired, a selected mammal (e.g., mouse, rabbit, goat, horse, etc.) is immunised with an immunogenic polypeptide. Serum from the immunised animal is collected and treated according to known procedures. If serum containing polyclonal antibodies contains antibodies to other antigens, the polyclonal antibodies can be purified by immunoaffinity chromatography. Techniques for producing and processing polyclonal antisera are known in the art. In order that such antibodies may be made, the invention also provides polypeptides of the invention or fragments thereof haptenised to another polypeptide for use as immunogens in animals or humans. Monoclonal antibodies directed against polypeptides of the invention can also be readily produced by one skilled in the art. The general methodology for making WO 2005/103253 PCT/AU2005/000571 62 monoclonal antibodies by hybridomas is well known. Immortal antibody-producing cell lines can be created by cell fusion, and also by other techniques such as direct transformation of B lymphocytes with oncogenic DNA, or transfection with Epstein-Barr virus. Panels of monoclonal antibodies produced can be screened for various properties; i.e., for isotype and epitope affinity. An alternative technique involves screening phage display libraries where, for example the phage express scFv fragments on the surface of their coat with a large variety of complementarity determining regions (CDRs). This technique is well known in the art. For the purposes of this invention, the term "antibody", unless specified to the contrary, includes fragments of whole antibodies which retain their binding activity for a target antigen. Such fragments include Fv, F(ab') and F(ab') 2 fragments, as well as single chain antibodies (scFv). Furthermore, the antibodies and fragments thereof may be humanised antibodies, for example as described in EP-A-239400. Antibodies of the invention may be bound to a solid support and/or packaged into kits in a suitable container along with suitable reagents, controls, instructions and the like. Preferably, antibodies of the present invention are detectably labeled. Exemplary detectable labels that allow for direct measurement of antibody binding include radiolabels, fluorophores, dyes, magnetic beads, chemiluminescers, colloidal particles, and the like. Examples of labels which permit indirect measurement of binding include enzymes where the substrate may provide for a coloured or fluorescent product. Additional exemplary detectable labels include covalently bound enzymes capable of providing a detectable product signal after addition of suitable substrate. Examples of suitable enzymes for use in conjugates include horseradish peroxidase, alkaline phosphatase, malate dehydrogenase and the like. Where not commercially available, such antibody-enzyme conjugates are readily produced by techniques known to those skilled in the art. Further exemplary detectable labels include biotin, which binds with high affinity to avidin or streptavidin; fluorochromes (e.g., phycobiliproteins, phycoerythrin and allophycocyanins; fluorescein and Texas red), which can be used with a fluorescence activated cell sorter; haptens; and the like. Preferably, the detectable label allows for direct measurement in a plate luminometer, e.g., biotin. Such labeled antibodies can be used in techniques known in the art to detect proteins of the invention.

WO 2005/103253 PCT/AU2005/000571 63 EXAMPLES Example 1. Materials and Methods Culturing Pavlova salina Pavlova salina isolates including strain CS-49 from the CSIRO Collection of Living Microalgae was cultivated under standard culture conditions (http://www.marine.csiro.au/microalgae). A stock culture from the Collection was sub cultured and scaled-up in a dilution of 1 in 10 over consecutive transfers in 1 L Erlenmeyer flasks and then into 10 L polycarbonate carboys. The culture medium was f/2, a modification of Guillard and Ryther's (1962) f medium containing half-strength nutrients, with a growth temperature of 20J1C. Other culturing conditions included a light intensity of 100 pimol. photons PAR.m- 2 .s- 1 , 12:12 hour light:dark photoperiod, and bubbling with 1% C02 in air at a rate of 200 mL.L1.min-. Yeast culturing and feeding with precursor fatty acids Plasmids were introduced into yeast by heat shock and transformants were selected on yeast minimal medium (YMMv) plates containing 2% raffinose as the sole carbon source. Clonal inoculum cultures were established in liquid YMM with 2% raffmose as the sole carbon source. Experimental cultures in were inoculated from these, in YMM + 1% NP-40, to an initial OD 600 of ~ 0.3. Cultures were grown at 30*C with shaking (-60 rpm) until OD 6 00 was approximately 1.0. At this point galactose was added to a final concentration of 2% and precursor fatty acids were added to a final concentration of 0.5mM. Cultures were incubated at 20C with shaking for a further 48 hours prior to harvesting by centrifugation. Cell pellets were washed with 1% NP-40, 0.5% NP-40 and water to remove any unincorporated fatty acids from the surface of the cells. Gas chromatography (GC) analysis of fatty acids Fatty acid preparation Fatty acid methyl esters (FAME) were formed by transesterification of the centrifuged yeast pellet or Arabidopsis seeds by heating with MeOH-CHCl 3 -HCl (10:1:1, v/v/v) at 90-100 OC for 2 h in a glass test tube fitted with a Teflon-lined screw-cap. FAME were extracted into hexane-dichloromethane (4:1, v/v) and analysed by GC and

GC-MS.

WO 2005/103253 PCT/AU2005/000571 64 Capillary gas-liquid chromatography (GC) FAME were analysed with Hewlett Packard (HP) 5890 GC or Agilent 6890 gas chromatograph fitted with HP 7673A or 6980 series automatic injectors respectively and a flame-ionization detector (FID). Injector and detector temperatures were 290 0 C and 310 0 C respectively. FAME samples were injected at 50 0 C onto a non-polar cross-linked methyl-silicone fused-silica capillary column (HP-5; 50 m x 0.32 mm i.d.; 0.17 pm film thickness.). After 1 min, the oven temperature was raised to 210 0 C at 30 0 C min, then to a final temperature of 280 0 C at 3 0 C min~ 1 where it was kept for 5 min. Helium was the carrier gas with a column head pressure of 65 KPa and the purge opened 2 min after injection. Identification of peaks was based on comparison of relative retention time data with standard FAME with confirmation using mass-spectrometry. For quantification Empower software (Waters) or Chemstation (Agilent) was used to integrate peak areas. Gas chromatography-mass spectrometry (GC-MS) GC-MS was carried out on a Finnigan GCQ Plus GC-MS ion-trap fitted with on column injection set at 40C. Samples were injected using an AS2000 auto sampler onto a retention gap attached to an HP-5 Ultra 2 bonded-phase column (50 in x 0.32 mm i.d. x 0.17 pm film thickness). The initial temperature of 45 0 C was held for 1 min, followed by temperature programming at 30 0 C.min' to 140 0 C then at 3 0 C.min' to 3 10 0 C where it was held for 12 min. Helium was used as the carrier gas. Mass spectrometer operating conditions were: electron impact energy 70 eV; emission current 250 pamp, transfer line 310 0C; source temperature 2400C; scan rate 0.8 scans.s - and mass range 40-650 Dalton. Mass spectra were acquired and processed with XcaliburM software. Construction of P. salina cDNA library mRNA, for the construction of a cDNA library, was isolated from P. salina cells using the following method. 2 g (wet weight) of P. salina cells were powdered using a mortar and pestle in liquid nitrogen and sprinkled slowly into a beaker containing 22 ml of extraction buffer that was being stirred constantly. To this, 5% insoluble polyvinylpyrrolidone, 90mM 2-mercaptoethanol, and 10mM dithiotheitol were added and the mixture stirred for a further 10 minutes prior to being transferred to a CorexTM tube. 18.4 ml of 3M ammonium acetate was added and mixed well. The sample was then centrifuged at 6000xg for 20 minutes at 4"C. The supernatant was transferred to a new tube and nucleic acid precipitated by the addition of 0.1 volume of 3M NaAc (pH 5.2) and 0.5 volume of cold isopropanol. After a 1 hour incubation at -20*C, the sample was WO 2005/103253 PCT/AU2005/000571 65 centrifuged at 6000xg for 30 minutes in a swing rotor. The pellet was resuspended in 1 ml of water extracted with phenol/chloroform. The aqueous layer was transferred to a new tube and nucleic acids were precipitated once again by the addition of 0.1 volume 3M NaAc (pH 5.2) and 2.5 volume of ice cold ethanol. The pellet was resuspended in water, the concentration of nucleic acid determined and then mRNA was isolated using the Oligotex mRNA system (Qiagen). First strand cDNA was synthesised using an oligo-dT primer supplied with the ZAP-cDNA synthesis kit (Stratagene - cat # 200400) and the reverse transcriptase SuperscriptIll (Invitrogen). Double stranded cDNA was ligated to EcoRIIXhoI adaptors and from this a library was constructed using the ZAP-cDNA synthesis kit as described in the accompanying instruction manual (Stratagene - cat # 200400). The titer of the primary library was 2.5 x 10 5 plaque forming units (pfu)/ ml and that of the amplified library was 2.5 x 10 9 pfu/ ml. The average insert size of cDNA inserts in the library was 1.3 kilobases and the percentage of recombinants in the library was 74%. Example 2. Microalgae and Polyunsaturated Fatty Acid Contents Thereof The CSIRO Collection of Living Microalgae CSIRO established and maintained a Collection of Living Microalgae (CLM) containing over 800 strains from 140 genera representing the majority of marine and some freshwater microalgal classes (list of strains available downloadable from http://www.marine.csiro.au). Selected micro-heterotrophic strains were also maintained. This collection is the largest and most diverse microalgal culture collection in Australia. The CLM focused on isolates from Australian waters - over 80% of the strains were isolated from diverse localities and climatic zones, from tropical northern Australia to the Australian Antarctic Territory, from oceanic, inshore coastal, estuarine, intertidal and freshwater environments. Additionally, emphasis has been placed on representation of different populations of a single species, usually by more than one strain. All strains in the culture collection were unialgal and the majority were clonal. A subset of strains were axenic. Another collection is the NIES-Collection (National Institute for Environmental Studies, Environment Agency) maintained in Japan. Microalgae are known for their cosmopolitanism at the morphological species level, with very low endemicity being shown. However this morphological cosmopolitanism can hide a plethora of diversity at the intra-specific level. There have been a number of studies of genetic diversity on different microalgae using approaches such as interbreeding, isozymes, growth rates and a range of molecular techniques. The diversity identified by these studies ranges from large regional and global scales (Chinain WO 2005/103253 PCT/AU2005/000571 66 et al., 1997) to between and within populations (Gallagher, 1980; Medlin et al., 1996; Bolch et al., 1999a,b). Variation at the intra-specific level, between morphologically indistinguishable microalgae, can usually only be identified using strains isolated from the environment and cultured in the laboratory. It is essential to have identifiable and stable genotypes within culture collections. While there are recorded instances of change or loss of particular characteristics in long term culture (Coleman, 1977), in general, culturing guarantees genetic continuity and stability of a particular strain. Cryopreservation strategies could also be used to limit the potential for genetic drift. Microalgae and their use in aquaculture Because of their chemical/nutritional composition including PUFAs, microalgae are utilized in aquaculture as live feeds for various marine organisms. Such microalgae must be of an appropriate size for ingestion and readily digested. They must have rapid growth rates, be amenable to mass culture, and also be stable in culture to fluctuations in temperature, light and nutrients as may occur in hatchery systems. Strains fulfilling these attributes and used widely in aquaculture include northern hemisphere strains such as Isochrysis sp. (T.ISO) CS-177, Pavlova lutheri CS-182, Chaetoceros calcitrans CS-178, C. muelleri CS-176, Skeletonema costatum CS-181, Thalassiosira pseudonana CS-173, Tetraselmis suecica CS-187 and Nannochloropsis oculata CS-189. Australian strains used include Pavlovapinguis CS-375, Skeletonema sp. CS-252, Nannochloropsis sp. CS 246, Rhodomonas salina CS-24 and Naviculajeffreyi CS-46. Biochemical assessment of over 50 strains of microalgae used (or of potential use) in aquaculture found that cells grown to late-logarithmic growth phase typically contained 30 to 40% protein, 10 to 20% lipid and 5 to 15% carbohydrate (Brown et al., 1997). Lipid composition including PUFA content of microalgae There is considerable interest in microalgae containing a high content of the nutritionally important long-chain polyunsaturated fatty acids (LC-PUFA), in particular EPA [eicosapentaenoic acid, 20:5(o3)] and DHA [docosahexaenoic acid, 22:6(o3)] as these are essential for the health of both humans and aquacultured animals. While these PUFA are available in fish oils, microalgae are the primary producers of EPA and DHA. The lipid composition of a range of microalgae (46 strains) and particularly the proportion and content of important PUFA in the lipid of the microalgae were profiled.

C

1

-C

22 PUFA composition of microalgal strains from different algal classes varied considerably across the range of classes of phototrophic algae (Table 3, Figure 2, see also WO 2005/103253 PCT/AU2005/000571 67 Dunstan et. al., 1994, Volkman et al., 1989; Mansour et al., 1999a). Diatoms and eustigmatophytes were rich in EPA and produced small amounts of the less common PUFA, ARA [arachidonic acid, 20:4(0o6)] with negligible amounts of DHA. In addition, diatoms made unusual C 16 PUFA such as 16:4(0ol) and 16:3(o4). In contrast, dinoflagellates had high concentrations of DHA and moderate to high proportions of EPA and precursor C 1 s PUFA [18:5(0o3) and 18:4(co3) SDA, stearidonic acid]. Prymnesiophytes also contained EPA and DHA, with EPA the dominant PUFA. Cryptomonads were a rich source of the C 18 PUFA 18:3(03) (ALA a-linolenic acid) and SDA, as well as EPA and DHA. Green algae (e.g. Chlorophytes such as Dunaliella spp. and Chlorella spp.) were relatively deficient in both C20 and C22 PUFA, although some species had small amounts of EPA (up to 3%) and typically contained abundant ALA and 18:2(co6), and were also able to make 16:4(03). The biochemical or nutritional significance of uncommon C 1 6 PUFA [e.g. 16:4(03), 16:4(ol), 16:3(0o4)] and C 1 PUFA (e.g. 18:5(o)3) and STA] is unclear. However there is current interest in Cis PUFA such as SDA that are now being increasingly recognized as precursors for the beneficial EPA and DHA, unlike ALA which has only limited conversion to EPA and DHA. New strains of Australian thraustochytrids were isolated. When examined, these thraustochytrids showed great morphological diversity from single cells to clusters of cells, complex reticulate forms and motile stages. Thruastochytrids are a group of single cell organisms that produce both high oil and LC-PUFA content. They were initially thought to be primitive fungi, although more recently have been assigned to the subclass Thraustochytridae (Chromista, Heterokonta), which aligns them more closely with other heterokont algae (e.g. diatoms and brown algae). Under culture, thraustochytrids can achieve considerably higher biomass yield (>20 g/L) than other microalgae. In addition, thraustochytrids can be grown in fermenters with an organic carbon source and therefore represent a highly attractive, renewable and contaminant-free, source of omega-3 oils.

WO 2005/103253 PCT/AU2005/000571 68 TABLE 3. Distribution of selected PUFA and LC-PUFA in microalgae and other groups, and areas of application. Group Genus / Species PUFA Application Eustigmatophytes Nannochloropsis EPA Aquaculture Diatoms Chaetoceros Dinoflagellates Crypthecodinium cohnii DHA Aquaculture, health Thraustochytrids Schizochytrium supplements, infant formula Red algae Phorphyridium ARA Aquaculture, infant formula Thraustochytrids undescribed species Pharmaceutical industry Fungi Mortiella (precursor to prostaglandins) Blue green algae Spirulina GLA health supplements Abbreviations: y-linolenic acid, GLA, 18:3o6; 20:5o3, eicosapentaenoic acid, EPA, 20:5o3; docosahexaenoic acid, DHA, 22:603; arachidonic acid, ARA, 20:40o6. Representative fatty acid profiles for selected Australian thraustochytrids are shown in Table 4. Strain 0 was particularly attractive as it contained very high levels of DHA (61%). Other PUFA were present at less than 5% each. High DHA-containing thraustochytrids often also contained high proportions of 22:5(o6, docosapentaenoic acid WO 2005/103253 PCT/AU2005/000571 69 (DPA), as was observed for strains A, C and H. DPA was only a minor component in strain 0 under the culture conditions employed, making this strain particularly interesting. Strain A contained both DHA (28%) and EPA (16%) as the main LC-PUFA. Strains C and H differed from the other strains with ARA (10-13%) also being present as a major LC-PUFA. A number of other LC-PUFA were present in the thraustochytrids including DPA(3) and 22:4o6 and other components. TABLE 4. Fatty acid composition (% of total) of thraustochytrid strains. Fatty acid Percentage composition Strain A C H 0 16:0 18.0 16.4 13.5 22.1 20:40o6 ARA 4.0 10.5 13.4 0.7 20:503 EPA 15.8 7.7 5.2 4.1 22:5o6 DPA(6) 16.6 9.3 12.7 3.4 22:603 DHA 28.2 21.6 19.2 61.0 The microalgal and thraustochytrid isolates in the CLM that may be used for isolation of genes involved in the synthesis of LC-PUFA are of the genera or species as follows: Class Bacillariophyceae (Diatoms) Attheya septentrionalis, Aulacoseira sp., Chaetoceros affinis, Chaetoceros calcitrans, Chaetoceros calcitrans f. pumilum, Chaetoceros cf. mitra, Chaetoceros cf. peruvianus, Chaetoceros cf. radians, Chaetoceros didymus, Chaetoceros difficile, Chaetoceros gracilis, Chaetoceros muelleri, Chaetoceros simplex, Chaetoceros socialis, Chaetoceros sp., Chaetoceros cf. minus, Chaetoceros cf. tenuissimus, Coscinodiscus wailesii, other Coscinodiscus spp., Dactyliosolen fragilissimus, Detonula pumila, Ditylum brightwellii, Eucampia zodiacus, Extubocellulus spinifera, Lauderia annulata, Leptocylindrus danicus, Melosira monilifformis, Melosira sp., Minidiscus WO 2005/103253 PCT/AU2005/000571 70 trioculatus, Minutocellus polymorphus, Odontella aurita, Odontella mobiliensis, Odontella regia, Odontella rhombus, Odontella sp., Papiliocellulus simplex, Planktosphaerium sp., Proboscia alata, Rhizosolenia imbricata, Rhizosolenia setigera, Rhizosolenia sp., Skeletonema costatum, Skeletonema pseudocostatum, Skeletonema sp., Skeletonema tropicum, other Skeletonema spp., Stephanopyxis turris, Streptotheca sp., Streptotheca tamesis, Streptotheca spp., Striatella sp., Thalassiosira delicatula, Thalassiosira eccentrica, Thalassiosira mediterranea, Thalassiosira oceanica, Thalassiosira oestrupii, Thalassiosira profunda, Thalassiosira pseudonana, Thalassiosira rotula, Thalassiosira stellaris, other Thalassiosira spp., Achnanthes cf amoena, Amphiprora cf. alata, Amphiprora hyalina, Amphora spp., Asterionella glacialis, Asterionellopsis glacialis, Biddulphia sp., Cocconeis sp., Cylindrotheca closterium, Cylindrotheca fusiformis, Delphineis sp., Diploneis sp., Entomoneis sp., Fallacia carpentariae, Grammatophora oceanica, Haslea ostrearia, Licmophora sp., Manguinea sp., Navicula cf jeffreyi, Naviculajeffreyi, other Navicula spp., Nitzschia of bilobata, Nitzschia cf. constricta, Nitzschia cf cylindrus , Nitzschia of frustulum, Nitzschia cf paleacea, Nitzschia closterium, Nitzschia fraudulenta, Nitzschia frustulum, Nitzschia sp., Phaeodactylum tricornutum, Pleurosigma delicatulum, other Pleurosigma spp., Pseudonitzschia australis, Pseudonitzschia delicatissima, Pseudonitzschia fraudulenta, Pseudonitzschia pseudodelicatissima, Pseudonitzschia pungens, Pseudonitzschia sp., Pseudostaurosira shiloi, Thalassionema nitzschioides, or Thalassiothrix heteromorpha. Class Chrysophyceae Chrysolepidomonas cf. marina, Hibberdia spp., Ochromonas danica, Pelagococcus subviridis, Phaeoplaca spp., Synura shagnicola or other Chrysophyte spp. Class Cryptophyceae Chroomonas placoidea, Chroomonas sp., Geminigera cryophila, Hemiselmis simplex, Hemiselmis sp., Rhodomonas baltica, Rhodomonas maculata, Rhodomonas salina, Rhodomonas sp. or other Cryptomonad spp. Class Dinophyceae (Dinoflagellates) Alexandrium affine, Alexandrium catenella, Alexandrium margalefi, Alexandrium minutum, Alexandrium protogonyaulax, Alexandrium tamarense, Amphidinium carterae, Amphidinium cf britannicum, Amphidinium klebsii, Amphidinium sp., Amphidinium steinii, Amylax tricantha, Clyptothecodinium cohnii, Ensicul'fera WO 2005/103253 PCT/AU2005/000571 71 sp., Fragilidium spp., Gambierdiscus toxicus, Gymnodinium catenatum, Gymnodinium galathaneum, Gymnodinium galatheanum, Gymnodinium nolleri, Gymnodinium sanguineum, or other Gymnodinium spp., Gyrodinium pulchellum, or other Gyrodinium spp., Heterocapsa niei, Heterocapsa rotundata, Katodinium cf. rotundatum, Kryptoperidinium foliaceum, Peridinium balticum, Prorocentrum gracile, Prorocentrum mexicanum, Prorocentrum micans, Protoceratium reticulatum, Pyrodinium bahamense, Scrippsiella cf. precaria, or other Scrippsiella spp. Symbiodinium microadriaticum, or Woloszynskia sp. Class Euglenophyceae Euglena gracilis. Class Prasinophyceae Pycnococcus sp., Mantoniella squamata, Micromonas pusilla, Nephroselmis minuta, Nephroselmis pyriformes, Nephroselmis rotunda, Nephroselmis spp., or other Prasinophyte spp., Pseudoscourfieldia marina, Pycnococcus provasolii, Pyramimonas cordata, Pyramimonas gelidicola, Pyramimonas grossii, Pyramimonas oltmansii, Pyramimonas propulsa, other Pyramimonas spp., Tetraselmis antarctica, Tetraselmis chuii, Tetraselmis sp., Tetraselmis suecica, or other Tetraselmis spp. Class Prymnesiophyceae Chrysochromulina acantha, Chrysochromulina apheles, Chrysochromulina brevifilum, Chrysochromulina camella, Chrysochromulina hirta, Chrysochromulina kappa, Chrysochromulina minor, Chrysochromulina- pienaar, Chrysochromulina simplex, Chrysochromulina sp., Chrysochromulina spinifera, Chrysochromulina strobilus, and other Chrysophyte spp., Chrysotila lamellosa, Cricosphaera carterae, Cystallolithus hyalinus, Diacronema v/kianum, Dicrateria inornata, Dicrateria sp., Emiliania huxleyi, Gephyrocapsa oceanica, Imantonia rotunda, and other Isochrysis spp., Ochrosphaera neapolitana, Pavlova cf. pinguis, Pavlova gyrans, Pavlova lutheri, Pavlova pinguis, Pavlova salina, Pavlova sp., Phaeocystis of. pouchetii, Phaeocystis globosa, Phaeocystis pouchetii, other Phaeocystis spp., Pleurochrysis aff carterae, Prymnesium parvum, Prymnesium patelh'ferum, other Prymnesium spp., or Pseudoisochtysis paradoxa.

WO 2005/103253 PCT/AU2005/000571 72 Class Raphidophyceae Chattonella antiqua, other Chattonella spp., Fibrocapsa japonica, other Fibrocapsa spp., Heterosigma akashiwo, Heterosigma carterae, or other Heterosigma spp. Class Thraustochytridae Schizochytrium spp., Thraustochytrium aureum, Thraustochytrium roseum, or other Thraustochytrium spp. Class Eustigmatophytae as a source of genes for EPA production: Eustigmatos vischeri, Monodus subterraneus, Nannochloropsis oculata, Nannochloropsis salina, Vischeria helvetica, Vischeria punctata, Chloridella neglecta, Chloridella simplex, Chlorobotrys regularis, Ellipsoidon parvum, Ellipsoidon solitare, Eustigmatos magnus, Eustigmatos polyphem, Goniochloris sculpta, Monodus subterraneus, Monodus unipapilla, Nannochloropsis gaditana, Nannochloropsis granulata, Nannochloropsis limnetica, Pseudocharaciopsis, ovalis, Pseudocharaciopsis texensis, Pseudostaurastrum limneticum, or Vischeria stellata Example 3. Isolation of Zebrafish A5/6 Desaturase and Functional Characterization in Yeast As well as microalgae, some other organisms have the capacity to synthesise LC PUFA from precursors such as a-linolenic acid (18:3, ALA) (see Figure 1) and some of the genes responsible for such synthesis have been isolated (see Sayanova and Napier, 2004). The genes involved in omega-3 C 20 +PUFA biosynthesis have been cloned from various organisms including algae, fungi, mosses, plants, nematodes and mammals. Based on the current understanding of genes involved in the synthesis of omega-3 C 20 +PUFA, synthesis of EPA in plants would require the transfer of genes encoding at least two desaturases and one PUFA elongase. The synthesis of DHA from EPA in plants would require the additional transfer of a further desaturase and a further elongase (Sayanova and Napier, 2004). These enzymes are: for the synthesis of EPA, the sequential activities of a A6 desaturase, A6 elongase and a A5 desaturase is required. Based on an alternative pathway operative in some algae, EPA may also be synthesised by the sequential activities of a A9 elongase, a A8 desaturase and a A5 desaturase (Wallis and Browse, 1999; Qi et al., 2002). For the further conversion of EPA to DHA in plants, a further transfer of a A5 elongase and A4 desaturase will be required (Sayanova and Napier, 2004).

WO 2005/103253 PCT/AU2005/000571 73 Hastings et al. (2001) isolated a gene encoding a A5/A6 bifunctional desaturase from zebra fish (Danio rerio) and showed that, when expressed in yeast, the desaturase was able to catalyse the synthesis of both A6 (GLA and SDA) and A5 (20:4 and EPA) fatty acids. The desaturase was therefore able to act on both o6 and o3 substrates. Isolation of the zebrafish A5/A6 desaturase RNA was extracted using the RNAeasy system according to the manufacturers instructions (Qiagen) from freshly dissected zebrafish livers. Based on the published sequence (Hastings et al. 2001), primers, sense, 5' CCCAAGCTTACTATGGGTGGCGGAGGACAGC-3' (SEQ ID NO:39) and antisense 5'-CCGCTGGAGTTATTTGTTGAGATACGC-3' (SEQ ID NO:40) at the 5' and 3' extremities of the zebrafish A5/6 ORF were designed and used in a one-step reverse transcription-PCR (RT-PCR, Promega) with the extracted RNA and using buffer conditions as recommended by the manufacturer. A single amplicon of size 1335bp was obtained, ligated into pGEM-T easy (Promega) and the sequence confirmed as identical to that published. A fragment containing the entire coding region (SEQ ID NO:38) was excised and ligated into the yeast shuttle vector pYES2 (Invitrogen). The vector pYES2 carried the URA3 gene, which allowed selection for yeast transformants based on uracil prototrophy. The inserted coding region was under the control of the inducible GAL] promoter and polyadenylation signal of pYES2. The resultant plasmid was designated pYES2-zfA5/6, for introduction and expression in yeast (Saccharomyces cerevisiae). Expression of zebrafish A5/A6 desaturase in yeast The gene construct pYES2-zfA5/6 was introduced into yeast strain S288. Yeast was a good host for analysing heterologous potential LC-PUFA biosynthesis genes including desaturases and elongases for several reasons. It was easily transformed. It synthesised no LC-PUFA of its own and therefore any new PUFA made was easily detectable without any background problems. Furthermore, yeast cells readily incorporated fatty acids from growth media into cellular lipids, thereby allowing the presentation of appropriate precursors to transformed cells containing genes encoding new enzymes, allowing for confirmation of their enzymatic activities. Biochemical analyses Yeast cells transformed with pYES2-zfA5/6 were grown in YMM medium and induced by the addition of galactose. The fatty acids 18:303 (ALA, 0.5 mIM) or 20:4 3 WO 2005/103253 PCT/AU2005/000571 74 (ETA, 0.5 mM) were added to the medium as described above. After 48 hours incubation, the cells were harvested and fatty acid analysis carried out by capillary gas liquid chromatography (GC) as described in Example 1. The analysis showed that 18:40)3 (1.9% of total fatty acid) was formed from 18:33 and 20:503 (0.24% of fatty acids) from 20:43, demonstrating A6 desaturase activity and A5 desaturase activity, respectively. These data are summarized in Table 5 and confirm the results of Hastings et al (2001). Example 4. Isolation of C. elegans Elongase and Functional Characterization in Yeast Cloning of C. elegans elongase gene Beaudoin and colleagues isolated a gene encoding an ELO-type fatty acid elongase from the nematode Caenorhabditis elegans (Beaudoin et al., 2000) and this gene was isolated as follows. Oligonucleotide primers having the sequences 5' GCGGGTACCATGGCTCAGCATCCGCTC-3' (SEQ ID NO:4 1) (sense orientation) and 5'- GCGGGATCCTTAGTTGTTCTTCTTCTT-3' (SEQ ID NO:42) (antisense orientation) were designed and synthesized, based on the 5' and 3' ends of the elongase coding region. These primers were used in a PCR reaction to amplify the 867 basepair coding region from a C. elegans N2 mixed-stage gene library, using an annealing temperature of 58 0 C and an extension time of 1 minute. The PCR amplification was carried out for 30 cycles. The amplification product was inserted into the vector pGEMTM T-easy (Promega) and the nucleotide sequence confirmed (SEQ ID NO:37). An EcoRI/BamHI fragment including the entire coding region was excised and inserted into the EcoRI/BgIl sites of pSEC-TRP (Stratagene), generating pSEC-Ceelo, for introduction and expression in yeast. pSEC-TRP contains the TRP1 gene, which allowed for the selection of transformants in yeast by tryptophan prototrophy, and the GAL1 promoter for expression of the chimeric gene in an inducible fashion in the presence of galactose in the growth medium.

WO 2005/103253 PCT/AU2005/000571 75 TABLE 5. Enzymatic activities in yeast and Arabidopsis Synthesised Observed Clone Precursor PUFA PUFA (of total FA) activity pYES2-zfA5/6 18:3o)3 18:4o)3 1.9 A6 desaturase pYES2-zfA5/6 20:403 20:5o3 0.24 A5 desaturase pYES2zfA5/6, pSEC Ceelo 18:303 18:403 0.82 A6 desaturase 20:33 0.20 A9 elongase 20:403 0.02 A6 elongase NOT pYES2-psA8 18:33 18:4o3 - A6 desaturase pYES2-psA8 20:303 20:403 0.12 A8 desaturase pYES2-psELOl 18:2o6 20:2co6 pYES2-psELO1 18:303 pYES2-psELO1 20:303 22:33 pYES2-psELOl 20:4(o3 22:4o3 pYES2-psELOl 20:503 22:503 0.82 A5 elongase pYES-psELO2 18:20o6 20:2o)6 0.12 A9 elongase pYES-psELO2 18:303 20:3o3 0.20 A9 elongase pYES-psELO2 20:30o3 22:3o3 pYES-psELO2 20:403 22:4o)3 pYES-psELO2 20:503 22:503 A5/6 desaturase, Arabidopsis + zfA5/6 - 18:3co6 0.32 A5/6/9 elongase & Ceelo - 18:403 1.1 (plant #1) - 20:4o6 1.1 - 20:5o3 2.1 - 20:30o6 1.1 - 20:403 0.40 - 20:20o6 3.2 - 20:303 TR - 22:40o6 0.06 - 22:503 0.13 22:30o6 0.03 TR, trace, not accurately determined.

WO 2005/103253 PCT/AU2005/000571 76 Functional characterization of C. elegans elongase gene in yeast Yeast strain S288 was transformed, using the method described in Example 1, with both vectors pYES2-zfA5/6 and pSEC-Ceelo simultaneously and double transformants were selected on YMM medium that lacked tryptophan and uracil. The transformants grew well on both minimal and enriched media, in contrast to transformants of strain S288 carrying pSEC-Ceelo alone, in the absence of pYES2 zfA5/6, which grew quite poorly. Double transformants were grown in YMM medium and induced by the addition of galactose. The fatty acid 18:3o3 (ALA, 0.5 mM?) was added to the medium and, after 48 hours incubation, cells were harvested and fatty acid analysis carried out by capillary gas-liquid chromatography (GC) as described .in Example 1. The analysis showed that 18:403 (0.82% of total fatty acid) and 20:303 (0.20%) were formed from 18:303, and 20:403 (0.02% of fatty acids) from either of those, demonstrating the concerted action of an elongase activity in addition to the A6 desaturase activity and A5 desaturase activity of the zebrafish desaturase (Table 5). The concerted action of a bifunctional A5/6 desaturase gene and an elongase gene has not been reported previously. In particular, the use of a bifunctional enzyme, if showing the same activities in plant cells, would reduce the number of genes that would need to be introduced and expressed. This also has not been reported previously. Example 5. Coordinate Expression of Fatty Acid Desaturase and Elongase in Plants Genetic construct for co-expression of the zebrafish A6/A5 desaturase and C. elegant elongase in plant cells Beaudoin and colleagues (2000) showed that the C. elegans A6 elongase protein, when expressed in yeast, could elongate the C18 A6 desaturated fatty acids GLA and SDA, i.e. that it had A6 elongase activity on C18 substrates. They also showed that the protein did not have A5 elongase activity on a C20 substrate in yeast. We tested, therefore, whether this elongase would be able to elongate the A6 desaturated fatty acids GLA and SDA in Arabidopsis seed. Arabidopsis thaliana seed have been shown to contain both omega-6 (18:2, LA) and omega-3 (18:3, ALA) fatty acids (Singh et al, 2001). The presence of 18:3 in particular makes Arabidopsis seed an excellent system to study the expression of genes that could lead to the synthesis of omega-3 C 20 +PUFA like EPA and DHA. The test for elongase activity in Arabidopsis required the coordinate expression of a A6 desaturase in the seed to first form GLA or SDA. We chose to express the elongase gene in conjunction with the zebrafish desaturase gene described above. There were no WO 2005/103253 PCT/AU2005/000571 77 previous reports of the expression of the zebra fish A6/A5 desaturase and C .elegans elongase genes in plant cells, either individually or together. Seed-specific co-expression of the zebra fish A6/A5 desaturase and C. elegans elongase genes was achieved by placing the genes independently under the control of a 309 napin promoter fragment, designated Fpl (Stalberg et al., 1993). For plant transformation, the genes were inserted into the binary vector pWvec8 that comprised an enhanced hygromycin resistance gene as selectable marker (Wang et al., 1997). To achieve this, the C. elegans elongase coding region from Example 4 was inserted as a blunt-end fragment between the Fpl and Nos 3' polyadenylation/terminator fragment in the binary vector pWvec8, forming pCeloPWvec8. The zebrafish A5/A6 desaturase coding region from Example 3 was initially inserted as a blunt end fragment between the Fp 1 and Nos 3' terminator sequences and this expression cassette assembled between the HindIII and ApaI cloning sites of the pBluescript cloning vector (Stratagene). Subsequently, the entire vector containing the desaturase expression cassette was inserted into the HindIII site of pCeloPWvec8, forming pZebdesatCeloPWvec8. The construct, shown schematically in Figure 3, was introduced into Agrobacterium strain AGLI (Valvekens et al., 1988) by electroporation prior to transformation into Arabidopsis thaliana, ecotype Columbia. The construct was also designated the "DO" construct, and plants obtained by transformation with this construct were indicated by the-prefix "DO". Plant transformation and analysis Plant transformation was carried out using the floral dipping method (Clough and Bent, 1998). Seeds (TI seeds) from the treated plants (TO plants) were plated out on hygromycin (20 mg/1) selective media and transformed plants selected and transferred to soil to establish Ti plants. One hygromycin resistant plant was recovered from a first screen and established in soil. The transformation experiment was repeated and 24 further confirmed T1 transgenic plants were recovered and established in soil. Most of these T1 plants were expected to be heterozygous for the introduced transgenes. T2 seed from the 25 transgenic plants were collected at maturity and analysed for fatty acid composition. As summarised in Table 6, seed of untransformed Arabidopsis (Columbia ecotype) contained significant amounts of both the o6 and 3, C18 fatty acid precursors LA and ALA but did not contain any A6-desaturated C18 (18:3o6 or 18:43), o6-desaturated C20 PUFA or o3-desaturated C20 PUFA. In contrast, fatty acids of the seed oil of the transformed plants comprising the zebra fish A5/A6 desaturase and C. elegans elongase gene constructs contained 18:3o6, 18:43 and a whole series of o6- WO 2005/103253 PCT/AU2005/000571 78 S 0 C\J . co0 0 00 000 0 0 0 00 0L0 V- V- V- V- T rn CD C00 0 00 0)C0 0 o C\j C.)) wo H C14 C.)N Co 00000100CD00C000C)C)C)000)CD00d0 C000d00 w 6 0- 0< 0 ~t C) V- ) C )C D0 C C) o o 6 6 6 o o 6 6 6 6 6 C.)l co 0C )C lCDC )C )aC 0C )C )C )C w C) o C)6 C rnn oY D00C C )C )C <5 ) 00 C)o0 CD 0 CNJ ) C) L 0 :1 r-:c ,fl q q '1: C Y 4 q 1:( q 17: )U 0 CDOOO0OV OOOOOOOC) C) 0 CD 00 '66666666666666.6 Id I.- qC )C C ) )0 C )C )C a C )C >C DC N C1 Y O m' qMc 1 qm mc C (E 00O O OO 5 5 55 5~ oQ QO OO oo oo4oooo WO 2005/103253 PCT/AU2005/000571 79 and o)3-C20 PUFA. These resulted from the sequential action of the desaturase and elongase enzymes on the respective C18 precursors. Most importantly and unexpectedly, the transgenic seed contained both 20:5o3 (EPA), reaching at least 2.3% of the total fatty acid in the seedoil, and 22:503 (DPA), reaching at least 0.2% of this omega-3 LC-PUFA in the fatty acid of the seedoil. The total C20 fatty acids produced in the transgenic seed oil reached at least 9.0%. The total o3 fatty acids produced that were a product of A6 desaturation (i.e. downstream of 18:303 (ALA), calculated as the sum of the percentages for 18:43 (SDA), 20:403 (ETA), 20:50o3 (EPA) and 22:5w3 (DPA)) reached at least 4.2%. These levels represent a conversion efficiency of ALA, which is present in seed oil of the wild-type Arabidopsis plants used for the transformation at a level of about 13 15%, to 03 products through a A6 desaturation step of at least 28%. Stated otherwise, the ratio of ALA products to ALA (products:ALA) in the seed oil was at least 1:3.6. Of significance here, Arabidopsis has a relatively low amount of ALA in its seed oil compared to some commercial oilseed crops. The T2 lines described above included lines that were homozygous for the transgenes as well as heterozygotes. To distinguish homozygotes and heterozygotes for lines expressing the transgenes at the highest levels, T2 plants were established from the T2 seed for the 5 lines containing the highest EPA levels, using selection on MS medium containing hygromycin (15mg/L) to determine the presence of the transgenes. For example, the T2 seed was used from the Ti plant designated DOll, containing 2.3% EPA and showing a 3:1 segregation ratio of resistant to susceptible progeny on the hygromycin medium, indicating that DO 11 contained the transgenes at a single genetic locus. Homozygous lines were identified. For example, T2 progeny plant DO1 1-5 was homozygous as shown by the uniformly hygromycin resistance in its T3 progeny. Other T2 plants were heterozygous for the hygromycin marker. The fatty acid profiles of T3 seed lots from DO11-5 and other T2 progeny of DOl1 were analysed and the data are presented in Table 7. As expected, the EPA contents reflected segregation of the DO construct. The levels of EPA in the fatty acid of the seedoil obtained from the T3 lines were in three groups: negligible (nulls for the DO construct), in the range 1.6-2.3% (heterozygotes for the DO construct) and reaching at least 3.1% (homozygotes for the DO construct). The levels obtained were higher in homozygotes than heterozygotes, indicating a gene dosage effect. T3 seed from the DO 11-5 plant synthesized a total of 9.6% new 3 and o06 PUFAs, including 3.2% EPA, 1.6% ARA, 0.1% DPA, 0.6% SDA and 1.8% GLA (Table 7). This level of EPA synthesis in seed was four fold higher than the 0.8% level previously achieved in linseed WO 2005/103253 PCT/AU2005/000571 80 0 c ar 0 00i :. C) co) m6 d6o NNOO0 No0 a0 0 7 7c *'7.J M N00wa000o 0 MN N 0 0 CO OOO qDq0q0 oc.q C o0 0 CD 0- N CD mNc0O 0 O 0000 0 m Noo ooo N NO O N CN 0 0' qq(0 qq0 R N7 04~00-~)~ 0 0 Nf-DCO -00000N N N 0 ZN 0 a- r-OO c c W L 0 ~0 O0 o~ 00o o i, NN0oo 0 r,: 66 6 6c6 6 iC No do d -co d -odood 0 T 7 C! CRC oR q q m ~ ORRR No N-00q0 M)'N M N 0 o o o7 OON CDo o 6 L6 N C~) o0 0 oI-~ N 00 '-0 00 6). 0 000CO Occu doCa0d N 0O 6 oo - ~ ~ ~ N rq q~q q Z:qqRqqq 1:qqqqq C 0 )00000r)- 0 DC N.~M~ -t-ocnnrO r-o oo,-, m~o rccis NoN - ~ 66 6 6 6 6 66 o N0 0-66r o-- o - 0 0 N N Nc .' ' N 0 NN, A LT0 WO 2005/103253 PCT/AU2005/000571 81 (Abbadi et al., 2004). Considering also that the level of ALA precursor for EPA synthesis in Arabidopsis seed was less than a third of that present in linseed, it appeared that the LC-PUFA pathway as described above which included a desaturase that was capable of using an acyl-CoA substrate, was operating with significantly greater efficiency than the acyl-PC dependent desaturase pathway expressed in linseed. The relative efficiencies of the individual enzymatic steps encoded by the EPA construct can be assessed by examining the percentage conversion of substrate fatty acid to product fatty acids (including subsequent derivatives) in DO 11-5. The zebra-fish A5/A6 desaturase exhibited strong A5 desaturation, with 89% of 20:43 being converted to EPA and DPA, and 45% of 20:3co6 being converted to ARA, consistent with the previously reported preference of this enzyme for 03 PUFA over 0o6 PUFA substrates (Hastings et al., 2001). In comparison, A6-desaturation occurred at significantly lower levels, with 32% of ALA and 14% LA being converted to A6-desaturated PUFA. Given that previous studies in yeast showed this enzyme to actually have higher A6-desaturase activity than A5-desaturase activity, the lower A6-desaturation levels achieved in Arabidopsis seeds could be reflect a limited availability of ALA and LA substrates in the acyl-CoA pool (Singh et al., in press). The A6-elongase operated highly efficiently, with 86% of GLA and 67% of SDA being elongated, suggesting that this enzyme may have a slight preference for elongation of co6-PUFA substrate. The germination ability of the T2 (segregating) and T3 seed (homozygous population) was assessed on MS medium and on soil. Seed from the EPA and DPA containing lines DO11 and DO11-5 showed the same timing and frequency of germination as wild-type seed, and the T2 and T3 plants did not have any apparent abnormal morphological features. Plant growth rates in vitro or in soil and the quantities of seed obtained from the plants were also unaffected. Including the germination of the T1 seed from which plant DOl1 was obtained, the normal germination of seed of the DO 11 line was thus observed over three generations. In addition, normal germination rates and timing were also observed for the other EPA and DPA containing seed. This feature was both important and not predictable, as higher plants do not naturally produce EPA or DPA and their seed therefore has never previously contained these LC-PUFA. Germination requires the catabolism of stored seed oils and use for growth and as an energy supply. The observed normal germination rates showed that plant seed were able to carry out these processes using EPA and DPA, and that these compounds were not toxic. It has been reported that a A4 desaturase encoded by a gene isolated from Thraustochytrium spp and expressed in Brassica juncea leaves was able to convert WO 2005/103253 PCT/AU2005/000571 82 exogenously supplied DPA to DHA (Qiu et al., 2001).. DPA produced in the plant seed described herein can serve as a precursor for DHA production. This conversion of DPA to DHA may be achieved in plant cells by the introduction of a A4 desaturase gene into the DPA producing plant cells (Example 11). Discussion The presence of 22:503 in the Arabidopsis seed oil implied that the C. elegans elongase gene not only had A6 elongase activity, but also A5 elongase activity in plant cells. This result was most surprising given that the gene had been demonstrated to lack A5 elongase activity in yeast. Furthermore, this demonstrated that only two genes could be used for the synthesis of DPA from ALA in plant cells. The synthesis of DPA in a higher plant has not previously been reported. Furthermore, the conversion efficiency of ALA to its o3 products in seed, including EPA, DPA or both, of at least 28% was striking. Synthesis of LC-PUFA such as EPA and DHA in cells such as plant cells by the A6 desaturation pathway required the sequential action of PUFA desaturases and elongases. The required desaturases in one pathway had A6, A5 and A4 desaturating activity, in that order, and the required PUFA elongases had elongating activity on A6 and A5 substrates. This conventional pathway operates in algae, mosses, fungi, diatoms, nematodes and some freshwater fish (Sayanova and Napier, 2004). The PUFA desaturases from algae, fungi, mosses and worms are selective for desaturation of fatty acids esterified to the sn-2 position of phosphatidylcholine (PC) while the PUFA elongases act on fatty acids in the form of acyl-CoA substrates represented in the acyl CoA pool of tissues. In contrast, vertebrate A6 desaturases have been shown to be able. to desaturate acyl-CoA substrates (Domergue et al., 2003a). Attempts to reconstitute LC-PUFA pathways in plant cells and other cells have to take into account the different sites of action and substrate requirements of the desaturases and elongase enzymes. For example, PUFA elongases are membrane bound, and perhaps even integral membrane proteins, which use acyl-CoAs which are present as a distinct pool in the endoplasmic reticulum (ER). This acyl-CoA pool is physiologically separated from the PC component of the ER, hence for a PUFA fatty acid to be sequentially desaturated and elongated it has to be transferred between PC and acyl-CoA pools in the ER. Therefore, earlier reported attempts to constitute LC-PUFA biosynthesis in yeast using desaturases and elongase from lower and higher plants, fungi and worms, have been inefficient, at best. In addition, the constituted pathways have led to the synthesis of only C20 PUFA such as ARA and EPA. There is no previous report of the WO 2005/103253 PCT/AU2005/000571 83 synthesis of C22 PUFA such as DPA and DHA in yeast (Beaudoin et al., 2000, Domergue et al., 2003a). The strategy described above of using a vertebrate desaturase, in this example a A5/A6 desaturase from zebra fish, with a A6 PUFA elongase from C. elegans had the advantage that both the desaturase and the elongase have activity on acyl-CoA substrates in the acyl-CoA pool. This may explain why this strategy was more efficient in the synthesis of LC-PUFA. Furthermore, using a bifunctional desaturase displaying dual A5/A6 desaturase activities allowed the synthesis of EPA by the action of only 2 genes instead of the 3 genes used by other researchers (Beaudoin et al., 2000, Domergue et al., 2003a). The use of a bifunctional A5/A6 elongase in plant cells also allowed the formation of DPA from ALA by the insertion of only three genes (one elongase and two desaturases) or, as exemplified, of only two genes (bifunctional elongase and bifunctional desaturase). Both of these aspects were surprising and unexpected. Biochemical evidence suggests that the fatty acid elongation consists of 4 steps: condensation, reduction, dehydration and a second reduction. Two groups of condensing enzymes have been identified so far. The first are involved in the synthesis of saturated and monosaturated fatty acids (C18-22). These are the FAE-like enzymes and do not appear to have a role in LC-PUFA biosynthesis. The other class of elongases identified belong to the ELO family of elongases named after the ELO gene family whose activities are required for the synthesis of the very long-chain fatty acids of sphingolipids in yeast. Apparent paralogs of the ELO-type elongases isolated from LC-PUFA synthesizing organisms like algae, mosses, fungi and nematodes have been shown to be involved in the elongation and synthesis of LC-PUFA. It has been shown that only the expression of the condensing component of the elongase is required for the elongation of the respective acyl chain. Thus the introduced condensing component of the elongase is able to successfully recruit the reduction and dehydration activities from the transgenic host to carry out successful acyl elongations. Thus far, successful elongations of C16 and C18 PUFA have been demonstrated in yeast by the heterologous expression of ELO type elongases. In this regard, the C. elegans elongase used as described above was unable to elongate C20 PUFA when expressed in yeast (Beaudoin et al, 2000). Our demonstration that the C. elegans elongase, when expressed in plants, was able to elongate the C20:5 fatty acid EPA as evidenced by the production of DPA in Arabidopsis seed was a novel and unexpected result. One explanation as to why the C. elegans elongase was able to elongate C20 PUFA in plants, but not in yeast, might reside in its ability to interact successfully with the other components of the elongation machinery of plants to bind and act on C20 substrates.

WO 2005/103253 PCT/AU2005/000571 84 This example showed that an ELO-type elongase from a non-verteberate organism was able to elongate C20 PUFA in plant cells. Leonard et al. (2002) reported that an ELO-type elongase gene isolated from humans, when expressed in yeast, was able to elongate EPA to DPA but in a non-selective fashion. Example 6. Isolation of a A8 Desaturase Gene from P. salina and Functional Characterization in Yeast Microalgae are the only organisms which have been reported to contain A8 desaturases, aside from the A8 sphingolipid desaturases in higher plants that are not involved in LC-PUFA biosynthesis. A gene encoding a A8 desaturase has been isolated from Euglena gracilis (Wallis and Browse, 1999). The existence of a A8 desaturase in Isochrysis galbana may be presumed because it contains a A9 elongase (Qi et al., 2002), the product of which, 20:3n-3, is the precursor for a A8 desaturase (see Figure 1). The fatty acid profiles of microalgae alone, however, do not provide sufficient basis for identifying which microalgae will contain A8 desaturase genes since multiple pathways may operate to produce the LC-PUFA. Isolation of a A8 desaturase gene fragment An alignment of A6 desaturase amino acid sequences with those from the following Genbank accession numbers, AF465283, AF007561, AAC15586 identified the consensus amino acid sequence blocks DHPGGS (SEQ ID NO:43), WWKDKHN (SEQ ID NO:44) and QIEHHLF (SEQ ID NO:45) corresponding to amino acid positions 204 210 and 394-400, respectively, of AF465283. DHPGSS corresponded to the "cytochrome b5 domain" block that had been identified previously (Mitchell and Martin, 1995). WWKDKHN was a consensus block that had not previously been identified or used to design degenerate primers for the isolation of desaturase genes. The QIEHHLF block, or variants thereof, corresponded to a required histidine-containing motif that was conserved in desaturases. It had been identified and used before as the "third His box" to design degenerate oligonucleotides for desaturase gene isolation (Michaelson et al., 1998). This combination of blocks had not been used previously to isolate desaturase genes. Based on the second and third conserved amino acid blocks, the degenerate primers 5'-TGGTGGAARCAYAARCAYAAY-3' (SEQ ID NO:46) and 5' GCGAGGGATCCAAGGRAANARRTGRTGYTC-3' (SEQ ID NO:47) were synthesised. Genomic DNA from P. salina was isolated using the DNAeasy system (Qiagen). PCR amplifications were carried out in reaction volumes of 20pL using WO 2005/103253 PCT/AU2005/000571 85 20pmol of each primer, 200ng of P. salina genomic DNA and Hotstar Taq DNA polymerase (Qiagen) with buffer and nucleotide components as specified. The cycling conditions were: 1 cycle of 95 0 C for 15 minutes; 5 cycles of 95'C 1min; 38'C, 1min; 72 0 C, 1min; followed by 35 cycles of 95 0 C, 35 sec; 52'C, 30 sec; 72 0 C, 1min; and finishing with 1 cycle of 72*C, 10min. A 515 basepair amplicon was generated, ligated into pGEM-T easy (Promega), sequenced and used as a probe to screen a P. salina cDNA library. Isolation of a cDNA encoding a A8 desaturase from P. salina A P. salina cDNA library in X-bacteriophage was constructed using the Zap cDNA Synthesis Kit (Stratagene) (see Example 1). The library was plated out at a concentration of -50,000 plaques per plate and lifts taken with Hybond N+ membrane and treated using standard methods (Ausubel et al., 1988, supra). The 515bp desaturase fragment, generated by PCR, was radio-labelled with 32 P-dCTP and used to probe the lifts under the following high stringency conditions: Overnight hybridisation at 65 0 C in 6X SSC with shaking, a 5 minute wash with 2x SSC/0.1% SDS followed by two .10 minute washes with 0.2x SSC/0.1% SDS. Fifteen primary library plates (150mm) were screened for hybridization to the labeled 515bp fragment. Forty strongly hybridizing plaques were identified and ten of these were carried through to a secondary screen. Plasmids from five secondary plaques hybridizing to the 515bp probe were excised with ExAssist Helper phage according to the suppliers protocol (Stratagene). The nucleotide sequences of the inserts were obtained using the ABI Prism Big Dye Terminator kit (PE Applied Biosystems). The nucleotide sequences were identical where they overlapped, indicating that all five inserts were from the same gene. One of the five inserts was shown to contain the entire coding region, shown below to be from a A8 desaturase gene. This sequence is provided as SEQ ID NO:6. The full-length amino acid sequence (SEQ ID NO: 1) revealed that the isolated cDNA encoded a putative A6 or A8 desaturase, based on BLAST analysis. These two types of desaturases are very similar at the amino acid level and it was therefore not possible to predict on sequence alone which activity was encoded. The maximum degree of identity between the P. salina desaturase and other desaturases (BLASTX) was 27 30%, while analysis using the GAP program which allows the insertions of "gaps" in the alignment showed that the maximum overall amino acid identity over the entire coding regions of the P. salina desaturase and AAD45877 from Euglena gracilis was 45%. A WO 2005/103253 PCT/AU2005/000571 86 Pileup diagram of other sequences similar to the Pavlova salina desaturase is provided in Figure 4. The entire coding region of this clone, contained within an EcoRI/XhoI fragment, was inserted into pYES2 (Invitrogen), generating pYES2-psA8, for introduction and functional characterisation in yeast. Cells of yeast strain S288 were transformed with pYES2-psA8 as described in Example 1, and transformants were selected on medium without uracil. The yeast cells containing pYES2-psA8 were grown in culture and then induced by galactose. After the addition of 18:303 or 20:303 (0.5 mM?) to the culture medium and 48 hours of further culturing at 30'C, the fatty acids in cellular lipids were analysed as described in Example 1. When 18:303 (A9, 12, 15) was added to the medium, no 18:4o3 (A6, 9, 12, 15) was detected. However, when 20:303 (All ,14,17) was added to the medium, the presence of 20:403 (A8,11,14,17) in the cellular lipid of the yeast transformants was detected (0.12%). It was concluded the transgene encoded a polypeptide having A8 but not A6 desaturase activity in yeast cells. Isolation of a gene encoding a A8 fatty acid desaturase that does not also have A6 desaturase activity has not been reported previously. The only previously reported gene encoding a A8 desaturase that was isolated (from Euglena gracilis) was able to catalyse the desaturation of both 18:303 and 20:303 (Wallis and Browse, 1999). Moreover, expression of a gene encoding a A8 desaturase has not previously been reported in higher plants. As shown in Figure 1, expression of a A8 desaturase in concert with a A9 elongase (for example the gene encoding ELO2 - see below) and a A5 desaturase (for example, the zebrafish A5/A6 gene or an equivalent gene from P. salina or other microalgae) would cause the synthesis of EPA in plants. Aside from providing an alternative route for the production of EPA in cells, the strategy of using a A9 elongase in combination with the A8 desaturase may provide an advantage in that the elongation, which occurs on fatty acids coupled to CoA, precedes the desaturation, which occurs on fatty acids coupled to PC, thereby ensuring the availability of the newly elongated C20 PUFA on PC for subsequent desaturations by A8 and A5 desaturases, leading possibly to a more efficient synthesis of EPA. That is, the order of reactions - an elongation followed by two desaturations - will reduce the number of substrate linking switches that need to occur. The increased specificity provided by the P. salina A8 desaturase is a further advantage.

WO 2005/103253 PCT/AU2005/000571 87 Example 7. Isolation of P.salina ELO1 and ELO2 Fatty Acid Elongases ELO-type PUFA elongases from organisms such as nematodes, fungi and mosses have been identified on the basis of EST or genome sequencing strategies. A gene encoding a A9 elongase with activity on 18:303 (ALA) was isolated from Jsochrysis galbana using a PCR approach with degenerate primers, and shown to have activity in yeast cells that were supplied with exogenous 18:2(o6 (LA) or 18:3o3 (ALA), forming C20 fatty acids 20:2o6 and 20:303 respectively. The coding region of the gene IgASE1 encoded a protein of 263 amino acids with a predicted molecular weight of about 30kDa and with limited homology (up to 27% identity) to other elongating proteins. Isolation of elongase gene fragments from P. salina Based on multiple amino acid sequence alignments for fatty acid elongases the consensus amino acid blocks FLHXYH (SEQ ID NO:48) and MYXYYF (SEQ ID NO:49) were identified and the corresponding degenerate primers 5' CAGGATCCTTYYTNCATNNNTAYCA-3' (SEQ ID NO:50) (sense) and 5' GATCTAGARAARTARTANNNRTACAT-3' (SEQ ID NO:51) (antisense) were synthesised. Primers designed to the motif FLHXYH or their use in combination with the MYXYYF primer have not previously been described. These primers were used in PCR amplification reactions in reaction volumes of 20jL with 20pmol of each primer, 200ng of P. salina genomic DNA and Hotstar Taq DNA polymerase (Qiagen) with buffer and nucleotide components as specified by the supplier. The reactions were cycled as follows: 1 cycle of 95 0 C for 15 minutes, 5 cycles of 95C, 1min, 38 0 C, 1min, 72 0 C, 1min, 35 cycles of 95 0 C, 35 sec, 52*C, 30 sec, 72 0 C, 1min, 1 cycle of 72 0 C, 10min. Fragments of approximately 150bp were generated and ligated into pGEM-Teasy for sequence analysis. Of the 35 clones isolated, two clones had nucleotide or amino acid sequence with similarity to known elongases. These were designated Elol and Elo2. Both gene fragments were radio-labelled with 3 2 P-dCTP and used to probe the P. salina cDNA library under the following high stringency conditions: overnight hybridisation at 65C in 6X SSC with shaking, 5 minute wash with 2x SSC/ 0.1% SDS followed by two 10 minute washes with 0.2x SSC/ 0.1% SDS. Ten primary library plates (150mm) were screened using the Elol or Elo2 probes. Elol hybridized strongly to several plaques on each plate, whilst Elo2 hybridised to only three plaques in the ten plates screened. All Elol-hybridising plaques were picked from a single plate and carried through to a secondary screen, whilst all three Elo2-hybridising plaques were carried through to a secondary screen. Each secondary plaque was then used as a PCR template using the WO 2005/103253 PCT/AU2005/000571 88 forward and reverse primers flanking the multiple cloning site in the pBluescript phagemid and the PCR products electrophoresed on a 1% TAE gel. Following electrophoresis, the gel was blotted onto a Hybond N+ membrane and the membrane hybridised overnight with 3 2 P-labelled Elol and Elo2 probes. Six of the amplified Elol secondary plaques and one of the amplified Elo2 secondary plaques hybridised to the Elo1/2 probe (Figure 5). Two classes of elongase-like sequences were identified in the P. salina cDNA library on the basis of their hybdridisation to the Elol and Elo2 probes. Phagemids that hybridised strongly to either labelled fragment were excised with ExAssist Helper phage (Stratagene), and sequenced using the ABI Prism Big Dye Terminator kit (PE Applied Biosystems). All of the 5 inserts hybridizing to the Elol- probe were shown to be from the same gene. Similarly DNA sequencing of the 2 inserts hybridising to the Elo2 probe showed them to be from the same gene. The cDNA sequence of the Elol clone is provided as SEQ ID NO:8, and the encoded protein as SEQ ID NO:2, whereas the cDNA sequence of the Elo2 clone is provided as SEQ ID NO: 10, and the encoded proteins as SEQ ID NO:3, SEQ ID NO:85 and SEQ ID NO:86 using three possible start methionines). A comparison was performed of the Elol and Elo2 and other known PUFA elongases from the database using the PILEUP software (NCBI), and is shown in Figure 6. The Elol cDNA was 1234 nucleotides long and had a open reading frame encoding a protein of 302 amino acid residues. According to the PILEUP analysis, Elol clustered with other Elo-type sequences associated with the elongation of PUFA including A6 desaturated fatty acids (Figure 6). The Elol protein showed the greatest degree of identity (33%) to an elongase from the moss, P. patens (Accession No. AF428243) across the entire coding regions. The Elo1 protein also displayed a conserved amino acids motifs found in all other Elo-type elongases. The Elo2 cDNA was 1246 nucleotides long and had an open reading frame encoding a protein of 304 amino acid residues. According to PILEUP analysis, Elo2 clustered with other Elo-type sequences associated with the elongation of PUFA, including those with activity on A6 or A9 PUFA (Figure 6). Elo2 was on the same sub branch as the A9 elongase isolated from Isochrysis galbana (AX571775). Elo2 displayed 31% identity to the Isochrysis gene across its entire coding region. The Elo2 ORF also displayed a conserved amino acid motif found in all other Elo-type elongases.

WO 2005/103253 PCT/AU2005/000571 89 Example 8. Functional Characterization of A5 Fatty Acid Elongase in Yeast and Plant Cells Yeast The entire coding region of the P. salina Elol gene was ligated into pYES2, generating pYES2-psELOl, for characterisation in yeast.-This genetic construct was introduced into yeast strains and tested for activity by growth in media containing exogenous fatty acids as listed in the Table 8. Yeast cells containing pYES2-psELO1 were able to convert 20:5o3 into 22:5co3, confirming A5 elongase activity on C20 substrate. The conversion ratio of 7% indicated high activity for this substrate. The same yeast cells converted 18:403 (A6,9,12,15) to 20:4 3 and 18:30)6 (A6,9,12) to 20:3o6, demonstrating that the elongase also had A6 elongase activity in yeast cells, but at approximately 10-fold lower conversion rates (Table 8). This indicated that the Elo 1 gene encodes a specific or selective A5 elongase in yeast cells. This represents the first report of a specific A5 elongase, namely an enzyme that has a greater A5 elongase activity when compared to A6 elongase activity. This molecule is also the first A5 elongase isolated from an algael source. This enzyme is critical in the conversion of EPA to DPA (Figure 1). Plants The A5 elongase, Elo1 isolated from Pavlova is expressed in plants to confirm its ability to function in plants. Firstly, a plant expression construct is made for constitutive expression of Elol. For this purpose, the Elol sequence is placed under the control of the 35S promoter in the plant binary vector pBI121 (Clontech). This construct is introduced into Arabidopsis using the floral dip method described above. Analysis of leaf lipids is used to determine the specificity of fatty acids elongated by the Elo1 sequence. In another approach, co expression of the Elo1 construct with the zebra fish A5/A6 desaturase/C. elegans elongase construct and the A4 desaturase isolated from Pavlova, results in DHA synthesis from ALA in Arabidopsis seed, demonstrating the use of the A5 elongase in producing DHA in cells. In a further approach, the Elol gene may be co expressed with A6-desaturase and A5 desaturase genes, or a A6/A5 bifunctional desaturase gene, to produce DPA from ALA in cells, particularly plant cells. In an alternative approach, the A5 elongase and A4 elongase genes are used in combination with the PKS genes of Shewanella which produce EPA (Takeyama et al., 1997), in plants, for the synthesis of DHA.

WO 2005/103253 PCT/AU2005/000571 90 TABLE 8. Conversion of fatty acids in yeast cells transformed With genetic constructs expressing Elo1 or Elo2. Clone Fatty acid precursor/ (% Fatty acid formed/ (% Conversion ratio of total FA) of total FA) () pYES2- 20:5n-3 / 3% 7% psELO1 22:5n-3 / 0.21% pYES2- 18:4n-3 / 16.9% 0.89% psELO1 20:4n-3 / 0.15% pYES2- 18:3n-6 / 19.8% 0.71% psELO1 20:3n-6 / 0.14% pYES2- 20:5n-3 /2.3% psELO2 22:5n-3 / tr pYES2- 18:4n-3 /32.5% 1.2% psELO2 20:4n-3 /0.38% pYES2- 18:3n-6 / 12.9% 0.62% psELO2 20:3n-6 /0.08% pYES2- 18:2n-6 /30.3% 0.40% psELO2 20:2n-6 / 0.12% pYES2- 18:3n-3 / 42.9% 0.47% psELO2 18:3n-3 / 0.20% tr: trace amounts (<0.02%) detected. Example 9. Functional Characterization of A9 Fatty Acid Elongase in Yeast and Plant Cells Expression in yeast cells The entire coding region of the P. salina Elo2 gene encoding a protein of 304 amino acids (SEQ ID NO:3) was ligated into pYES2, generating pYES2-psELO2, for characterisation in yeast. This genetic construct was introduced into yeast strains and tested for activity by growth in media containing exogenous fatty acids. Yeast cells containing pYES2-psELO2 were able to convert 18:2o6 into 20:2o6 (0.12% of total fatty acids) and 18:3o3 into 20:3o3 (0.20%), confirming A9 elongase activity on C18 substrates (Table 8). These cells were also able to convert 18:3o6 into 20:3o6 and 18:43 into 20:4o3, confirming A6 elongase activity on C18 substrates in yeast. However, since the 18:3co6 and 18:4o3 substrates also have a desaturation in the A9 WO 2005/103253 PCT/AU2005/000571 91 position, it could be that the Elo2 enzyme is specific for A9-desaturated fatty acids, irrespective of whether they have a A6 desaturation as well. The cells were able to convert 20:5o3 into the 22:5 product DPA. This is the first report of a A9 elongase that also has A6 elongase activity from a non-vertebrate source, in particular from a fungal or algal source. As the coding region contained three possible ATG start codons corresponding to methionine (Met) amino acids at positions 1, 11 (SEQ ID NO:85) and 29 (SEQ ID NO:86) of SEQ ID NO:3, the possibility that polypeptides beginning at amino acid positions 11 or 29 would also be active was tested. Using 5' oligonucleotide (sense) primers corresponding to the nucleotide sequences of these regions, PCR amplification of the coding regions was performed, and the resultant products digested with EcoRI. The fragments are cloned into pYES2 to form pYES2-psELO2-11 and pYES2-psELO2-29. Both plasmids are shown to encode active A9-elongase enzymes in yeast. The three polypeptides may also be expressed in Synechococcus or other cells such as plant cells to demonstrate activity. Expression in plant cells The A9 elongase gene, Elo2, isolated from Pavlova was expressed in plants to confirm its ability to function in plants. Firstly, a plant expression construct is made for constitutive expression of Elo2. For this purpose, the Elo2 coding sequence from amino acid position 1 of SEQ ID NO:3 was placed under the control of the 35S promoter in the plant binary vector pBI121 (Clontech). This construct is introduced into Arabidopsis using the floral dip method described above. Analysis of leaf lipids indicates the specificity of fatty acids that are elongated by the Elo2 sequence. Co-expression of A9 elongase and A8-desaturase genes in transformed cells The P. salina A8-desaturase and A9-elongase were cloned into a single binary vector, each under the control of the constitutive 35S promoter and nos terminator. In this gene construct, pBIl21 containing the A8-desaturase sequence was cut with HindIII and ClaI (blunt-ended) to release a fragment containing the 35S promoter and the A8 desaturase gene, which was then ligated to the HindIII + SacI (blunt ended) cut pXZP143/A9-elongase vector to result in the intermediate pJRP013. This intermediate was then opened with HindII and ligated to a pWvec8/A9-elongase binary vector (also HindIII-opened) to result in the construct pJRP014, which contains both genes between the left and right T-DNA borders, together with a hygromycin selectable marker gene suitable for plant transformation.

WO 2005/103253 PCT/AU2005/000571 92 This double-gene construct was then used to transform tobacco using a standard Agrobacterium-mediated transformation technique. Following introduction of the construct into Agrobacterium strain AGLI, a single transformed colony was used to inoculate 20 mL of LB media and incubated with shaking for 48 hours at 28'C. The cells were pelleted (1000 g for 10 minutes), the supernatant discarded, and the pellet resuspended in 20 mL of sterile MS media. This was step was then repeated before 10 ml of this Agrobacterial solution was added to freshly cut (1 cm squares) tobacco leaves from cultivar W38. After gentle mixing, the tobacco leaf pieces and Agrobacterium solution were allowed to stand at room temperature for 10 min. The leaf pieces were transferred to MS plates, sealed, and incubated (co-cultivation) for 2 days at 24'C. Transformed cells were selected on medium containing hygromycin, and shoots regenerated. These shoots were then cut off and transferred to MS-rooting media pots for root growth, and eventually transferred to soil. Both leaf and seed lipids from these plants are analysed for the presence of 20:2o6, 20:3co6, 20:3co3 and 20:403 fatty acids, demonstrating the co-expression of the two genes. Discussion Biochemical evidence suggests that the fatty acid elongation consists of 4 steps: condensation, reduction, dehydration and a second reduction, and the reaction is catalysed by a complex of four proteins, the first of which catalyses the condensation step and is commonly called the elongase. There are 2 groups of condensing enzymes identified so far. The first are involved in the synthesis of saturated and monounsaturated fatty acids (C18-22). These are the FAE-like enzymes and do not play any role in LC PUFA biosynthesis. The other class of elongases identified belong to the ELO family of elongases named after the ELO gene family whose activities are required for the synthesis of very LC fatty acids of sphingolipids in yeast. Apparent paralogs of the ELO type elongases isolated from LC-PUFA synthesizing organisms like algae, mosses, fungi and nematodes have been shown to be involved in the elongation and synthesis of LC PUFA. It has been shown that only the expression of the condensing component of the elongase is required for the elongation of the respective acyl chain. Thus the introduced condensing component of the elongase is able to successfully recruit the reduction and dehydration activities from the transgenic host to carry out successful acyl elongations. This was also true for the P. salina A9-elongase.

WO 2005/103253 PCT/AU2005/000571 93 Example 10. Isolation of a Gene Encoding a A4-Desaturase from P. salina The final step in the aerobic pathway of DHA synthesis in organisms other than vertebrates, such as microorganisms, lower plants including algae, mosses, fungi, and possibly lower animals, is catalysed by a A4-desaturase that introduces a double bond into the carbon chain of the fatty acid at the A4 position. Genes encoding such an enzyme have been isolated from the algae Euglena and Pavlova and from Thraustochytrium, using different approaches. For example, A4-desaturase genes from Pavlova lutheri and Euglena gracilis were isolated by random sequencing of cloned ESTs (EST approach, Meyer et al., 2003; Tonon et al., 2003), and a A4-desaturase gene from Thraustochytrium sp. ATCC21685 was isolated by RT-PCR using primers corresponding to a cytochrome b 5 HPGG domain and histidine box III region (Qiu et al., 2001). The cloned A4 desaturase genes encoded front-end desaturases whose members are characterised by the presence of an N-terminal cytochrome b 5 -like domain (Napier et al., 1999; Sayanova and Napier, 2004). Isolation of a gene fragment from a A4-desaturase gene from P. salina Comparison of known moss and microalgae A4-desaturases revealed several conserved motifs including a HPGG (SEQ ID NO:52) motif within a cytochrome b 5 - like domain and three histidine box motifs that are presumed to be required for activity. Novel degenerate PCR primers PavD4Des-F3 (5'-AGCACGACGSSARCCACGGCG 3') (SEQ ID NO:53) and PavD4Des-R3 (5'-GTGGTGCAYCABCACGTGCT-3') (SEQ ID NO:54) corresponding to the conserved amino acid sequence of histidine box I and complementary to a nucleotide sequence encoding the amino acid sequence of histidine box II, respectively, were designed as to amplify the corresponding region of P. salina desaturase genes, particularly a A4-desaturase gene. The use of degenerate PCR primers corresponding to histidine box I and histidine box II regions of A4-desaturase has not been reported previously. PCR amplification reactions using these primers were carried out using P. salina first strand cDNA as template with cycling of 95 0 C, 5min for 1 cycle, 94 0 C 30 sec, 57'C 30 sec, 72 0 C 30 see for 35 cycles, and 72 0 C 5 min for 1 cycles. The PCR products were cloned into pGEM-T-easy (Promega) vectors, and nucleotide sequences were determined with an AB13730 automatic sequencer using a reverse primer from the pGEM-Teasy vector. Among 14 clones sequenced, three clones showed homology to A4-desaturase genes. Two of these three clones are truncated at one primer end. The nucleotide sequence of the cDNA insert of the third, clone 1803, is provided as SEQ ID NO:11.

WO 2005/103253 PCT/AU2005/000571 94 The amino acid sequence encoded by SEQ ID NO: 11 was used to search the NCBI protein sequence database using the BLASTX software. The results indicated that this sequence was homologous to known A4-desaturases. The amino acid sequence of the P. salina gene fragment showed 65%, 49%, 46% and 46% identity to that of A4 desaturases of P. lutheri, Thraustochytrium sp. ATCC21685, Thraustochytrium aureum and Euglena gracilis respectively. Isolation of a full-length A4-desaturase gene The insert from clone 1803 was excised, and used as probe to isolate full-length cDNAs corresponding to the putative A4-desaturase gene fragment. About 750,000 pfu of the P. salina cDNA library were screened at high stringency. The hybridization was performed at 60 0 C overnight and washing was done with 2xSSC/0.1%SDS 30min at 65 0 C then with 0.2xSSC/0.1%SDS 30min at 65C. Eighteen hybridising clones were isolated and secondary screening with six clones was performed under the same hybridization conditions. Single plaques from secondary screening of these six clones were isolated. Plasmids from five single plaques were excised and the nucleotide sequences of the inserts determined with an ABI 3730 automatic sequencer with reverse and forward primers from the vector. Sequencing results showed that four clones each contained A4-desaturase cDNA of approximately 1.7kb in length, each with the same coding sequence and each apparently full-length. They differed slightly in the length of the 5' and 3' UTRs even though they contained identical protein coding regions. The cDNA sequence of the longest P. salina A4-desaturase cDNA is provided as SEQ ID NO:13, and the encoded protein as SEQ ID NO:4. The full-length cDNA was 1687 nucleotides long and had a coding region encoding 447 amino acids. The Pavlova salina A4-desaturase showed all the conserved motifs typical of 'front-end desaturases' including the N-terminal cytochrome b 5 -like domain and three conserved histidine-rich motifs. Comparison of the nucleotide and amino acid sequences with other A4-desaturase genes showed that the greatest extent of homology was for the P. lutheri A4-desaturase (Accession No. AY332747), which was 69.4% identical in nucleotide sequence over the protein coding region, and 67.2% identical in amino acid sequence. Demonstration of enzyme activity of Paylova salina A4-desaturase gene A DNA fragment including the Pavlova salina A4-desaturase cDNA coding region was excised as an EcoRI-SalI cDNA fragment and inserted into the pYES2 yeast expression vector using the EcoRl and XhoI sites. The resulted plasmid was transformed WO 2005/103253 PCT/AU2005/000571 95 into yeast cells. The transformants were grown in YMM medium and the gene induced by the addition of galactose, in the presence of added (exogenous) o6 and o3 fatty acids in order to demonstrate enzyme activity and the range of substrates that could be acted upon by the expressed gene. The fatty acids 22:503 (DPA, 1.0mM), 20:4n-3 (ETA, 1.0mM), 22:4(o6 (DTAG, 1.0 mM) and 20:40o6 (ARA, 1.0mM) were each added separately to the medium. After 72 hours incubation, the cells were harvested and fatty acid analysis carried out by capillary gas-liquid chromatography (GC) as described in Example 1. The data obtained are shown in Table 9. TABLE 9. Yeast PUFA feeding showing activity of delta-4 desaturase gene. Exogenous fatty acid added to -growth medium Fatty acid composition (% of total fatty acid) 22:4co6 22:5o3 14:0 0.63 0.35 15:0 0.06 0.06 16:1o7c 43.45 40.52 16:10o5 0.20 0.13 16:0 18.06 15.42 17:10o8 0.08 0.09 17:0 0.08 18:10o9 26.73 30.07 18:10o7 (major) & 18:30o3 1.43 1.61 18:1o5c 0.02 tr 18:0 7.25 8.87 20:503 0.40 0.62 20:1o9A/ o11 0.03 tr 20:0 0.08 0.09 22:5(o6 0.03 0.00 22:60o3 0.04 22:4o6 0.97 22:5o3 0.00 1.66 22:0 0.06 0.06 24:10o7 0.31 0.37 24:0 0.12 0.04 Sum 100.00% 100.00% WO 2005/103253 PCT/AU2005/000571 96 This showed that the cloned gene encoded a A4-desaturase which was able to desaturate both C22:4o6 (3.0% conversion to 22:5o)6) and C22:5o3 (2.4% conversion to 22:603) at the A4 position. The enzyme did not show any A5 desaturation activity when the yeast transformants were fed C20:3o6 or C20:4o3. Example 11. Expression of P. salina A4-desaturase Gene in Plant Cells and Production of DHA To demonstrate activity of the A4-desaturase gene in plant cells, the coding region may be expressed either separately to allow the conversion of DPA to DHA, or in the context of other LC-PUFA synthesis genes such as, for example, a A5-elongase gene for the conversion of EPA to DHA. For expression as a separate gene, the A4-desaturase coding region may be excised as a BamHl-SaI fragment and inserted between a seed specific promoter and a polyadenylation/transcription termination sequence, such as, for example, in vector pGNAP (Lee et al., 1998), so that it is expressed under the control of the seed specific promoter. The expression cassette may then be inserted into a binary vector and introduced into plant cells. The plant material used for the transformation may be either untransformed plants or transformed plants containing a construct which expressed the zebrafish A5/A6-dual desaturase gene and C. elegans elongase gene each under the control of a seed specific promoter (Example 5). Transgenic Arabidopsis containing the latter, dual-gene construct had successfully produced EPA and DPA in seeds, and the combination with the A4-desaturase gene would allow the conversion of the DPA to DHA in the plant cells, as demonstrated below. To demonstrate co-expression of a A5-elongase gene with the A4-desaturase gene in recombinant cells, particularly plant cells, and allow the production of DHA, the A4 desaturase and the A5-elongase genes from P. salina (Example 8) were combined in a binary vector as follows. Both coding regions were placed under the control of seed specific (napin) promoters and nos3' terminators, and the binary vector construct had a kanamycin resistance gene as a selectable marker for selection in plant cells. The coding region of the A5-elongase gene was excised from its cDNA clone as a PstI-SacI fragment and inserted into an intermediate plasmid (pXZP143) between the promoter and terminator, resulting in plasmid pXZP144. The coding region of the A4-desaturase gene was excised from its cDNA clone as a BamHI -Sal fragment and inserted into plasmid pXZP143 between the promoter and nos 3' transcription terminator, resulting in plasmid pXZP150. These two expression cassettes were combined in one vector by inserting the HindII-ApaI fragment from pXZP144 (containing promoter-Elo1-nos3') between the StuI and ApaI sites of pXZP150, resulting in plasmid pXZP191. The HindII-StuI WO 2005/103253 PCT/AU2005/000571 97 fragment from pXZP191 containing both expression cassettes was then cloned into the binary vector pXZP330, a derivative of pBIl21, resulting in plant expression vector pXZP355. This vector is shown schematically in Figure 7. Plant transformation The A5-elongase and the A4-desaturase genes on pXZP355 were introduced by the Agrobacteriun-mediated floral dip transformation method into the Arabidopsis plants designated DOll (Example 5) which were already transgenic for the zebrafish A5/A6 bifunctional desaturase and the C. elegans A5/A6 bifunctional elongase genes. Since those transgenes were linked to a hygromycin resistance gene as a selectable marker gene, the secondary transformation with pXZP355 used a kanamycin resistance selection, thus distinguishing the two sets of transgenes. Five transgenic plants are obtained, designated "DW" plants. Since the DOl 1 plants were segregating for the zebrafish A5/A6 bifunctional desaturase and the C. elegans A5/A6 bifunctional elongase genes, some of the transformed plants were expected to be heterozygous for these genes, while others were expected to be homozygous. Seed (T2 seed) of the five transformed plants were analysed and shown to contain up to at least 0.1% DPA and up to at least 0.5% DHA in the seed oils. Data are presented for two lines in Table 10. Analysis, by mass spectrometry (GC-MS), of the fatty acids in the peaks identified as EPA and DHA from the GC analysis proved that they were indeed EPA and DHA (Figure 8). The fatty acid analysis of the T2 seedoil demonstrated that significant conversion of EPA to DHA had occurred in the DW2 and DW5 lines, having 0.2% and. 0.5% DHA, respectively. Examination of the enzyme efficiencies in plant DW5 containing the higher level of DHA showed that 17% of the EPA produced in its seed was elongated to DPA by the P. salina A5-elongase, and greater than 80% of this DPA was converted to DHA by the P. salina A4-desaturase. Since the A5-elongase and A4-desaturase genes were segregating in the T2 seed, the fatty acid composition data represented an average of pooled null, heterozygous and homozygous genotypes for these genes. It is expected that levels of DHA in progeny lines of DW5 will be greater in seed that is uniformly homozygous for these genes.

WO 2005/103253 PCT/AU2005/000571 98 TABLE 10. Fatty acid composition (% of total fatty acids) of seed oils fromArabidopsis thaliana (ecotype Columbia) and derivatives carrying EPA and DHA gene constructs EPA, DPA and DHA synthesis in transgenic seed. Fatty acid Wild type DO 1l + DHA construct Columbia DW2 DW5 Usualfatty acids Total Total TAG PL 16:0 7.2 6.7 6.1 5.5 12.5 18:0 2.9 3.8 4.4 4.3 4.5 18:1A 9 20.0 20.6 16.6 18.9 13.7 18:2A 9

',

12 (LA) 27.5 26.0 25.9 25.5 33.1 18:3A 9 1 2

,

5 (ALA) 15.1 13.2 15.0 13.6 15.1 20.0 2.2 2.1 1.8 1.9 0.6 20: 1A 19.8 14.8 10.5 10.5 3.2 20:1A' 3 2.2 3.0 4.2 4.8 1.4 20:2A"' 1 4 0.1 1.7 3.5 3.8 3.7 22:1A1 3 1.5 1.4 1.0 0.3 0.4 Other minor 1.5 2.9 2.7 2.4 3.8 Total 100.0 96.0 91.7 91.5 92.0 New co6-PUFA 18:3A 6 ,9,1 2 (GLA) 0 0.2 0.4 0.4 0.2 20:3A, 1 '1, 1 4 0 0.8 1.5 1.5 1.7 20:4A' 8

'

4 (ARA) 0 0.4 1.0 1.1 1.2 22:4A7,10,13,16 0 0 0 0 0.2 22:5A4,7,10,13,1 6 0 0 0.1 0.1 0.1 Total 0 1.4 3.0 3.1 3.4 New m3-PUFA 18:4A', 9 ,12,1 5 (SDA) 0 0.7 1.5 1.6 0.5 20:4A8,''14,1 7 0 0.5 0.8 0.7 0.9 20:5A5' 8 "''14,1 7 (EPA) 0 1.1 2.4 2.5 2.3 22:5A 7 ,10,13,16,1 9 (DPA) 0 0.1 0.1 0.2 0.7 22:6A 4 ,7,10,13,16,19 (DHA) 0 0.2 0.5 0.4 0.2 Total 0 2.6 5.3 5.4 4.6 Totalfatty acids 100.0 100.0 100.0 100.0 100.0 Total MUFAa 41.3 36.8 28.1 29.7 17.3 WO 2005/103253 PCT/AU2005/000571 99 Total C 18 -PUFAb 42.6 39.2 40.9 39.1 48.2 Total new PUFA* 0 4.0 8.3 8.5 8.0 a Total of 18:1A 9 and derived LC-MIUFA (=18: 1A 9 + 20:1A" + 22:1A) b 18:2 +18:3 'Total of all new o6 and o3-PUFA Germination of 50 T2 seed from each of DW2 and DW5 on hygromycin containing medium showed that the DW5 TI plant was homozygous (50/50) for the A5/A6 bifunctional desaturase and A5/A6 bifunctional elongase genes, while the DW2 seed segregating in a 3:1 ration (resistant: susceptible) for these genes and DW2 was therefore heterozygous. This was consistent with the higher levels of EPA observed in DW5 seed compared to DW2 seed, and explained the increased level of DHA produced in the seed homozygous for these transgenes. This further demonstrated the desirability of seed that are homozygous for the trait. We also noted the consequences of LC-PUFA synthesis on the overall fatty acid profile in these seed. Although we observed accumulation of new o6 and o3 PUFA (i.e. products of A6-desaturation) at levels of greater than 8% in DW5 seed, these seed had levels of the precursor fatty acids LA and ALA that were almost the same as in the wild type seed. Rather than depleting LA and ALA, the levels of monounsaturated fatty acid C18: 1A 9 and its elongated derivatives (20: 1A 11 and 22: 1A 3 ) were significantly reduced. Thus it appeared that conversion of C 18 -PUFA to LC-PUFA resulted in increased conversion of 18:1 to LA and ALA, and a corresponding reduction in 18:1 available for elongation. The plant expression vector pXZP355 containing the A4-desaturase and the A5-elongase genes was also used to introduce the genes into plants of the homozygous line DOl 1-5, and 20 transgenic TI plants were obtained. The levels of DHA and DPA in T2 seed from these plants were similar to those observed in seed from DW5. Reductions in the levels of the monounsaturated fatty acids were also observed in these seed. Fractionation of the total seed lipids of DW5 seed revealed them to be comprised of 89% TAG and 11% polar lipids (largely made up phospholipids). Furthermore, fatty acid analysis of the TAG fraction from DW5 seed showed that the newly synthesised EPA and DHA were being incorporated into the seed oil and that the proportion of EPA and DHA in the fatty acid composition of the total seed lipid essentially reflected that of the TAG fraction (Table 10).

WO 2005/103253 PCT/AU2005/000571 100 Example 12. Isolation of Homologous Genes from Other Sources Homologs of the desaturase and elongase genes such as the P. salina genes described herein may be readily detected in other microalgae or other sources by hybridization to labelled probes derived from the genes, particularly to parts or all of the coding regions, for example by Southern blot hybridization or dot-blot hybridisation methods. The homologous genes may be isolated from genomic or cDNA libraries of such organisms, or by PCR amplification using primers corresponding to conserved regions. Similarly, homologs of vertebrate desaturases with high affinity for Acyl-CoA and/or freshwater fish bifunctional desaturases can be isolated by similar means using probes to the zebrafish A5/A6 desaturase. Dot Blot Hybridisations Genomic DNA from six microalgae species was isolated using a DNAeasy kit (Qiagen) using the suppliers instructions, and used in dot blot hybridization analyses for identification of homologous genes involved in LC-PUFA synthesis in these species. This also allowed evaluation of the sequence divergence of such genes compared to those isolated from Pavlova salina. The species of microalga examined in this analysis were from the genera Melosira, Rhodomonas, Heterosigma, Nannochloropsis, Heterocapsa and Tetraselm is. They were identified according to Hasle, G. R. & Syvertsen, E. E. 1996 Dinoflagellates. In: Tomas, C. R.(ed.) Identifying Marine Phytoplankton. Academic Press, San Diego, CA. pp 531-532. These microalga were included in the analysis on the basis of the presence of EPA, DHA, or both when cultured in vitro (Example 2). Genomic DNA (approximately 100ptg) isolated from each of the microalga was spotted onto strips of Hybond N+ membrane (Amersham). After air drying, each membrane strip was placed on a layer of 3MM filter paper saturated with 0.4 M NaOH for 20 min, for denaturation of the DNA, and then rinsed briefly in 2x SSC solution. The membrane strips were air dried and the DNA cross linked to the membranes under UV light. Probes labeled with 32 P nucleotides and consisting of the coding regions without the untranslated regions of a number of Pavlova-derived genes, including the A8, A5 and A4 desaturases and A9 and A5 elongases, were prepared and hybridized to each membrane strip/DNA dot blot. The membranes were hybridized with each probe overnight in a buffer containing 50 mM Tris-HCl, pH7.5, 1M NaCl, 50% formamide, lOx Denhardt's solution, 10% dextran sulfate, 1%SDS, 0.1% sodium pyrophosphate, and 0.1 mg/ml herring sperm DNA, at 42*C, then washed three times in a solution containing 2x SSC, 0.5% SDS at 50 0 C for 15 min each (low stringency wash in this experiment) or for a high stringency wash in 0.2x SSC, 0.5% SDS at 65*C for 20 minutes each.

WO 2005/103253 PCT/AU2005/000571 101 It is well understood that the stringency of the washing conditions employed in DNA blot/hybridizations can reveal useful information regarding the sequence relatedness of genes. Thus hybridizations maintained when subjected to a high stringency wash indicate a high level of sequence relatedness (e.g. 80% or greater nucleotide identity over at least 100-200 nucleotides), while hybridizations maintained only during low stringency washes indicate a relatively lower degree of DNA conservation between genes (e.g. 60% or greater nucleotide identity over at least 200 nucleotides). The hybridized dot blots were exposed to BioMax X-ray film (Kodak), and the autoradiograms are shown in Figure 9. The autoradiograms reveal the presence of homologs to the P. salina LC-PUFA genes in these species, and moreover reveal a range of homologies based on the different levels of hybridization seen under the high and low stringency conditions. It appeared that some of the microalgal species examined have LC-PUFA genes that may differ substantially from the genes in P. salina, while others are more related in sequence. For example, genes from Tetraselmis sp appeared to be highly similar to the A4- and A5-desaturases and the A5 elongase from Pavlova salina on the basis of the strength of hybridizations. In contrast, all of the LC-PUFA genes identified in Melosira sp appeared to have lower degrees of similarity to the P. salina genes. Isolation of an LC-PUFA elongase gene from Heterocapsa sp. Heterocapsa spp. such as Heterocapsa niei in the CSIRO collection (Example 2) are dinoflagellates that were identified as producers of LC-PUFA including EPA and DHA. To exemplify the isolation of LC-PUFA synthesis genes from these dinoflagellates, DNA was purified from cells of a Heterocapsa niei strain originally isolated in Port Hacking, NSW, Australia in 1977. DNA was isolated using a DNAeasy kit (Qiagen) using the suppliers instructions. Based on published multiple amino acid sequence alignments for fatty acid elongases (Qi et al., 2002; Parker-Barnes et al., 2000), the consensus amino acid blocks FLHXYH (SEQ ID NO:48) and MYXYYF (SEQ ID NO:49) were identified and corresponding degenerate primers encoding these sequences 5'-CAGGATCCTTYYTNCATNNNTAYCA-3' (SEQ ID NO:50) (sense) or complementary to these sequences 5'-GATCTAGARAARTARTANNNRTACAT-3' (SEQ ID NO:51) (antisense) were synthesised. PCR amplification reactions were carried out in reaction volumes of 20pL with 20pmol of each primer, 200ng of Heterocapsa sp. genomic DNA and Hotstar Taq DNA polymerase (Qiagen) with buffer and nucleotide components as specified by the supplier. The reactions were cycled as follows: 1 cycle of WO 2005/103253 PCT/AU2005/000571 102 95*C for 15 minutes, 5 cycles of 95"C, 1min, 38 0 C, 1min, 72 0 C, 1min, 35 cycles of 95'C, 35 sec, 52 0 C, 30 sec, 72 0 C, 1min, 1 cycle of 72'C, 10min. Fragments of approximately 350bp were generated and ligated into pGEM-Teasy for sequence analysis. Of eight clones isolated, two identical clones had nucleotide and encoded amino acid sequences with similarity to regions of known elongases. These were designated Het350Elo, and the nucleotide and amino acid sequences are provided as SEQ ID NO:79 and SEQ ID NO:80 respectively. BLAST analysis and the presence of an in-frame stop codon suggested the presence of an intron between approximate positions 33 and 211. The best matches to the amino acid sequence were animal elongase sequences, see for example Meyer et al. (2004), indicating that the isolated Heterocapsa gene sequence was probably involved in elongation of C18 and C20 fatty acid substrates. Full-length clones of the elongase can readily be isolated by screening a Heterocapsa cDNA library or by 5'- and 3' RACE techniques, well known in the art. Construction of Melosira sp. cDNA library and EST sequencing mRNA, for the construction of a cDNA library, was isolated from Melosira sp. cells using the following method. 2 g (wet weight) of Melosira sp. cells were powdered using a mortar and pestle in liquid nitrogen and sprinkled slowly into a beaker containing 22 ml of extraction buffer that was being stirred constantly. To this, 5% insoluble polyvinylpyrrolidone, 90mM 2-mercaptoethanol, and 10mM dithiotheitol were added and the mixture stirred for a further 10 minutes prior to being transferred to a CorexTM tube. 18.4 ml of 3M ammonium acetate was added and mixed well. The sample was then centrifuged at 6000xg for 20 minutes at 4*C. The supernatant was transferred to a new tube and nucleic acid precipitated by the addition of 0.1 volume of 3M NaAc (pH 5.2) and 0.5 volume of cold isopropanol. After 1 hour incubation at -20 0 C, the sample was centrifuged at 6000xg for 30 minutes in a swing-out rotor. The pellet was resuspended in 1 ml of water and extracted with phenol/chloroform. The aqueous layer was transferred to a new tube and nucleic acids were precipitated once again by the addition of 0.1 volume 3M NaAc (pH 5.2) and 2.5 volume of ice cold ethanol. The pellet was resuspended in water, the concentration of nucleic acid determined and then mRNA was isolated using the Oligotex mRNA system (Qiagen). First strand cDNA was synthesised using an oligo(dT) linker-primer supplied with the ZAP-cDNA synthesis kit (Stratagene - cat # 200400) and the reverse transcriptase SuperscriptIII (Invitrogen). Double stranded cDNA was ligated to EcoRI adaptors and from this a library was constructed using the ZAP-cDNA synthesis kit as described in the accompanying instruction manual (Stratagene - cat # 200400). A primary library of 1.4 x WO 2005/103253 PCT/AU2005/000571 103 106 plaque forming units (pfa) was obtained. The average insert size of cDNA inserts in the library was 0.9 kilobases based on 47 random plaques and the percentage of recombinants in the library was 99%. Single pass nucleotide sequencing of 8684 expressed sequence tags (ESTs) was performed with SK primer (5'-CGCTCTAGAACTAGTGGATC-3') (SEQ ID NO:87) using the ABI BigDye system. Sequences of 6750 ESTs were longer than 400 nucleotides, showing the inserts were at least this size. ESTs showing homology to several fatty acid desaturases and one PUFA elongase were identified by BlastX analysis. The amino acid sequence (partial) (SEQ ID NO:88) encoded by the cDNA clone Mm301461 showed 75% identity to Thalassiosira pseudonana fatty acid elongase 1 (Accession No. AY591337). The nucleotide sequence of EST clone Mm301461 is provided as SEQ ID NO:89. The high degree of identity to a known elongase makes it highly likely that Mn301461 encodes a Melosira fatty acid elongase. RACE techniques can readily be utilized to isolate the full-length clone encoding the elongase. Example 13. Isolation of FAE-Iike elongase gene fragment from P. salina Random cDNA clones from the P. salina cDNA library were sequenced by an EST approach. In an initial round of sequencing, 73 clones were sequenced. One clone, designated 11.B1, was identified as encoding a protein (partial sequence) having sequence similarity with known beta keto-acyl synthase-like fatty acid elongases, based on BLASTX analysis. The nucleotide sequence of 1 1.B1 from the 3' end is provided as (SEQ ID NO:55). These plant elongases are different to the ELO class elongase in that they are known to be involved in the elongation of C16 to C18 fatty acids and also the elongation of very-long-chain saturated and monounsaturated fatty acids. Clone 11 .B 1, represents the first non-higher plant gene in this class isolated. Example 14. Isolation of a Gene Encoding a A5-Desaturase from P. saina Isolation of a gene fragment from a A5-desaturase gene from P. salina In order to isolate a A5-desaturase gene from P. salina, oligonucleotides were designed for a conserved region of desaturases. The oligonucleotides designated d5A and d5B shown below were made corresponding to a short DNA sequence from a A5 desaturase gene from Pavlova lutheri. Oligo d5A: 5' TGGGTTGAGTACTCGGCCAACCACACGACCAACTGCGCGCCCTCGTGGTGGT GCGACTGGTGGATGTCTTACCTCAACTACCAGATCGAGCATCATCTGT-3' (nucleotides 115-214 of International patent application published as WOO3078639-A2, WO 2005/103253 PCT/AU2005/000571 104 Figure 4a) (SEQ ID NO:56) and oligo d5B: 5' ATAGTGCAGCCCGTGCTTCTCGAAGAGCGCCTTGACGCGCGGCGCGATCGTC GGGTGGCGGAATTGCGGCATGGACGGGAACAGATGATGCTCGATCTGG-3' (corresponding to the complement of nucleotides 195-294 of W003078639-A2, Figure 4a) (SEQ ID NO:57). These oligonucleotides were annealed and extended in a PCR reaction. The PCR product from was inserted into pGEM-T Easy vector and the nucleotide sequence confirmed. The cloned fragment was labelled and used as a hybridization probe for screening of a Pavlova salina cDNA library under moderately high stringency conditions, hybridizing at 55C overnight with an SSC hybridization solution and washing the blots at 60 0 C with 2x SSC/0.1% SDS three times each for 10 minutes. From screening of about 500,000 plaques, 60 plaques were isolated which gave at least a weak hybridization signal. Among 13 clones that were sequenced, one clone designated p1918 contained a partial-length cDNA encoding an amino acid sequence with homology to known A5 desaturase genes. For example, the amino acid sequence was 53% identical to amino acid residues 210-430 from the C-terminal region of a Thraustochytrium A5-desaturase gene (Accession No. AF489588). Isolation of a full-length A5-desaturase gene The partial-length sequence in p1918 was used to design a pair of sequence specific primers, which were then used in PCR screening of the 60 isolated plaques mentioned above. Nineteen of the 60 were positive, having the same or similar cDNA sequence. One of the clones that showed a strong hybridization signal using the partial length sequence as a probe was used to determine the full-length sequence provided as SEQ ID NO:58, and the amino acid sequence (425 amino acids in length) encoded thereby is provided as SEQ ID NO:60. The amino acid sequence was used to search the NCBI protein sequence database using the BLASTX software. The results indicated that this sequence was homologous to known A5-desaturases. The amino acid sequence of the P. salina protein showed 81% identity to a P. lutheri sequence of undefined activity in W003/078639-A2, and 50% identity to a A5-desaturase from Thraustochytrium (Accession No. AF489588). The Pavlova salina A5-desaturase showed all the conserved motifs typical of 'front-end desaturases' including the N-terminal cytochrome b 5 -like domain and three conserved histidine-rich motifs.

WO 2005/103253 PCT/AU2005/000571 105 Co-expression of A9 elongase, A8-desaturase and A5-desaturase genes in transformed cells Co-expression of the A5-desaturase gene together with the A9 elongase gene (Elo2, Example 7) and the A8-desaturase gene (Example 6) was achieved in cells as follows. The plant expression vector pXZP354 containing the three genes, each from P. salina, and each expressed from the seed specific napin promoter was constructed. The P. salina A8-desaturase coding region from the cDNA clone (above) was first inserted as a BamHI-NcoI fragment into pXZP143 between the seed specific napin promoter and Nos terminator, resulting in plasmid pXZP146. The P. salina A9-elongase gene was likewise inserted, as a PstI-XhoI fragment from its cDNA clone, into pXZP143 resulting in plasmid pXZP143-Elo2. The P. salina A5-desaturase gene was also inserted, as a PstI BssHII fragment from its cDNA clone, into pXZP143, resulting in plasmid pXZP147. Then, the HindIII-ApaI fragment containing the A9-elongase expression cassette from pXZP143-Elo2 was inserted into pXZP146 downstream of the A8-desaturase expression cassette, resulting in plasmid pXZP148. The HindIII-ApaI fragment containing the A5 desaturase expression cassette from pXZP147 was inserted into pXZP148 downstream of A8-desaturase and A9-elongase expression cassettes, resulting in plasmid pXZP149. Then, as a final step, the HindII-ApaI fragmetit containing the three genes from pXZP149 was inserted into a derivative of the binary vector pART27, containing a hygromycin resistance gene selection marker, resulting in plant expression plasmid pXZP354. Plasmid pXZP354 was introduced into Arabidopsis by the Agrobacterium mediated floral dip method, either in the simultaneous presence or the absence of expression plasmid pXZP355 (Example 11) containing the P. salina A5-elongase and A4 desaturase genes. Co-transformation of the vectors could be achieved since they contained different selectable marker genes. In the latter case, the transgenic plants (designated "DR" plants) were selected using hygromycin as selective agent, while in the former case, the plants ("DU" plants) were selected with both hygromycin and kanamycin. Twenty-one DR plants (T1 plants) were obtained. Fatty acid analysis of seedoil from T2 seed from ten of these plants showed the presence of low levels of 20:2o (EDA), 20:3o6 (DGLA) and 20:4o6 (ARA), including up to 0.4% ARA. Fatty acid analysis of seedoil from T2 seed from seven DU plants showed similar levels of these fatty acids. From the relative ratios of these fatty acids, it was concluded that the A5-desaturase and A8-desaturase genes were functioning efficiently in seed transformed with pXZP354 but that the activity of the A9 elongase gene was suboptimal. It is likely that shortening of the WO 2005/103253 PCT/AU2005/000571 106 coding region at the N-terminal end, to initiate translation at amino acid position 11 or 29 of SEQ ID NO:3 (Example 9) (see SEQ ID NO's 85 and 86) will improve the level of activity of the A9 elongase gene. Expression of one or two of the genes from seed specific promoters other than the napin promoter, so they are not all expressed from the napin promoter, is also expected to improve the expression level of the A9 elongase gene. Example 15. Isolation of a Gene Encoding a A6-Desaturase from Echium plantagineum Some plant species such as evening primrose (Oenothera biennis), common borage (Borago officinalis), blackcurrant (Ribes nigrum), and some Echium species belonging to the Boragenacae family contain the o6- and co3-desaturated C18 fatty acids, y-linolenic acid (18:3co6, GLA) and stearidonic acid (18:4o3, SDA) in their leaf lipids and seed TAGs (Guil-Guerrero et al., 2000). GLA and SDA are recognized as beneficial fatty acids in human nutrition. The first step in the synthesis of LC-PUFA is a A6-desaturation. GLA is synthesized by a A6-desaturase that introduces a double bond into the A6-position of LA. The same enzyme is also able to introduce a double bond into A6-position of ALA, producing SDA. A6-Desaturase genes have been cloned from members of the Boraginacae, like borage (Sayanova et al., 1997) and two Echium species (Garcia-Maroto et al., 2002). Echium plantagineum is a winter annual native to Mediterranean Europe and North Africa. Its seed oil is unusual in that it has a unique ratio of 03 and o6 fatty acids and contains high amounts of GLA (9.2%) and SDA (12.9%) (Guil-Guerrero et al., 2000), suggesting the presence of A6-desaturase activity involved in desaturation of both 3 and o6 fatty acids in seeds of this plant. Cloning of E. platanmineum EplD6Des gene Degenerate primers with built-in XbaI or SacI restriction sites corresponding to N and C-termini amino acid sequences MANAIKKY (SEQ ID NO: 61) and EALNTHG (SEQ ID NO: 62) of known Echium pitardii and Echium gentianoides (Garcia-Maroto et al., 2002) A6-desaturases were used for RT-PCR amplification of A6-desaturase sequences from E. platangineum using a proofreading DNA polymerase Pfu Turbo@ (Stratagene). The 1.35kb PCR amplification product was inserted into pBluescript SK(+) at the XbaI and SacI sites to generate plasmid pXZP106. The nucleotide sequence of the insert was determined (SEQ ID NO:63). It comprised an open reading frame encoding a polypeptide of 438 amino acid residues (SEQ ID NO:64) which had a high degree of homology with other reported A6- and A8-desaturases from E. gentianoides (SEQ ID WO 2005/103253 PCT/AU2005/000571 107 NO:65), E. pitardii (SEQ ID NO:66), Borago officinalis (SEQ ID NO:67 and 68), Helianthus annuus (SEQ ID NO:69) and Arabidopsis thaliana (SEQ ID NO:70 and SEQ ID NO:71) (Figure 10). It has a cytochrome b 5 domain at the N-terminus, including the HPGG (SEQ ID NO:72) motif in the heme-binding region, as reported for other A6- and A8-desaturases (Sayanova et al. 1997; Napier et al. 1999). In addition, the E. plantagineum A6 desaturase contains three conserved histidine boxes, including the third histidine box containing the signature QXXHH (SEQ ID NO:73) motif present in majority of the 'front-end' desaturases (Figure 10) (Napier et al., 1999). Cluster analysis including representative members of A6 and A8 desaturases showed a clear grouping of the cloned gene with other A6 desaturases especially those from Echium species. Heterologous expression of E. plantagineum A6-desaturase gene in yeast Expression experiments in yeast were carried out to confirm that the cloned E. platangineum gene encoded a A6-desaturase enzyme. The gene fragment was inserted as an XbaI-SacI fragment into the SmaI-SacI sites of the yeast expression vector pSOS (Stratagene) containing the constitutive ADH1 promoter, resulting in plasmid pXZP27 1. This was transformed into yeast strain S288Cc by a heat shock method and transformant colonies selected by plating on minimal media plates. For the analysis of enzyme activity, 2mL yeast clonal cultures were grown to an O.D.

600 of 1.0 in yeast minimal medium in the presence of 0.1% NP-40 at 30'C with shaking. Precursor free-fatty acids, either linoleic or linolenic acid as 25mM stocks in ethanol, were added so that the final concentration of fatty acid was 0.5mM. The cultures were transferred to 20*C and grown for 2-3 days with shaking. Yeast cells were harvested by repeated centrifugation and washing first with 0.1% NP-40, then 0.05%NP-40 and finally with water. Fatty acids were extracted and analyzed. The peak identities of fatty acids were confirmed by GC MS. The transgenic yeast cells expressing the Echium EplD6Des were able to convert LA and ALA to GLA and SDA, respectively. Around 2.9% of LA was converted to GLA and 2.3% of ALA was converted to SDA, confirming the A6-desaturase activity encoded by the cloned gene. Functional expression of E. platangineum A6-desaturase gene in transgenic tobacco In order to demonstrate that the EplD6Des gene could confer the synthesis of A6 desaturated fatty acids in transgenic plants, the gene was expressed in tobacco plants. To do this, the gene fragment was excised from pXZP106 as an XbaI-SacI fragment and cloned into the plant expression vector pBI121 (Clonetech) at the XbaI and SacI sites WO 2005/103253 PCT/AU2005/000571 108 under the control of a constitutive 35S CaMV promoter, to generate plant expression plasmid pXZP341. This was introduced into Agrobacterium tumefaciens AGLI, and used for transformation of tobacco W3 8 plant tissue, by selection with kanamycin. Northern blot hybridization analysis of transformed plants was carried out to detect expression of the introduced gene, and total fatty acids present in leaf lipids of wild-type tobacco W38 and transformed tobacco plants were analysed as described above. Untransformed plants contained appreciable amounts of LA (21 % of total fatty acids) and ALA (37% of total fatty acids) in leaf lipids. As expected, neither GLA nor SDA, products of A6-desaturation, were detected in the untransformed leaf. Furthermore, transgenic tobacco plants transformed with the pBIl21 vector had similar leaf fatty acid composition to the untransformed W38 plants. In contrast, leaves of transgenic tobacco plants expressing the EplD6Des gene showed the presence of additional peaks with retention times corresponding to GLA and SDA. The identity of the GLA and SDA peaks were confirmed by GC-MS. Notably, leaf fatty acids of plants expressing the EpiD6Des gene consistently contained approximately a two-fold higher concentration of GLA than SDA even when the total A6-desaturated fatty acids amounted up to 30% of total fatty acids in their leaf lipids (Table 11). TABLE 11. Fatty acid composition in lipid from transgenic tobacco leaves (%). Total A6 Plant 16:0 18:0 18:1 18:2 GLA 18:3 SDA desaturate products W38 21.78 5.50 2.44 21.21 - 37.62 - ET27-1 20.33 1.98 1.25 10.23 10.22 41.10 6.35 16.57 ET27-2 18.03 1.79 1.58 14.42 1.47 53.85 0.48 1.95 ET27-4 19.87 1.90 1.35 7.60 20.68 29.38 9.38 30.07 ET27-5 15.43 2.38 3.24 11.00 0.84 49.60 0.51 1.35 ET27-6 19.85 2.05 1.35 11.12 4.54 50.45 2.19 6.73 ET27-8 19.87 2.86 2.55 11.71 17.02 27.76 7.76 24.78 ET27-11 17.78 3.40 2.24 12.62 1.11 51.56 0.21 1.32 ET27-12 16.84 2.16 1.75 13.49 2.71 50.80 1.15 3.86 WO 2005/103253 PCT/AU2005/000571 109 Northern analysis of multiple independent transgenic tobacco lines showed variable levels of the EpiD6Des transcript which generally correlated with the levels of A6-desaturated products synthesized in the plants. For example, transgenic plant ET27-2 which contained low levels of the EplD6Des transcript synthesised only 1.95% of its total leaf lipids as A6-desaturated fatty acids. On the other hand, transgenic plant ET27-4 contained significantly higher levels of EplD6Des transcript and also had a much higher proportion (30%) of A6-desaturated fatty acids in its leaf lipids. Analysis of the individual tobacco plants showed that, without exception, GLA was present at a higher concentration than SDA even though a higher concentration of ALA than LA was present in untransformed plants. In contrast, expression of EplD6Des in yeast had resulted in approximately equivalent levels of conversion of LA into GLA and ALA into SDA. Echium plantagineum seeds, on the other hand, contain higher levels of SDA than GLA. EplD6Des probably carries out its desaturation in vivo in Echium plantagineum seeds on LA and ALA esterified to phosphatidyl choline (PC) (Jones and Harwood 1980). In the tobacco leaf assay, the enzyme is most likely desaturating LA and ALA esterified to the chloroplast lipid monogalactosyldiacylglyerol (MGDG) (Browse and Slack, 1981). In the yeast assay, free fatty acid precursors LA and ALA added to the medium most likely enter the acyl-CoA pool and are available to be acted upon by EplD6Des in this form. Functional expression of E. piatangineum A6-desaturase gene in transgenic seed To show seed-specific expression of the Echium A6-desaturase gene, the coding region was inserted into the seed-specific expression cassette as follows. An NcoI-SacI fragment including the A6-desaturase coding region was inserted into pXZP6, a pBluescriptSK derivative containing a Nos terminator, resulting in plasmid pXZP157. The SmaI-ApaI fragment containing the coding region and terminator EplD6Des-NosT was cloned into pWVec8-Fpl downstream of the Fp1 prompter, resulting in plasmid pXZP345. The plasmid pXZP345 was used for transforming wild type Arabidopsis plants, ecotype Columbia, and transgenic plants selected by hygromycin B selection. The transgenic plants transformed with this gene were designated "DP" plants. Fatty acid composition analysis of the seed oil from T2 seed from eleven T1 plants transformed with the construct showed the presence of GLA and SDA in all of the lines, with levels of A6-desaturation products reaching to at least 11% (Table 12). This demonstrated the efficient A6-desaturation of LA and ALA in the seed.

WO 2005/103253 PCT/AU2005/000571 110 N cq U) CO It 00 ml 00 O u) L) 0 Cq 00 ~- co N 0) 'IT 't co LO N 6)LO t o 6 L6 L cli 6 c'J j L Cd 1- 00 't Ne Nt m - 0 *. - N N; c 5 6 1: C= C ; C U)) CCD CL) q, N N q q CO N cl Cl) N- NY Vl) LO N C) N 00) NO 0 C') 00 'JQ a~ NC) m O o , It clc' co N O N oCl) 0, ) U ) 6 ' N- '6 i CO C) 6 cc CN Nl N N N N C ') C~ o co N 0M < 0( N q 0 ) o L Ci) cli 6 C0 i- co N- c6 4C 4C - ci CC 00 c N 0) LO C0 0) 0) 0 0 c (P ~ CO 00 co N N 0) CO N- C') N 0) C.0 o6 P, t. c6 o6 ci c6 o6 c c6 cc c'li ENc d - L C EL NC ') CL CL CL C)o00o0 o 0 0 0 0 a~ WO 2005/103253 PCT/AU2005/000571 111 Example 16. Mutagenesis of E. platangineum EpID6Des Gene To determine whether variability could be introduced into the A6-desaturase gene and yet retain desaturase activity, the E. platangineum A6-desaturase cDNA was randomly mutated by PCR using Taq polymerase and EPD6DesF1 and EPD6DesR1 5 primers in the presence of dITP as described by Zhou and Christie (1997). The PCR products were cloned as XbaI-SacI fragments in pBluescript SK(+) at XbaI and SacI sites, and sequences of randomly selected clones determined. Random variants with amino acid residue changes were chosen to clone as XbaI-SacI fragments into pBIl21 and the enzyme activities of proteins expressed from these variants characterized in 10 transgenic tobacco leaves as described above for the wild-type gene. Figure llA represents the activity of the EplD6Des sequence variants when expressed in tobacco plants. The variants could be divided -into two broad classes in terms of their ability to carry out A6-desaturation. Mutations represented as empty diamonds showed substantial reductions in the A6-desaturation activity while mutations 15 denoted as solid diamonds had little or no effect on the activity of the encoded A6 desaturase enzyme. Figure 11B represents the quantitative effect that a selection of mutations in the EplD6Des gene had on the A6-desaturase activity. An L14P mutation in the cytochrome b 5 domain and an S301P mutation between histidine box II and histidine box III of EplD6Des caused substantial reductions in their A6-desaturase 20 activities, resulting in a 3- to 5-fold reduction in total A6-desaturated fatty acids when compared to the wild-type enzyme in W38 plants. Surprisingly, significant activity was retained for each. In contrast, most of the variants examined, as exemplified by the S205N mutation, had no effect on the A6-desaturation activity of EplD6Des gene. 25 Example 17. Comparison of acyl-CoA and acyl-PC substrate dependent desaturases for production of LC-PUFA in cells As described above, the synthesis of LC-PUFA such as EPA and DHA in cells by the conventional A6 desaturation pathway requires the sequential action of PUFA desaturases and elongases, shown schematically in Figure 12 part A. This conventional 30 pathway operates in algae, mosses, fungi, diatoms, nematodes and some freshwater fish (Sayanova and Napier, 2004). The PUFA desaturases from algae, fungi, mosses and worms are selective for desaturation of fatty acids esterified to the sn-2 position of phosphatidylcholine (PC) while the PUFA elongases act on fatty acids in the form of acyl-CoA substrates represented in the acyl-CoA pool in the endoplasmic reticulum 35 (ER), which is physiologically separated from the PC component of the ER. Therefore, sequentially desaturation and elongation reactions on a fatty acid substrate requires that WO 2005/103253 PCT/AU2005/000571 112 the fatty acid is transferred between the acyl-PC and acyl-CoA pools in the ER. This requires acyltransferases that are able to accommodate LC-PUFA substrates. This "substrate switching" requirement may account for the low efficiency observed in earlier reported attempts to re-constitute LC-PUFA biosynthesis (Beaudoin et al., 5 2000, Domergue et al., 2003a). The alternative A8 desaturation pathway (Figure 12 part B) suffers from the same disadvantage of requiring "substrate switching". As described in Example 5, the strategy of using a vertebrate desaturase which was able to desaturate acyl-CoA substrates, provided relatively efficient production of LC-PUFA in plant cells including in seed. In Example 5, the combination of a A5/A6 10 desaturase from zebra fish with a A6 elongase from C. elegans had the advantage that both the desaturase and the elongase enzymes had activity on acyl-CoA substrates in the acyl-CoA pool. This may explain why this strategy was more efficient in the synthesis of LC-PUFA. To provide a direct comparison of the relative efficiencies of using an acyl-CoA substrate-dependent desaturase compared to an acyl-PC substrate 15 dependent desaturase, we conducted the following experiment. This compared the use of the Echium A6 desaturase (Example 15) and the P. salina A5 desaturase (Example 14), both of which are thought to use acyl-PC substrates, with the zebrafish A6/A5 desaturase which uses an acyl-CoA substrate (Example 5). A construct was prepared containing two acyl-PC dependent desaturases, 20 namely the Echium A6 desaturase and P. salina A5 desaturase, in combination with the C. elegans A6 elongase. The Echium A6 desaturase gene on an NcoI-SacI fragment was inserted into pXZP143 (Example 15) resulting in pXZP192. The C. elegans A6 elongase gene (Fpl-CeElo-NosT expression cassette) on the HindIII-ApaI fragment of pCeloPWVec8 (Example 5) was inserted into the StuI-ApaI sites of pXZP147 25 (Example 14) to make pXZP193. The HindIll-ApaI fragment of pXZP193 containing both genes (Fpl-PsD5Des-NosT and Fpl-CeElo-NosT) was inserted into the ApaI-StuI sites of pXZP192, resulting in plasmid pXZP194 containing the three expression cassettes. The XbaI-ApaI fragment from pXZP194 was inserted in a pWvec8 derivative, resulting in pXZP357. 30 The plasmid pXZP357 was used to transform plants of wild-type Arabidopsis ecotype Columbia by Agrobacterium-mediated floral dip method, and six transgenic plants were obtained after hygromycin B (20 mg/L) selection. The transgenic TI plants were designated "DT" plants. The hygromycin resistant transformed plants were transferred into soil and self-fertilised. The T2 seed were harvested and the seed fatty 35 acid composition of two lines, DT1 and DT2, was analysed. The seed fatty acids of DTl and DT2 contained low levels of 18:3o6 and 18:4(o4 (0.9 and 0.8 % of GLA, WO 2005/103253 PCT/AU2005/000571 113 0.3% and 0.1% of SDA, respectively, Table 13). In addition, both DT1 and DT2 seed also contained 0.3% and 0.1% of the 20:40o6 (ARA). However, there was no apparent synthesis of the o3 fatty acid EPA in either of the T2 seed lines, which probably reflected the greater desaturation ability of the Echium A6 desaturase on the co6 5 substrate LA compared to the o3 substrate ALA (Example 15). TABLE 13. Fatty acid composition of seed-oil from T2 seed of DTl and DT2. Fatty acid values are % of total fatty acids. Fatty acid Control DTI DT2 16:0 7.2 6.5 6.5 18:0 2.9 3.6 3.3 18:1o9 20.0 23.2 22.3 18:20o6 27.5 23.6 24.4 18:3(o3 15.1 15.4 16.1 20:0 2.2 2.0 1.9 20:1o9/o11 19.9 19.4 19.5 20:1co7 2.2 3.4 3.0 20:26 0.1 0.0 0.0 22:1o)7 0.0 0.0 0.0 Other minor 2.8 1.5 1.9 Total 100.0 98.6 98.9 New A6-PUFA 18:36 0.0 0.9 0.8 20:3co6 0.0 0.0 0.0 20:4o6 0.0 0.3 0.1 Total 0.0 1.2 0.9 New W3-PUFA 18:40o3 0.0 0.3 0.2 20:43 0.0 0.0 0.0 20:53 0.0 0.0 0.0 Total 0.0 0.3 0.2 Total fatty acids 100.0 100.0 100.0 10 This data was in clear contrast to Example 5, above, where expression of the acyl-CoA dependent desaturase from zebrafish in combination with a A6 elongase resulted in the production of at least 1.1% ARA and 2.3% EPA in T2 seed fatty acids. Thus it would appear that acyl-PC dependent desaturases were less effective than acyl 15 CoA dependent desaturases in driving the synthesis of LC-PUFA in plants cells.

WO 2005/103253 PCT/AU2005/000571 114 Example 18. Expression of LC-PUFA genes in Synechococcus Synechococcus spp. (Bacteria; Cyanobacteria; Chroococcales; Synechococcus species for example Synechococcus elongatus, also known as Synechocystis spp.) are unicellular, photosynthetic, marine or freshwater bacteria in the order cyanobacteria 5 that utilize chlorophyll a in the light-harvesting apparatus. The species include important primary producers in the marine environment. One distinct biochemical feature of Synechococcus is the presence of phycoerythrin, an orange fluorescent compound that can be detected at an excitation wavelength of 540 nm, and which can be used to identify Synechococcus. Members of the marine synechococccus group are 10 closely related at the level of 16s rRNA. They are obligately marine and have elevated growth requirements for Na±, Cl~, Mg 2 +, and Ca 2 +, but can be grown readily in both natural and artificial seawater liquid media as well as on plates (Waterbury et al. 1988). Since they have a rapid heterotrophic or autotrophic growth rate, contain fatty acid precursors such as LA and ALA, and are relatively simple to transform, they are 15 suitable for functional studies involving LC-PUFA synthesis genes, or for production of LC-PUFA in fermenter type production systems. Strains such as Synechococcus sp. strain WH8102, PCC7002 (7002, marine), or PCC7942 (freshwater) can be grown easily and are amenable to biochemical and genetic manipulation (Carr, N.G., and N. H. Mann. 1994. The oceanic cyanobacterial picoplankton, p. 27-48. In D. A. Bryant 20 (ed.), the Molecular biology of cyanobacteria. Kluwer Academic publishers, Boston). For example, Synechococcus has been used as a heterologous expression system for desaturases (Domergue 2003b). Wildtype Synechococcus 7002 Fatty Acid Profile and Growth Rates 25 To show that cyanobacterium Synechococcus 7002 was a suitable host for the transformation of fatty acid synthesis genes and that this expression system could be used to rapidly test the functions and specificities of fatty acid synthesis genes, the growth of the wildtype strain 7002 was first analysed at 22"C, 25'C and 30"C and the resultant fatty acid profiles analysed by gas chromatography for growth at 22*C and 30 30 0 C (Table 14). 35 WO 2005/103253 PCT/AU2005/000571 115 TABLE 14. Synechococcus 7002 wildtype fatty acid profiles at 22 0 C and 30 0 C growth temperatures (% of total fatty acid). Palmit- Lino Temp Myristic Palmitic . Stearic Oleic 18:1 iso Linoleic GLA oleic lenic 22 0 C 0.79 42.5 10.6 0.92 8.4 1.5 7.5 0.54 27.1 30 0 C 0.76 47.1 10.9 0.67 17.0 0.34 20.4 2.9 Growth at 30 0 C was much more rapid than at 22*C, with intermediate rates at 5 25'C (Figure 13). The cells were found to contain both linoleic (LA, 18:2 06) and linolenic (ALA, 18:3 co3) acids which could be used as precursors for LC-PUFA synthesis. Although some of the preferred precursor ALA was produced at the 30'C, higher levels were obtained at 22 0 C. Tests were also carried out to determine whether cells could be grown at 30 0 C, followed by reducing the incubation temperature to 22 0 C 10 after sufficient biomass had been achieved, to see if this would result in a shift to higher production of linolenic acid (Figure 14). In this experiment, levels of ALA obtained were greater than 5%. In further experiments, 25 0 C was used as the preferred temperature for strain 7002, providing adequate growth rates and suitable precursor fatty acid profile. 15 Transformation Strategy Both replicative plasmid vectors and non-replicative homologous recombination vectors have been used previously to transform various cyanobacterial species, including Synechococcus 7002 (Williams and Szalay, 1983; Ikeda et al., 2002; 20 Akiyama et al., 1998a). The recombination vectors may be preferred in certain applications, and have been used to inactivate a gene, rather than create an expression strain. A recombination vector was constructed that was suitable for introduction of one or more fatty acid synthesis genes into the chromosome of Synechococcus strains 25 such as strain 7002. This vector contained the Synechococcus 7002 sul2 gene in a pBluescript plasmid backbone, which provided an ampicillin gene as a selectable marker and allowed bacterial replication in species such as E. coli. The vector was engineered to contain a plac promoter from E. coli fused to a downstream multiple cloning site, with the two elements inserted approximately in the centre of the sul2 30 gene. The sul2 gene in Synechococcus encodes a low affinity sulfate which is not essential under normal growth conditions. Any gene other than sul2, preferably a non essential gene, could have been chosen for incorporation in the recombination vector.

WO 2005/103253 PCT/AU2005/000571 116 The sul2 gene was amplified from Synechococcus 7002 genomic DNA using gene-specific primers, based on the near-identical sequence in strain PCC6803 (Genbank Accession No. NC_000911, nucleotides 2902831 to 2904501) and inserted into the vector pGEM-T. The plac promoter from pBluescript was amplified using the 5 primers 5'-getacgcccggggatcctcgaggctggcgcaacgcaattaatgtga-3' (SEQ ID NO:81) (sense) and 5' cacaggaaacagcttgacatcgattaccggcaattgtacggcggccgctacggatatctcgctcgagctcgcccgggg tagct-3' (SEQ ID NO:82) (antisense), which also introduced a number of restriction sites at the ends of the promoter sequence. The amplified fragment was then digested 10 with SmaI and ligated to the large PvuII fragment of pBluescript including the beta lactamase gene. This intermediate vector was then digested with EcoRV and SacI and ligated to the HpaI to SacI fragment (designated sul2b) of the sul2 gene. The resultant plasmid was digested with BamHI, treated with DNA polymerase I (Klenow fragment) to fill in the ends, and ligated to the SmaI to HpaI fragment (designated sul2a) of the 15 sul2 gene. Excess restriction sites were then removed from this vector by digestion with SacI and SpeI, blunting the ends with T4 DNA polymerase, and religation. Finally, a multiple cloning site was introduced downstream of the plac promoter by digesting the vector with ClaI and NotI, and ligating in a ClaI to NotI fragment from pBluescript, generating the recombination vector which was designated pJRP3.2. 20 Various genes related to LC-PUFA synthesis were adapted by PCR methods to include flanking restriction sites as well as ribosome binding site (RBS) sequences that were suitable for expression in the prokaryote, Synechococcus. For example, the Echium plantagineum A6-desaturase (Example 15) was amplified with the primers 5' AGCACATCGATGAAGGAGATATACCCatggctaatgcaatcaagaa-3' (SEQ ID NO:83) 25 (sense) and 5'-ACGATGCGGCCGCTCAACCATGAGTATTAAGAGCTT-3' (SEQ ID NO:84) (antisense). The amplified product was digested with Clal and NotI and cloned into the Clal to NotI sites of pJRP3.2. A selectable marker gene comprising a chloramphenicol acetyl transferase coding region (CAT) (catB3 gene, Accession No. AAC53634) 30 downstream of a pbsA promoter (psbA-CAT) was inserted into the XhoI site of pJRP3.2, producing the vector pJRP3.3. The selectable marker gene was inserted within the sulB gene to enable easy selection for homologous recombination events after introduction of the recombination vector into Synechococcus. Transformation of Synechococcus 7002 was achieved by mixing vector DNA 35 with cells during the exponential phase of growth, during which DNA uptake occurred, as follows. Approximately 1 g of the recombination vector DNA resuspended in WO 2005/103253 PCT/AU2005/000571 117 100 L of 10 mM Tris-HCl was added to 900gL of mid-log phase cells growing in BG 11 broth. The cells were incubated for 90 min at 30 0 C and light intensity of 20pmol photons.m -2.s~ . 250pL aliquots were then added to 2mL BG-1 1 broth, mixed with 2mL molten agar (1.5%) and poured onto BG-11 agar plates containing 50pg/mL 5 chloramphenicol (Cm) for selection of recombinant cells. The plates were incubated for 10-14 days at the same temperature/light conditions before the Cm-resistant colonies were clearly visible. These colonies were then re-streaked several times onto fresh BG 1 1/Cm50 plates. After several rounds of restreaking on selective plates, liquid medium was inoculated with individual colonies and the cultures incubated at 25C 10 Synechococcus 7002 cells containing the Echium A6-desaturase gene inserted into the sulB gene via the recombination vector and expressed from the plac promoter are shown to produce GLA (18:3 A6,9,12) and SDA (18:4, A6,9,12,15) from endogenous linoleic acid (LA) and linolenic acid (ALA), respectively, as substrates. Episomal vectors can also be used in Synechococcus rather than the 15 integrative/recombinational vectors described above. Synechococcus species have native plasmids that have been adapted for use in transformation, for example pAQ EXI, where a fragment of the native plasmid pAQ1 (Accession No. NC_005025) was fused with an E. coli plasmid to form a shuttle vector with both E. coli and Synechococcus origins of replication (Ikeda et al., 2002; Akiyama et al., 1998b). 20 It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly 25 described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive. All publications discussed above are incorporated herein in their entirety. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a 30 context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each claim of this application.

WO 2005/103253 PCT/AU2005/000571 118 REFERENCES Abbadi, A. et al., (2001) Eur. J. Lipid. Sci. Technol. 103:106-113. Abbadi, A., et al. (2004) Plant Cell 16:2734-2748. 5 Abbott et al., (1998) Science 282:2012-2018. Agaba, M. et al., (2004) Marine Biotechnol (NY) 6:251-261. Akiyama, H. et al. (1998a) DNA Res. 5:327-334. Akiyama, H. et a. (1998b) DNA Res. 5:127-129. Baumlein, H. et al., (1991) Mol. Gen. Genet. 225:459-467. 10 Baumlein, H. et a., (1992) Plant J. 2:233-239. Beaudoin, F. et al., (2000) Proc. Natl. Acad. Sci U.S.A. 97:6421-6426. Berberich, T. et a., (1998) Plant Mol. Biol. 36:297-306. Bolch, C.J. et a., (1999a) J. Phycology 35:339-355. Bolch, C.J. et al., (1999b) J. Phycology 35:356-367. 15 Broun, P. et al., (1998) Plant J. 13:201-210. Brown, M.R. et a., (1997) Aquaculture 151:315-331. Browse, J.A. and Slack, C.R. (1981) FEBS Letters 131:111-114. Chinain, M. et al., (1997) J. Phycology 33:36-43. Cho, H.P. et a., (1999a) J. Biol. Chem. 274:471-477. 20 Cho, H.P. et a., (1999b) J. Biol Chem 274:37335-37339. Clough, S.J. and Bent, A.F. (1998) Plant J. 16:735-43. Coleman, A.W. (1977) Am. J. Bot. 64:361-368. Domergue, F. et a., (2002) Eur. J. Biochem. 269:4105-4113. Domergue, F. et al., (2003a) J. Biol. Chem. 278:35115-35126. 25 Domergue, F. et al., (2003b) Plant Physiol. 131:1648-1660. Drexler, H. et al., (2003) J. Plant Physiol. 160:779-802. Dunstan, G. A. et a., (1994) Phytochemistry 35:155-161. Gallagher, J.C. (1980) J. Phycology 16:464-474. Garcia-Maroto, F. et al., (2002) Lipids 37:417-426. 30 Girke, T. et a., (1998) Plant J. 15:39-48. Guil-Guerrero, J.L. et a., (2000). Phytochemistry 53:451-456. Haseloff, J. and Gerlach, W.L. (1988) Nature 334:585-591. Hastings, N. et a., (2001) Proc. Natl. Acad. Sci. U.S.A. 98:14304-14309. Hong, H. et a!., (2002) Lipids 37:863-868. 35 Hong, H. etal., (2002a) Lipids 37:863-868. Horiguchi, G. et a., (1998) Plant Cell Physiol. 39:540-544.

WO 2005/103253 PCT/AU2005/000571 119 Huang, Y.S. et al., (1999) Lipids 34:649-659. Ikeda, K. et al. (2002) World J. Microbiol. Biotech. 18:55-56. Inagaki, K. et al., (2002) Biosci Biotechnol Biochem 66:613-621. Jones, A.V. and Harwood, J.L (1980) Biochem J. 190:851-854. 5 Kajikawa, M. et al., (2004) Plant Mol Biol 54:335-52. Knutzon, D.S. et al., (1998) J. Biol Chem. 273:29360-6. Lee, M. et al., (1998) Science 280:915-918. Leonard, A.E. et al., (2000) Biochem. J. 347:719-724. Leonard, A.E. et al., (2000b) Biochem. J. 350:765-770. 10 Leonard, A.E. et al., (2002) Lipids 37:733-740. Lo, J. et al., (2003) Genome Res. 13:455-466. Mansour, M.P. et al., (1999a) J. Phycol. 35:710-720. Medlin, L.K. et al., (1996) J. Marine Systems 9:13-3 1. Metz, J.G. et al., (2001) Science 293:290-293. 15 Meyer, A. et al., (2003) Biochemistry 42:9779-9788. Meyer, A. et al., (2004) Lipid Res 45:1899-1909. Michaelson, L.V. et al., (1998a) J. Biol. Chem. 273:19055-19059. Michaelson, L. V. et al., (1998b) FEBS Lett. 439:215-218. Mitchell, A.G. and Martin, C.E. (1995) J. Biol. Chem. 270:29766-29772. 20 Morita, N. et al., (2000) Biochem. Soc. Trans. 28:872-879. Napier, J.A. et al., (1998) Biochem J. 330:611-614. Napier, J.A et al., (1999) Trends in Plant Sci 4:2-4. Napier, J.A. et al., (1999) Curr. Op. Plant Biol. 2:123-127. Needleman, S.B. and Wunsch, C.D. (1970) J. Mol. Biol. 48:443-453. 25 Parker-Barnes, J.F. et al., (2000) Proc Natl Acad Sci USA 97:8284-8289. Pereira, S.L. et al., (2004) Biochem. J. 378:665-671. Perriman, R. et al., (1992) Gene 113:157-163. Qi, B. et al., (2002) FEBS Lett. 510:159-165. Qiu, X. et al., (2001) J. Biol. Chem. 276:31561-31566. 30 Reddy, A.S. et al., (1993) Plant Mol. Biol. 22:293-300. Saito, T. et al., (2000) Eur. J. Biochem. 267:1813-1818. Sakuradani, E. et al., (1999) Gene 238:445-453. Sayanova, O.V. et al., (1997) Proc. Natl. Acad. Sci. U.S.A. 94:4211-4216. Sayanova, O.V. et al., (1999) Plant Physiol. 121:641-646. 35 Sayanova, O.V. et al., (2003) FEBS Lett. 542:100-104. Sayanova, O.V. and Napier, J.A. (2004) Phytochemistry 65:147-158.

WO 2005/103253 PCT/AU2005/000571 120 Shippy, R. et al., (1999) Mol. Biotech. 12:117-129. Simopoulos, A.P. (2000) Poultry Science 79:961-970. Singh, S. et al., (2001) Planta 212:872-879. Smith, N.A. et al., (2000) Nature 407:319-320. 5 Sperling, P. et al., (2000) Eur. J. Biochem. 267:3801-3811. Sperling, P. and Heinz, E. (2001) Eur. J. Lipid Sci. Technol 103:158-180. Sprecher, H. et al., (1995) J. Lipid Res. 36:2471-2477. Spychalla, P.J. et al., (1997) Proc. Natl. Acad. Sci. U.S.A. 94:1142-1147. Stalberg, K. et al., (1993) Plant. Mol. Biol. 23:671-683. 10 Takeyama, H. et al., (1997) Microbiology 143:2725-273 1. Tanaka, M. et al., (1999) Biotechnol. Lett. 21:939-945. Tonon, T. et al., (2003) FEBS Lett. 553:440-444. Trautwein, E.A. (2001) Eur. J. Lipid Sci. Technol. 103:45-55. Tvrdik, P. (2000) J. Cell Biol. 149:707-718. 15 Valvekens, D. et al., (1988) Proc. Natl. Acad. Sci. USA. 85:5536-5540. Volkman, J.K. et al., (1989) J. Exp. Mar. Biol. Ecol. 128:219-240. Wallis, J.G. and Browse, J. (1999) Arch. Biochem. Biophys. 365:307-316. Wang, M.B. et al., (1997) J. Gen. Breed. 51:325-334. Waterbury, J. B. et al. (1988) Methods Enzymol. 167:100-105. 20 Waterhouse, P.M. et al., (1998) Proc. Natl. Acad. Sci. USA 95:13959-13964. Watts, J.L. and Browse, J. (1999b) Arch Biochem Biophys 362:175-182. Williams, J.G. and Szalay, A.A. (1983) Gene 24:37-5 1. Whitney, H.M. et al., (2003). Planta 217:983-992. Yazawa, K. (1996) Lipids 3 1:S297-S300. 25 Yu, R. et al., (2000) Lipids 35:1061-1064. Zank, T.K. et al., (2000) Plant J. 31:255-268. Zank, T.K. et al., (2002) Plant J. 31:255-268. Zhang, Q. et al., (2004) FEBS Lett. 556:81-85. Zhou, X.R. and Christie, P.J. (1997) J Bacteriol 179:5835-5842. 30

Claims (25)

1. A recombinant plant cell which is capable of synthesising a long chain polyunsaturated fatty acid(s) (LC-PUFA) derived from a cell that is not capable of synthesising said LC-PUFA, or a recombinant cell with an enhanced capacity to synthesize an LC-PUFA relative to an isogenic non-recombinant cell, the cell comprising one or more exogenous polynucleotides which encode at least two enzymes each of which is a A5/A6 bifunctional desaturase, A5 desaturase, A6 desaturase, A5/A6 bifunctional elongase, A5 elongase, A6 elongase, A4 desaturase, A9 elongase, or A8 desaturase, wherein the one or more polynucleotides are operably linked to one or more promoters that are capable of directing expression of said polynucleotides in the cell, wherein at least one of the enzymes is a A5 elongase which catalyses the conversion of EPA to DPA in the cell, and wherein the plant cell is homozygous for at least one of the exogenous polynucleotides, and wherein the A5 desaturase catalyses the conversion of ETA to EPA in the cell, and wherein the plant cell comprises DPA which is synthesized from EPA in the cell, and wherein the efficiency of conversion of EPA to DPA in the cell is at least 5%.

2. The cell of claim 1, wherein at least one of the enzymes is a desaturase which is able to act on an acyl-CoA substrate.

3. The cell of claim 2, wherein the cell comprises a A5 desaturase, a A6 desaturase, or both, which is able to act on an acyl-CoA substrate.

4. The cell according to any one of claims 1 to 3, which has one or more of the following features; i) wherein the A5 elongase also has A6 elongase activity, and wherein the elongase is more efficient at synthesizing DPA from EPA than it is at synthesizing ETA from SDA, ii) wherein at least the protein coding region of one, two or more of the polynucleotides is obtained from an algal gene, iii) which comprises at least one C20 LC-PUFA which is synthesized in the cell, and wherein the total fatty acid of the cell comprises at least 2% C20 LC-PUFA, iv) which comprises at least one o3 C20 LC-PUFA which is synthesized in the cell, and wherein the total fatty acid of the cell comprises at least 2.5% 03 C20 LC PUFA, v) which comprises co3 polyunsaturated fatty acids that are the products of A6 desaturation of ALA and/or the products of A9 elongation of ALA, which products 122 are synthesized in the cell, and wherein the efficiency of conversion of ALA to said products in the cell is at least 22%, and vi) which comprises docosahexaenoic acid (DHA).

5. The cell according to any one of claims 1 to 4, wherein the LC-PUFA is docosahexaenoic acid (DHA), and wherein the introduced polynucleotide(s) encode five enzymes wherein the enzymes are any one of the following combinations; i) a A4 desaturase, a A5 desaturase, a A6 desaturase, a A5 elongase and a A6 elongase, or ii) a A4 desaturase, a A5 desaturase, a AS desaturase, a A5 elongase and a A9 elongase.

6. The cell according to any one of claims 1 to 5, wherein the LC-PUFA is docosapentaenoic acid (DPA), and wherein the introduced polynucleotide(s) encode four enzymes wherein the enzymes are any one of the following combinations; i) a A5 desaturase, a A6 desaturase, a AS elongase and a A6 elongase, or ii) a A5 desaturase, a AS desaturase, a A5 elongase and a A9 elongase.

7. The cell according to any one of claims I to 7, wherein i) the total fatty acid of the cell comprises at least 0.1% DPA, and/or ii) the efficiency of conversion of EPA to DPA in the cell is at least 7%.

8. The cell according to any one of claims I to 7, wherein the plant cell from an angiosperm, an oilseed plant cell, or a cell in a seed.

9. The cell according to any one of claims 1 to 8, wherein at least one promoter is a seed specific promoter.

10. The cell according to any one of claims I to 9, wherein at least 50% of the LC PUFA that is produced in said cell is incorporated into triacylglycerols.

11. The cell according to any one of claims 4 to 10, wherein the algal gene is from the genus Pavlova such as from the species Pavlova salina,

12. A transgenic plant comprising at least one cell according to any one of claims 1 to 11.

13. The plant of claim 12, which has one or more of the following features; 123 i) wherein the plant is an angiosperm, ii) which comprises at least one plant part synthesizing EPA, and wherein the total fatty acid of the plant part comprises at least 1.5% EPA, iii) which comprises at least one plant part synthesizing DPA, wherein the total fatty acid of the plant part comprises at least 0.1% DPA, iv) which comprises at least one plant part synthesizing at least one m3 C20 LC-PUFA, and wherein the total fatty acid of the plant part comprises at least 2.5% o3 C20 LC-PUFA v) which comprises at least one plant part synthesizing EPA, and wherein the efficiency of conversion of ALA to EPA in the plant part is at least 2%, vi) which comprises at least one plant part synthesizing m3 polyunsaturated fatty acids that are the products of A6-desaturation of ALA and/or the products of A9 elongation of ALA, and wherein the efficiency of conversion of ALA to said products in the plant part is at least 22%, vii) wherein the efficiency of conversion of EPA to DPA in the plant part is at least 7%, viii) wherein the plant is an oilseed plant, and ix) which comprises at least one plant part which synthesizes DHA.

14. A method of producing an oilseed, the method comprising i) growing a transgenic plant according to claim 12 or claim 13 which is an oilseed under suitable conditions, and ii) harvesting the seed of the plant.

15. Part of the transgenic plant of claim 12 or claim 13, wherein said part comprises an increased level of LC-PUFA in its fatty acid relative to the corresponding part from an isogenic non-transformed plant.

16. The plant part of claim 15 which is a seed, leaf, stem, flower, pollen, roots or specialised storage organ.

17. The plant part of claim 15 or claim 16 which is a seed, wherein the LC-PUFA is selected from the group consisting of: i) DPA, ii) DHA, iii) EPA and DPA, and iv) EPA, DHA, and DPA. 124

18. The plant part of claim 17, which has one or more of the following features; i) which is derived from an isogenic non-transgenic seed which produces LA and/or ALA, and wherein the isogenic non-transgenic seed comprises at least about 13% ALA in its total fatty acids, ii) wherein the total fatty acid in the oil of the seed comprises at least 9% C20 fatty acids, iii) wherein the seed derived from an oilseed plant, iv) wherein the seed has a germination rate which is substantially the same as that of the isogenic non-transgenic seed, v) which comprises EPA which is synthesized in the seed and wherein the total fatty acid of the seed comprises at least 1.5% EPA, vi) which comprises DPA which is synthesized in the seed and wherein the total fatty acid of the seed comprises at least 0.1% DPA, vii) which comprises DHA which is synthesized in the seed, viii) which comprises at least one co3 C20 LC-PUFA which is synthesized in the seed and wherein the total fatty acid of the seed comprises at least 2.5% ot3 C20 LC-PUFA, ix) which comprises EPA which is synthesized in the seed and wherein the efficiency of conversion of ALA to EPA in the seed is at least 2%, x) which comprises o3 polyunsaturated fatty acids that are the products of A6 desaturation of ALA and/or the products of A9 elongation of ALA, which products are synthesized in the seed, and wherein the efficiency of conversion of ALA to said products in the seed is at least 22%, xi) which comprises DPA which is synthesized from EPA in the seed and wherein the efficiency of conversion of EPA to DPA in the seed is at least 5%, and x) wherein at least 25% of the LC-PUFA in the seed form part of triacylglycerols.

19. An extract from the transgenic plant of claim 12 or claim 13, or a part according to any one of claims 15 to 18, wherein said extract comprises an increased level of LC-PUFA in its fatty acid relative to a corresponding extract from an isogenic non-transformed plant.

20. A method of producing a LC-PUFA, the method comprising culturing, under suitable conditions, a recombinant cell according to any one of claims I to 11, or cultivating, under suitable conditions, a transgenic plant of claim 12 or claim 13. 125

21. A method of producing oil comprising at least one LC-PUFA, comprising obtaining the transgenic plant of claim 12 or claim 13, or the plant part according to any one of claims 15 to 18, and extracting oil from said plant or plant part.

22. A composition comprising a cell according to any one of claims 1 to 11, transgenic plant of claim 12 or claim 13, or the plant part according to any one of claims 15 to 18, or an extract or portion of any of these comprising LC-PUFA, and a suitable carrier.

23. A method of increasing the levels of a LC-PUFA in an organism, the method comprising administering to the organism a cell according to any one of claims 1 to 11, a transgenic plant of claims 12 or claim 13, the plant part according to any one of claims 15 to 18, an extract of claim 19, the product of the method of claim 20 or claim 21, or a composition of claim 22.

24. A method of treating or preventing a condition which would benefit from a LC-PUFA, the method comprising administering to a subject a cell according to any one of claims 1 to 11, a transgenic plant of claim 12 or claim 13, the plant part according to any one of claims 15 to 18, an extract of claim 19, the product of the method of claim 20 or claim 21, or a composition of claim 22.

25. Use of a cell according to any one of claims 1 to 11, a transgenic plant of claim 12 or claim 13, the plant part according to any one of claims 15 to 18, an extract of claim 19, the product of the method of claim 20 or claim 21, or a composition of claim 22, for the manufacture of a medicament for treating or preventing a condition which would benefit from a LC-PUFA.