Background of the Invention Cellular membranes are dynamic structures in which variable amounts of proteins are embedded in a lipid bilayer whose hydrophobic characteristics are largely due to fatty acid moities of complex lipids (Singer and Nicolson 1972). The'fluidity'of the membranes are achieved by incorporating unsaturated fatty acyl chains of varying lengths and varying degrees of unsaturation into the lipids (Stubbs and Smith 1984). In animals, some of the unsaturated fatty acids need to be supplie by the diet (essentiel polyunsaturated fatty acids') but, in part, can also be synthesized de novo by oxidative desaturation (i. e. formation of double bonds) of saturated fatty acids of plant and animal origin. Polyunsaturated fatty acid formation requires acetyl-CoA dependent chain elongation and desaturation. Most mammalian tissues can modify acyl chains by introducing more than one double bond with the first one generally at the A-9 position between carbons C-9 and C- 10. Subsequent double bonds may then be inserted at the A-4, A-5, and A-6 positions by individual desaturase activities (Cook 1991).

For the two major precursors of the (n-6) and (n-3) series of polyunsaturated fatty acids, linoleic 18: 2 (n-6) and alpha-linolenic 18: 3 (n-3) acids, animals depend entirely on their dietary intake. By alternating sequences of desaturation (involving

the subsequent action of A4, A5-and A6-desaturases, respectively) and C2 chain elongation, linoleic and alpha-linolenic acids are utilized to form arachidonic acid, 20: 4 (n-6), and the (n-3) acyl chains eicosapentaenoic acid, 20: 5 (n-3), and docosahexaenoic acid, 22: 6 (n-3), respectively (Cook 1991).

Linoleic and arachidonic acid are the only members of the (n-6) family that accumulate in large quantities in liver and most other animal tissues. The intermediates 18: 3 (n-6) and 20: 3 (n-6) are formed from 18: 2 (n-6) by A6- desaturation, chain elongation and A5-desaturation (Horrobin 1993). As a component of phospholipids arachidonic acid is abundant in cellular membranes but also serves as the primary precursor of oxygenated derivatives such as prostaglandine E2 which is pro-inflammatory and regulates cell function of the immune system.

The (n-3) acyl chains eicosapentaenoic acid [20: 5 (n-3)] and docosahexaenoic acid [22: 6 (n-3)] are most abundant in cerebral cortex, retina, and spermatozoa.

Although it is generally assumed that the liver is the major source of 22: 6 (n-3), it has been shown that docosahexaenoic acid can also be produced by retinal pigment epithelium (Wang and Anderson 1993) as well as brain astrocytes (Moore et al. 1991, Delton-Vandenbrouke et al. 1997). In retinal rod outer segments, phospholipids may contain 40-60% of 22: 6 (n-3) which can markedly influence membrane fluidity due to the presence of six double bonds.

In recent years there has been increasing interest in the role of polyunsaturated fatty acids in the pathobiology of a number of chronic conditions such as coronary and peripheral vascular disease (Horrobin 1995), acute and chronic inflammatory immune responses (Calder 1998, Fan and Chapkin 1998, Grimble and Tappia 1998), cutaneous abnormalities (Horrobin 1989, Grattan et al. 1990), essential hypertension (Russo et al. 1997, Chi and Gupta 1998), diabetes mellitus (Mori et al. 1997), asthma (Leichsenring et al. 1995, Villani et al. 1998, Hodge et al. 1998) and rheumatoid arthritis (James and Cleland 1997, Ariza-Ariza et al. 1998, Grimble and Tappia 1998). A particular role has been attributed to gamma- linolenic acid [18: 3 (n-6)] as an anti-cancer polyunsaturated fatty acid. It has been

shown that 18: 3 (n-6) confers anticancer properties by a variety of mechanisms such as (i) up-regulation of E-cadherin, a cell-cell adhesion molecule which acts as a suppressor of metastasis (Jiang et al. 1995), (ii) regulation of desmosome- mediated cell-cell adhesion in human cancer cells (Jiang et al. 1997a), (iii) up- regulation of the metastasis-suppressor gene nm-23 thus contributing to the inhibition of the in vitro invasion of tumor cells (Jiang et al. 1998a), (iv) up- regulation of maspin expression, a mammary serine protase inhibitor, with profond effects on motility of cancer cells (Jiang et al. 1997b) and (v) finally inhibition of cell cycle progression via regulation of phosphorylation and subsequent degradation of cell cycle inhibitors p27kip1 and p57kip2 (Jiang et al.

1998b).

To further understand lipid-related function in human health and disease additional research into fatty acid biosynthesis and metabolism is required. In particular, we need to understand the pharmacological properties, the mechanisms of action and the tissue-specific regulation of composition of the polyunsaturated fatty acids and their metabolites. This will provide additional insight into the role of the polyunsaturated fatty acids in various chronic disease states and will make it feasible to focus pharmacogenomic research on drug design and evaluation with the goal of ameliorating acute health problems associated with impaire lipid function. As a prerequisite, the genes and their gene products involved in the above-mentioned processes need to be identifie and characterized.

It is the objective of the present invention to provide cDNA molecules of three novel members of the human membrane fatty acid desaturase gene family, termed FADS1, FADS2 and FADS3. The three genes share a nucleic acid identity of approximately 50-60% and an amino acid identity of about 77% with each other.

Similar to other membrane-bound desaturases from mammals, fungi, insects, plants and cyanobacteria FADS1, FADS2 and FADS3 reveal a hydropathy profile typical of membrane-bound desaturases and share three regions of highly conserve primary sequence of the general histidine motif HX2 (3) [XH] H (Shanklin et al. 1994). The histidine residues may act as metal-chelating ligands involved in the binding of oxygen in the rection center (Shanklin et al. 1995). Together, these

features confirm FADS1, FADS2 and FADS3 as novel members of the desaturase family of fatty acyl chain-modifying enzymes.

Amino acid identity of FADS1, FADS2 and FADS3 to known desaturases (e. g. from Arabidopsis thaliana, Brassica napus, Synechocystis spec., Borago officinalis, Helianthus annuus, Saccharomyces cerevisiae and Caenorhabditis elegans) is restricted to the respective carboxy terminal regions (amino acid positions 260 to 422) revealing an overall sequence identity of approximately 27%.

Interestingly, the respective amino-termini of the three novel proteins demonstrate similarities to cytochrome b5 (amino acid positions 4 to 75; Fig. 1). Cytochrome b5 is a small hemoprotein and functions as an intermediate donor in a number of oxidation/reduction rections including e. g. the NADH-dependent A9 stearoyl-CoA desaturation (Strittmatter et al. 1974) or the A5 desaturation in cholesterol biosynthesis (Reddy et al. 1977). From the amino acid alignments we conclue that FADS1, FADS2 and FADS3 are fusion proteins consisting of a N-terminal cytochrome b5 and a C-terminal desaturase-like enzyme. From a functional point of view, this fusion of two activities may increase the efficiency of electron transport required for desaturation by covalently bringing together the presumed electron donor (cytochrome b5) and its putative acceptor (desaturase-like enzyme). Other heme fusion proteins containing the cytochrome b5 domain have been identifie and represent a superfamily of fused proteins (Guiard and Lederer 1979). Besides others this superfamily inclues the yeast flavocytochrome b2, sulfite oxidase, nitrate reductase, the yeast A9 acyl-CoA desaturase and more recently the sunflower cytochrome b5-desaturase fusion protein (Sperling et al.

1995). The three novel desaturase-like enzymes reporte herein, FADS1, FADS2 and FADS3, can be added to the growing list of members of this superfamily of fused proteins (Fig. 2).

Summary of the invention The eukaryotic fatty acid desaturases represent a group of iron-containing enzymes that catalyze NAD (P) H- and 02-dependent introduction of double bonds into fatty acyl chains. Impairment of desaturase activities has been implicated in a variety of human conditions including liver disease, coronary artery disease and

cancer. With the present invention we are providing three isolated human cDNA molecules that encode three novel members of a cytochrome-b5-containing fusion protein with similarity to plant and lower animal desaturase enzymes, termed fatty acid desaturase-1 (FADS1) (represented by Fig. 3 and SEQ ID NO. 1), fatty acid desaturase-2 (FADS2) (represented by Fig. 4 and SEQ ID NO. 2) and fatty acid desaturase-3 (FADS3) (represented by Fig. 5 and SEQ ID NO. 3).

FADS1 protein MAPDPVAAETAAQGPTPRYFTWDEVAQRSGCEERWLVI DRKVYN ISEFTRRH P GGSRVISHYAGQDATDPFVAFHINKGLVKKYMNSLLIGELSPEQPSFEPTKNKEL <BR> <BR> <BR> <BR> TDEFRELRATVERMGLM KAN HVFFLLYLLH I LLLDGAAWLTLWVFGTSFLPFLLCA<BR> <BR> <BR> <BR> <BR> VLLSAVQAQAGWLQHDFGHLSVFSTSKWNHLLHHFVIGHLKGAPASWWNHMHF QHHAKPNCFRKDPDINMHPFFFALGKILSVELGKQKKKYMPYNHQHKYFFLIGPP ALLPLYFQWYIFYFVIQRKKWVDLAWMITFYVRFFLTYVPLLGLKAFLGLFFIVRFL ESNWFVWVTQMNHIPMHIDHDRNMDWVSTQLQATCNVHKSAFNDWFSGHLNF <BR> <BR> <BR> <BR> QIEHHLFPTMPRHNYHKVAPLVQSLCAKHGIEYQSKPLLSAFADIIHSLKESGQLW LDAYLHQ FADS2 protein <BR> <BR> <BR> <BR> MGKGGNQGEGAAEREVSVPTFSWEEIQKHNLRTDRWLVIDRKVYNITKWSIQHP GGQRVIGHYAGEDATDAFRAFHPDLEFVGKFLKPLLIGELAPEEPSQDHGKNSKI <BR> <BR> <BR> <BR> TEDFRALRKTAEDMNLFKTNHVFFLLLLAHIIALESIAWFTVFYFGNGWIPTLITAFV&l t;BR> <BR> <BR> <BR> <BR> <BR> LATSQAQAGWLQHDYGHLSVYRKPKWNHLVHKFVIGHLKGASANWWNHRHFQ HHAKPNIFHKDPDVNMLHVFVLGEWQPIEYGKKKLKYLPYNHQHEYFFLIGPPLLI <BR> <BR> <BR> <BR> PMYFQYQIIMTMIVHKNWVDLAWAVSYYIRFFITYIPFYGILGALLFLNFIRFLESHW FVWVTQMNHIVMEIDQEAYRDWFSSQLTATCNVEQSFFNDWFSGHLNFQIEHHL FPTMPRHNLHKIAPLVKSLCAKHGIEYQEKPLLRALLDIIRSLKKSGKLWLDAYLHK FADS3 protein MGGVGEPGPREGPAQPGAPLPTFCWEQIRAHDQPGDKWLVIERRVYDISRWA QRHPGGSRLIGHHGAEDATDAFRAFHQDLNFVRKFLQPLLIGELAPEEPSQDGP LNAQLVEDFRALHQAAEDMKLFDASPTFFAFLLGHILAMEVLAWLLIYLLGPGWV <BR> <BR> <BR> PSALAAFILAISQAQSWCLQHDLGHASIFKKSWWNHVAQKFVMGQLKGFSAHW

WNFRHFQHHAKPNIFHKDPDVTVAPVFLLGESSVEYGKKKRRYLPYNQQHLYFF <BR> <BR> <BR> <BR> LIGPPLLTLVNFEVENLAYMLVCMQWADLLWAASFYARFFLSYLPFYGVPGVLLF FVAVRVLESHWFVWITQMNHIPKEIGHEKHRDWVSSQLAATCNVEPSLFTNWFS GHLNFQIEHHLFPRMPRHNYSRVAPLVKSLCAKHGLSYEVKPFLTALVDIVRSLK KSGDIWLDAYLHQ Studies to clarify the specificity and the subcellular location of these ubiquitiously expressed fusion proteins are in progress. Also, the detailed cellular functions and dysfunctions of the desaturase-like domains are being investigated in appropriate cellular and animal systems. This will address the question whether and to which extent these novel enzymes are involved in human disease. The invention encompasses the three cDNA molecules, FADS1, FADS2, and FADS3, the nucleotide sequence of these cDNAs, and the putative amino acid sequences of the FADS1 (represented by Fig. 6 and SEQ ID NO. 4), FADS2 (represented by Fig. 7 and SEQ ID NO. 5), and FADS3 represented by Fig. 8 and SEQ ID NO. 6) proteins.

Also comprehended by this invention are oligonucleotide primers comprising the cDNA molecule or its complementary strand allowing the amplification of FADS1 (represented by Fig. 9 and SEQ ID NOS. 7-12), FADS2 (represented by Fig. 9 and SEQ ID NOS. 13-18), and FADS3 (represented by Fig. 9 and SEQ ID NOS. 19- 22), by the reverse transcriptase polymerase chain rection (RT-PCR). Such primers are particularly useful and will provide researchers and physicians with an enhanced ability to assess the role of FADS1, FADS2, and FADS3 in human disease. The present invention also relates to methods of screening for and detection of FADS1, FADS2, and FADS3 mutation carriers including prenatal FADS1, FADS2, and FADS3 screening and diagnosis.

Having provided the isolated human FADS1, FADS2, and FADS3 cDNA sequences, also comprehended by this invention are the FADS1, FADS2, and FADS3 proteins, and derivatives thereof, in aspects of diagnosis and treatment of human disease. Finally, the invention pertains to proteins which comprise the same or substantially the same amino acid sequence (at least 200 amino acids) as

that represented by Figs. 6,7,8 and SEQ ID NOS. 4,5,6 or a variant of the amino acid sequences having a deletion, addition or substitution of 1 to 10 amino acids, or its salt.

Another aspect of the invention is the use of the FADS1, FADS2, and FADS3 proteins as a target for drug and gene therapy in the treatment of human disease.

This inclues the generation and utilization of FADS1, FADS2, and FADS3- targeted animal models (knock-in, knock-out, transgenic animals) and anti-FADS1, -FADS2, and-FADS3 antibodies that specifically detect the FADS1, FADS2, and FADS3 proteins, respectively.

The foregoing and other features and avantages of the invention will become more apparent from the following detailed description and accompanying drawings.

One aspect of the invention are the isolated cDNAs selected from the group consisting of: (a) a polynucleotide having at least a 65 % homology, preferably at least a 80 % homology with a polynucleotide encoding a polypeptide selected from the group consisting of the polypeptides of SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6; (b) a polynucleotide having at least a 65 % homology, preferably at least a 80 % homology with a polynucleotide which by virtue of the redundancy of the genetic code, encodes the same polypeptide selected from the group consisting of the polypeptides of SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6; (c) a DNA molecule capable of hybridization under stringent conditions to a DNA molecule according to (a) or (b); (d) a polynucleotide which is complementary to the polynucleotide of (a), (b) or (c); and (e) a oligonucleotide comprising at least 15 consecutive nucleotides of the polynucleotide of (a), (b), (c) or (d)

(including DNAs which are synonymous to the DNAs of (a), (b), (c), (d) and (e) due to the degeneracy of the genetic code) especially isolated cDNAs selected from the group consisting of: (a) a polynucleotide having at least a 65 % homology, preferably at least a 80 % homology with a polynucleotide sequence selected from the group consisting of the polynucleotides of SEQ ID NO: 1, SEQ ID NO: 2 and SEQ ID NO: 3; (b) a DNA molecule capable of hybridization under stringent conditions to a DNA molecule according to (a); (c) a polynucleotide which is complementary to the polynucleotide of (a) or (b); (d) a oligonucleotide comprising at least 15 consecutive nucleotides of the polynucleotide of (a), (b) or (c); and (e) a DNA which is synonymous to the DNAs of (a), (b), (c) or (d) due to the degeneracy of the genetic code.

In the scope of the invention are polynucleotides having a polynucleotide encoding a polypeptide selected from the group consisting of the polypeptides of SEQ ID NO: 1, SEQ ID NO: 2 and SEQ ID NO: 3 and polynucleotides having a polynucleotide sequence selected from the group consisting of the polynucleotides of SEQ ID NO: 4, SEQ ID NO: 5 and SEQ ID NO: 6, but DNAs comprising a nucleotide sequence with at least a 65 % homology with these nucleotide sequences is also within the scope of the invention.

Furthermore within the scope of the invention are : A recombinant vector comprising the disclosed DNA molecules.

Transgenic host cells such as COS7, fibroblast cell lines or any other tissue- specific cell lines, as well as a transgenetic host cell tranformed by the DNA or the vector, a corresponding transgenetic organism or a corresponding transgenetic knock-in or knock-out animal model.

Polypeptides and corresponding proteins comprising at least 65 %, preferably 85 %, especially 100 % of a polypeptide sequence selected from the group consisting of the polypeptides of SEQ ID NO: 1, SEQ ID NO: 2 and SEQ ID NO: 3; polypeptides comprising a polypeptide sequence with at least a 65 % homology with the said polypeptides; peptides comprising at least 15, preferably 30, especially 60 consecutive amino acids of the said polypeptides; and polypeptides having substantially the same amino acid sequence as the said polypeptides, or having a variant of the amino acid sequence of the polypeptides with a deletion, addition or substitution of 1 to 10 amino acids. The salts of the peptides and proteins are also within the scope of the invention.

A process for preparing the proteins which comprises cultivating the transformants to form the proteins.

A method of screening for modulators in well known assays using constructs such as FADS1, FADS2, and FADS3 promoter luciferase or green fluorescent protein hybrids or screening for interacting proteins or factors using state of the art technologies like the interaction trap technology to screen for interacting substances of FADS1, FADS2, and FADS3 or isolated domains of FADS1, FADS2, and FADS3.

A method of screening chemical libraries comprising transformed cell lines A compound which alters/reacts with at least one epitope of the proteins and which is obtained by screening methods utilizing the FADS1, FADS2, and FADS3 cDNAs or protein molecules.

Use of antibodies against the FADS1, FADS2, and FADS3 proteins for diagnostic or therapeutic purposes.

A pharmaceutical composition comprising as an effective component of the proteins or a partial peptide of the proteins, and a pharmaceutically acceptable carrier or diluent.

The term"knock-out animal"as used herein is intended to describe an animal containing a gene which has been modifie by homologous recombination. The homologous recombination event may completely disrupt the gene such that a functional gene product can no longer be produced (hence the name"knock-out") or the homologous recombination event may modify the gene such that an altered, although still functional, gene product is produced.

The term"knock-in"as used herein is intended to describe a variation of gene targeting that uses homologous recombination but allows expression of added genetic sequences in place of the endogenous gene. This approach allows the test of more subtle mutations than is allowed by a simple knock-out.

The term"epitope"describes a region on a macromolecule which is recognized by an antibody. Frequently it is in a short region of primary sequence in a protein and it is generally about 5 to 12 amino acids long (the size of the antigen binding site on an antibody). Carbohydrates, nucleic acids and other macromolecules may be antigens and have epitopes.

Detailed Description of the Invention Materials and Methods Isolation of the FADS1 and FADS2 cDNAs cDNA fragments corresponding to FADS1 and FADS2 were identifie by direct cDNA selection. The cDNA selection was performed essentially as described (Rommens et al. 1993) with only minor modifications. Briefly, total RNA was prepared from human retina and from established human retinal pigment epithelium cell line ARPE-19 (Dunn et al. 1996). Prior to the use as templates for cDNA synthesis the isolated RNAs were separated on a 1.2% agarose gel in the presence of 3- (N-morpholino) propanesulfonic acid (MOPS) and formaldehyde to check their integrity (Sambrook et al., 1989).

RNAs were reverse transcribed using the SUPERSCRIPT preamplification system for first strand cDNA synthesis (Gibco, BRL) and the RXGT12 oligonucleotide primer (5'-CGG AAT TCT CGA GAT CTT TTT TTT TTT TT-3').

After poly (A)-tailing with terminal transferase (United States Biochemical, USB), a cDNA pool was generated by RXGT12-primed PCR at 94°C for 1 min; 2 cycles of 94°C, 30 sec; 37°C, 1 min, 72°C, 2 min followed by 22 cycles of 94°C, 30 sec; 58°C, 30 sec and 72°C, 2 min. Prior to hybridization the cDNA pools were pre- annealed to Cot-1 DNA (Gibco, BRL) enriched with sonicated LINE1 sequences.

Genomic PAC clones for cDNA selection were derived from 11q12-q13.1, a region known to contain the gene underlying Best's vitelliform macular dystrophy (Stöhr et al. 1998). The assembly and orientation of the clones have been described previously (Cooper et al. 1997). Inserts from PAC clones dJ465G21 and dJ139E20 (~1, ug) were isolated by Notl digestion, purifie using QIAEXII agarose gel extraction beads (Qiagen) and immobilized on Hybond-N+ membrane filters with an average concentration of 60 ng/mm2. The insert filters were subjected to two consecutive rounds of hybridization with a starting mixture of 20 zig of retina and ARPE-19 derived cDNAs. Hybridization time was four days at 58°C in Church hybridization buffer (Church and Gilbert 1984). Filters were washed three times in 2 x SSC/0.1 % SDS at room temperature, once each in 0.5 x SSC/0.1 % SDS, 0.2 x SSC/0.1 % SDS and 0.2 x SSC/0.05% SDS (all at 58°C). A final wash was in 2 x SSC. cDNAs were eluted in distille H20 by incubating for 10 min at 98°C and reamplified by PCR using the RXGT12 oligonucleotide primer. Fourg of the reamplified cDNAs were used for a second round of hybridization. After two rounds of selection the cDNAs were amplifie using the RXGT12 oligonucleotide primer, digeste with EcoRl and cloned into the EcoRl site of pBluescript (Stratagene).

The selected cDNAs represent segments of the 3'-untranslated region (3'-UTR) of FADS1 (clone IVC4 at FADS1 nucleotide position 3793-4204; clone IVB7 at nucleotide position 3132-3609; clone VIIC6 at nucleotide position 2077-2317) (Fig.

3) and of the 3'UTR/coding sequence of FADS2 (clone IVB8 at FADS2 nucleotide position 2626-3009; clone TUK8-4B at nucleotide position 753-1508) (Fig. 4).

Using the selected clone sequences extensive dbEST database searches were conducted and revealed a large number of additional overlapping expressed sequence tags (ESTs). More than 100 ESTs (e. g. zk09hO8, EST1 77650, yb28cO3, ym29bO5, yx67hO5) were assemble to an overlapping EST contig representing FADS1. The assemble EST sequences contain an open reading frame (ORF) of 1410 bp, with a first potential in-frame translation initiation codon, ATG, starting 79 nucleotides downstream the most 5'end of EST clone zk09hO8. r1 (GenBank acc. no. AA029030) (Fig. 1a). A consensus polyadenylation signal, AAUAAA, was identifie at nucleotide position 4.182. The mature protein predicted from the ORF consists of 444 amino acid residues resulting in a calculated molecular mass of 52.0 kDa (Fig. 6).

Another 30 overlapping ESTs (e. g. cp2485. seq, HSC2EA121, EST06759, ym42cO4, nc08cO5) were found facilitating the assembly of the FADS2 cDNA. The assemble EST sequences contain an open reading frame (ORF) of 1352 bp, with a first potential in-frame translation initiation codon, ATG, starting 21 nucleotides downstream the most 5'end of EST clone ub64eO1. r1 (GenBank acc. no.

A1036465) (Fig. 4). Consensus polyadenylation signals were predicted at nucleotide positions 2.996 and 4.056. The mature FADS2 protein predicted from the ORF consists of 444 amino acid residues resulting in a calculated molecular mass of 52.3 kDa (Fig. 7). Amino acid sequence identity between FADS1 and FADS2 is 62%.

Isolation of the FADS3 cDNA Additional 30 human EST clones were available to assemble a third individual cDNA, termed FADS3 (e. g. zs84eO6, zs84eO5, nq23fO5, ya49a19, zs86hO9). The existence of a third member of the FADS family was confirme by PCR mapping of FADS1-, FADS2-and FADS3-specific 3'-UTR fragments revealing three distinct gene loci within a 1.4 Mb PAC contig in 11q12-q13.1 (Cooper et al., 1997). The assemble EST sequences contain an open reading frame (ORF) of 1468 bp, with a first potential in-frame translation initiation codon, ATG, starting 134 nucleotides downstream the most 5'end of EST clone qa99dO6. s1 (GenBank acc. no.

Au123992) (Fig. 5). The mature protein predicted from the ORF consists of 445

amino acid residues resulting in a calculated molecular mass of 51.2 kDa (Fig. 8).

The 3'-UTR of the FADS3 cDNA is represented by several EST clones (e. g. zs86hO9. sl, AA279632). A potential polyadenylation signal, AUUAAA, is present at cDNA nucleotide position 1.757 and may be functional as AUUAAA is the most common natural variant of the consensus polyadenylation signal AAUAAA (Fig. 5) (Sheets et al., 1990).

Amino acid sequence identities between FADS1 and FADS3 as well as between FADS2 and FADS3 are 52% and 63%, respectively. All EST sequences in the dbEST databases could be aligned to one of the three cDNAs, FADS1, FADS2, and FADS3. This suggests that there are no additional members of the FADS family in the human genome.

Northern blot analysis Northern blot analysis was performed either with total RNA isolated using the guanidinium thiocyanate method (Chomczynski and Sacchi 1987) or with commercially available multiple tissue Northern (MTN) blots purchased from Clontech Laboratories Inc. (Palo Alto, CA). Each lane of the total RNA blot contained 12 pg of total RNA from lung, cerebellum, uterus, retina, liver, heart, RPE cell line ARPE-19, RPE tissue, lymphocytes and was electrophoretically separated in the presence of formaldehyde. The MTN blots were prepared from poly (A) + RNA isolated from human heart, brain, placenta, lung, liver, skeletal muscle, kidney and pancreas. Inserts of clones IVC4, IVB7 (FADS1), IVB8 (FADS2) and of the 362 bp PCR product F3/R (5'-ACAGCTTTCCCCCAATTCTC- 3'/5'-GGCCTCAGCTACGAAGTGAAG-3') (FADS3) derived from the 3'-UTRs of the respective genes were used for filter hybridization at 65°C in 0.5 mM sodium phosphate buffer, pH 7.2; 7% SDS, 1 mM EDTA at 65°C (Church and Gilbert 1984).

The three genes are ubiquitiously expressed and appear to have similar expression levels in all tissues analyzed. FADS1 revealed a transcrit size of 4.0 kb while FADS2 revealed a similar sized transcrit of 4.0 kb in addition to a smaller transcrit of approximately 3.1 kb. The two FADS2 variants may be due to

differential usage of polyadenylation signals (see above). Finally, FADS3 is represented by two transcripts of 1.75 kb and 1.25 kb in size. While the former is in agreement with the usage of the variant polyadenylation signal identifie at position 1738 of the cDNA, the small size of the latter transcrit can not be explained at present and it does not appear to be due to a differential usage of polyadenylation signals. Possibly, differential splicing and/or exon skipping may be involved in the generation of the variant transcript. However, there is no evidence from cDNA cloning or EST contig assembly to support this possibility.

Comparison with other desaturases Local sequence alignments of the deduced amino acid sequences of FADS1, FADS2, and FADS3 with known proteins or protein motifs were done using SwissProt (http ://wWw. nCbi. nim. nih. gov/cgi-bin/Blast/nph-blast? Jform=O) and the BLASTP and BEAUTY programs at Baylor College of Medicine (http ://dot. imgen. bcm. tmc. edu: 9331/seq-search/protein-search. html). Amino acid sequence alignments were performed using the CLUSTALW multiple alignment program at http ://pbil. ibcp. fr/NPSA/npsaclustalw. html. Phylogenetic tree assembly was done using the TREECON software Version 1.3b available at http ://bioc-www. uia. ac. be/u/yvdp/index. html.

Overall amino acid identities to known desaturases were found to be in the range of 22%-27% (Fig. 1). Phylogenetic tree construction revealed a genetic relationship of FADS1, FADS2, and FADS3 to the A5-, A6-and A8-desaturases with some distance to the A9-desaturases (Fig. 2). From these analyses it becomes obvious that sequence identity by itself is not a predictor of a specific desaturase activity. For example, A5-and A6-desaturases from C. elegans demonstrate a higher sequence identity to each other than to the A6-desaturases from other species. We therefore conclue that based on simple sequence comparisons it is not feasible to determine the specific functions of FADS1, FADS2, and FADS3. This will be done by transgene expression of the three desaturases combine with gas chromatograpy-mass spectometry.

Hydropathy plots of FADS1, FADS2, and FADS3 indicate two hydrophiobic sequences predicted to represent transmembrane-spanning domains similar to other desaturases identifie thus far (Fig. 1) (reviewed in Sperling et al. 1995). cDNA amplification of FADS1, FADS2, and FADS3 The coding sequences of the three genes are amplifie in overlapping fragments by performing RT-PCR using oligonucleotide primer pairs derived from the respective cDNA sequences: (1) FADS1 (Fig. 9 and SEQ ID NOS. 7-12) Sense primer TU12-R5 (5'-CGCCTGACAGCCCCTGCT-3') at cDNA position 31- 48 in combination with antisense primer TU12-F10 (5'- CAGGTGGCCAATCACAAAAT-3') at cDNA position 671-690 results in a product of 660 bp; sense primer TU 12-R7 (5'-CTCAAAGTGGAACCATCTGCTA-3') at cDNA position 645-666 in combination with antisense primer TU12-F9 (5'- GGAAACCCAGTCCATGTTCC-3') at cDNA position 1130-1149 results in a product of 505 bp; sense primer TU12-R6 (5'-CCTGGGCCTTTTCTTCATAGT-3') at cDNA position 1035-1055 in combination with antisense primer TU12-F5 (5'- CTCAAGCTCCCCTCTGCCT-3') at cDNA position 1465-1483 results in a product of 449 bp.

(2) FADS2 (Fig. 9 and SEQ ID NOS. 13-18) Sense primer TU1 3-R4 (5'-TCAGMGCATMCCTGCGC-3') at cDNA position 98- 116 in combination with antisense primer TU13-F7 (5'- CCAGTTCACCAATCAGCAGG-3') at cDNA position 284-303 results in a product of 206 bp; sense primer TU13-R3 (5'-CCCCTGCTGATTGGTGMCT-3') at cDNA position 282-301 in combination with antisense primer TU1 3-F4 (5'- TGTAGGGCAGGTATTTCAGC-3') at cDNA position 779-798 results in a product of 517 bp; sense primer TU13-R2 (5'-AGCCCATCGAGTACGGCAA-3') at cDNA position 754-772 in combination with antisense primer TU1 3-F1 (5'- CCTCAGAACAAAAGCCCATC-3') at cDNA position 1416-1435 results in a product of 682 bp.

(3) FADS3 (Fig. 9 and SEQ ID NOS. 19-22) Sense primer TU19-R2 (5'-TCTTGCTCGGACCTCGGC-3') at LLCDL3 cDNA position 81-98 in combination with antisense primer TU19-F2 (5'- GTGATCCACACGAACCAGTG-3') at cDNA position 1130-1149 position results in a product of 1069 bp; sense primer TU19-R3 (5'-GMGMCCCAGCCAGGATG-3') at cDNA position 428-446 in combination with antisense primer TU19-F3 (5'- ACAGCTTTCCCCCAATTCTC-3') at cDNA position 1709-1728 results in a product of 1301 bp.

Short description of Figures Fig. 1 Comparison of putative amino acid sequences from FADS1, FADS2, FADS3, Borago officinalis, Helianthus annuus and human cytochrome b5.

Arrowheads indicate eight invariant amino acid residues typical for the cytochrome b5 domain. Two potential transmembrane domains are boxed. Three histidine motifs HX2 (3) [XH] H that are conserve within the desaturase family are hatched.

Fig. 2 Phylogenetic tree of fatty acid desaturases.

Fig. 3 (SEQ ID NO. 1) shows the nucleotide sequence of the FADS1 cDNA Fig. 4 (SEQ ID NO. 2) shows the nucleotide sequence of the FADS2 cDNA Fig. 5 (SEQ ID NO. 3) shows the nucleotide sequence of the FADS33 cDNA Fig. 6 (SEQ ID NO. 4) shows the putative amino acid sequence of the predicted FADS1 protein Fig. 7 (SEQ ID NO. 5) shows the putative amino acid sequence of the predicted FADS2 protein Fig. 8 (SEQ ID NO. 6) shows the putative amino acid sequence of the predicted FADS3 protein Fig. 9 (SEQ ID NOS. 7-22) shows the oligonucleotide PCR primers utilized to amplify the FADS1, FADS2, FADS3 cDNA, respectively.

References Ariza-Ariza R, Mestanza-Peralta M, Cardiel MH. Omega-3 fatty acids in rheumatoid arthritis: an overview. Semin Arthritis Rheum 27: 366-370 (1998) Calder PC. Immunoregulatory and anti-inflammatory effects of n-3 polyunsaturated fatty acids. Braz J Med Biol Res 31: 467-490 (1998) Chi Y, Gupta RK. Alterations in membrane fatty acid unsaturation and chain length in hypertension as observe by 1 H NMR spectroscopy. Am J Hypertens 11: 340- 348 (1998) Chomczynski, P. and Sacchi, N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156- 159 (1987) Church, G. M., and W. Gilbert. Genomic sequencing. Proc Natl Acad Sci USA 81: 1991-1995 (1984) Cook HW. Fatty acid desaturation and chain elongation in eucaryotes. In: Biochemistry of lipids, lipoproteins and membranes, Vance DE & Vance JE (eds), Elsevier Amsterdam London, New York, pp. 141-169 (1991) Cooper P, Nowak NJ, Higgins MJ, Simpson SA, Stöhr H, Marquardt A, Weber BHF, Gerhard DS, deJong P, Shows TB. A sequence ready high resolution physical map of the Best's macular dystrophy gene region in 11q12-q13.

Genomics 41: 185-192 (1997) Delton-Vandenbrouke I, Grammas P, Anderson RE. Polyunsaturated fatty acid metabolism in retinal and cerebral microvascular endothelial cells. J Lipid Res 38: 147-159 (1997) Dunn KC, Aotaki-Keen AE, Putkey FR, Hjelmeland LM. ARPE-19, a human retinal pigment epithelial cell line with differentiated properties. Exp Eye Res 62: 155- 169 (1996) Fan YY, Chapkin RS. Importance of dietary gamma-linolenic acid in human health and nutrition. J Nutr 128: 1411-1414 (1998) Grattan C, Burton JL, Manku M, Stewart C, Horrobin DF. Essential-fatty-acid metabolites in plasma phospholipids in patients with ichthyosis vulgaris, acne vulgaris and psoriasis. Clin Exp Dermatol 15: 174-176 (1990)

Grimble RF, Tappia PS. Modulation of pro-inflammatory cytokine biology by unsaturated fatty acids. Z Ernährungswiss 37 (Suppl 1): 57-65 (1998) Guiard B, Lederer F. The"cytochrome b5 fold" : structure of a novel protein superfamily. J Mol Biol 135: 639-50 (1979) Hodge L, Salome CM, Hughes JM, Liu-Brennan D, Rimmer J, Allman M, Pang D, Armour C, Woolcock AJ. Effect of dietary intake of omega-3 and omega-6 fatty acids on severity of asthma in children. Eur Respir J 11: 361-536 (1998) Horrobin DF. Anormal membrane concentrations of 20 and 22-carbon essential fatty acids: a common link between risk factors and coronary and peripheral vascular disease? Prostaglandins Leukot Essent Fatty Acids 53: 385-396 (1995) Horrobin DF. Essential fatty acids in clinical dermatology. J Am Acad Dermatol 20: 1045-1053 (1989) Horrobin DF. Fatty acid metabolism in health and disease: the role of A6- desaturase. Am J Clin Nutr 57 (5 Suppl): 732S-737S (1993) James MJ, Cleland LG. Dietary n-3 fatty acids and therapy for rheumatoid arthritis.

Semin Arthritis Rheum 27: 85-97 (1997) Jiang WG, Bryce RP, Horrobin DF, Mansel RE. gamma-Linolenic acid blocks cell cycle progression by regulating phosphorylation of p27kip1 and p57kip2 and their interactions with other cycle regulators in cancer cells. Int J Oncol 13: 611-617 (1998b) Jiang WG, Hiscox S, Bryce RP, Horrobin DF, Mansel RE. The effects of n-6 polyunsaturated fatty acids on the expression of nm-23 in human cancer cells. Br J Cancer 77: 731-738 (1998a) Jiang WG, Hiscox S, Hallett MB, Horrobin DF, Mansel RE, Puntis MC. Regulation of the expression of E-cadherin on human cancer cells by gamma-linolenic acid (GLA). Cancer Res 55: 5043-5048 (1995) Jiang WG, Hiscox S, Horrobin DF, Bryce RP, Mansel RE. Gamma linolenic acid regulates expression of maspin and the motility of cancer cells. Biochem Biophys Res Commun 237: 639-644 (1997b) Jiang WG, Singhrao SK, Hiscox S, Hallett MB, Bryce RP, Horrobin DF, Puntis MC, Mansel RE. Regulation of desmosomal cell adhesion in human tumor cells by polyunsaturated fatty acids. Clin Exp Metastasis 15: 593-602 (1997a)

Leichsenring M, Kochsiek U, Paul K. (n-6)-Fatty acids in plasma lipids of children with atopic bronchial asthma. Pediatr Allergy Immunol 6: 209-212 (1995) Moore SA, Yoder E, Murphy S, Dutton GR, Spector AA. Astrocytes, not neurons, produce docosahexaenoic acid (22: 6 omega-3) and arachidonic acid (20: 4 omega-6). J Neurochem 56: 518-524 (1991) Mori Y, Murakawa Y, Katoh S, Hata S, Yokoyama J, Tajima N, Ikeda Y, Nobukata H, Ishikawa T, Shibutani Y. Influence of highly purifie eicosapentaenoic acid ethyl ester on insulin resistance in the Otsuka Long-vans Tokushima Fatty rat, a model of spontaneous non-insulin-dependent diabetes mellitus. Metabolism 46: 1458-1464 (1997) Reddy W, Kupfer D, Caspi E. Mechanism of C-5 double bond introduction in the biosynthesis of cholesterol by rat liver microsomes. J Biol Chem 252: 2797-2801 (1977) Rommens JM, Lin B, Hutchinson GB, Andrew SE, Goldberg YP, Glaves ML, Graham R, Lai V, McArthur J, Nasir J, Theilmann J, McDonald H, Kalchman M, Clarke LA, Schappert K, Hayden MR. A transcription map of the region containing the Huntington disease gene. Hum Mol Genet 2: 901-907 (1993) Russo C, Olivier 0, Girelli D, Guarini P, Pasqualini R, Azzini M, Corrocher R.

Increased membrane ratios of metabolite to precursor fatty acid in essential hypertension. Hypertension 29: 1058-1063 (1997) Sambrook J, Fritsch EF, Maniais T. Molecular Cloning, A Laboratory Manual, 2nd Edition, Cold Spring Harbor Laboratory Press, USA (1989) Shanklin J, Whittle E, Fox BG. Eight histidine residues are catalytically essential in a membrane-associated iron enzyme, stearoyl-CoA desaturase, and are conserve in alkane hydroxylase and xylene monooxygenase. Biochem 33: 12787-12794 (1994) Shanklin J, Whittle EJ, Fox BG. Membrane bound desaturases and hydroxylases: Structure function studies, in Plant lipid metabolism, Kader JC, Mazliak P (eds), Kluwer Academic Publishers, Netherlands, pp. 18-20 (1995).

Sheets MD, Ogg SC, Wickens MP. Point mutations in AAUAAA and the poly (A) addition site: effects on the accuracy and efficiency of cleavage and poiyadenylation in vitro. Nucal Acid Res 18: 5799-5805 (1990)

Singer SJ, Nicolson GL. The fluid mosaic model of the structure of cell membranes. Science 175: 720-731 (1972) Sperling P, Schmidt H, Heinz E. A cytochrome b5-containing fusion protein similar to plant acyl lipid desaturases. Eur J Biochem 232: 798-805 (1995) Stöhr H, Marquardt A, Rivera A, Cooper PR, Nowak NJ, Shows TB, Gerhard DS, Weber BHF. A gene map of the Best's vitelliform macular dystrophy region in chromosome 11q12-q13.1. Genome Res 8: 48-56 (1998) Strittmatter P, Spatz L, Corcoran D, Rogers MJ, Setlow B, Redline R. Purification and properties of rat liver microsomal stearyl coenzyme A desaturase. Proc Natl Acad Sci USA 71: 4565-4569 (1974) Stubbs CD, Smith AD. The modification of mammalian membrane polyunsaturated fatty acid composition in relation to membrane fluidity and function. Biochim Biophys Acta 779: 89-137 (1984) Villani F, Comazzi R, De Maria P, Galimberti M. Effect of dietary supplementation with polyunsaturated fatty acids on bronchial hyperreactivity in subjects with seasonal asthma. Respiration 65: 265-269 (1998) Wang N, Anderson RE. Synthesis of docosahexaenoic acid by retina and retinal pigment epithelium. Biochemistry 32: 13703-13709 (1993)

3. A DNA comprising a nucleotide sequence with at least a 65 % homology with the nucleotide sequences as defined in claim 1 or 2.

4. A recombinant vector comprising the DNA as claimed in any of claims 1 to 3.

5. A transgenic host cell comprising the DNA as claimed in any of claims 1 to 3.

6. A transgenetic host cell transformed by the DNA according to any of claims 1 to 3 or the vector according to claim 4, a corresponding transgenetic organism or a corresponding transgenetic knock-in or knock-out animal model.

7. A polypeptide comprising at least 65 % of a polypeptide sequence selected from the group consisting of the polypeptides of SEQ ID NO: 1, SEQ ID NO: 2 and SEQ ID NO: 3, or its salt.

8. A polypeptide comprising a polypeptide sequence with at least a 85 % homology with the polypeptide sequence as claimed in claim 7, or its salt.

9. A peptide comprising at least 15 consecutive amino acids of the polypeptide as claimed in claim 7, or its salt.

10. A polypeptide having substantially the same amino acid sequence as the polypeptide as claimed in claim 7, or having a variant of the amino acid sequence of the polypeptide as claimed in claim 7 with a deletion, addition or substitution of 1 to 10 amino acids, or its salt.

11. A process for producing a polypeptide comprising expressing from the host cell of claim 5 or 6 a polypeptide encoded by the DNA as claimed in any of claims 1 to 3.