WO2000000593A2

WO2000000593A2 - Sphingolipid-desaturase

Info

Publication number: WO2000000593A2
Application number: PCT/DE1999/001859
Authority: WO
Inventors: Ernst Heinz; Ulrich ZÄHRINGER; Petra Sperling; Hermann Schmidt
Original assignee: Gesellschaft Für Erwerb Und Verwertung Von Schutzrechten - Gvs Mbh; Forschungszentrum Borstel Zentrum für Medizin und Biowissenschaften
Priority date: 1998-06-27
Filing date: 1999-06-25
Publication date: 2000-01-06
Also published as: AU5407599A; DE19828850A1; WO2000000593A3

Abstract

The invention relates to enzymes which selectively introduce a double bond in the sphingobase of the ceramide residual of sphingolipids and of capnoids. The invention also relates to DNA sequences which code for these enzymes, to the use of the enzymes and DNA sequences for modifying the content and/or the structure of sphingolipids, the catabolites thereof and/or their synthetic subsequent products in transgenic cells and/or organisms. In addition, a method is provided for producing sphingolipids and capnoids with an unsaturated sphingobase.

Description

Name: Sphingolipid desaturase

The invention relates to enzymes which selectively introduce a double bond into the sphingobase of the ceramide residue of sphingolipids and of capnoids (hereinafter also called “spingolipid desaturase” or “desaturase”). The invention further relates to DNA sequences coding for these sphingoiipid desaturases, the use of the sphingoiipid desaturases and the DNA sequences to change the content and / or the structure of sphingolipids, their catabolites and / or their synthetic secondary products in transgenic cells and / or organisms and a process for the preparation of sphingolipids and capnoids with unsaturated sphingobase (hereinafter also referred to as “sphingolipid”) and the sphingolipids thus obtained.

Sphingolipids are membrane components that are found in all eukaryotic cells and only in a few bacteria. The hydrophobic component, ceramide, contains a long-chain base (2-amino-1,3-dihydroxyalkane) which is linked to a long-chain fatty acid via an acid amide bond. A distinction is made between sphingophosphatides, sphingoglycolipids and sphingophosphoglycolipids based on the polar head group. Such a ceramide residue can also be found in capnoids (1-deoxy-ceramide-1-sulfonic acid) of sliding bacteria. Corresponding enzymes are also responsible there for the desaturation of capnoids in sliding bacteria (RA Drijber et al., "Cytophaga hutchinsonii ATCC 33406 contains a structural variant of the sulfonolipid N-acylcapnine", Can. J. Microbiol. 43: 689-6931997) .

Sphingolipids act as inter- and intracellular messengers in cell growth and differentiation, in apoptosis effects and in pathogen defense against microorganisms (AH Merrill et al., "Sphingolipids: metabolism and cell signaling", in 'Biochemistry of Lipids, Lipoproteins and Membranes', Vol 31: 309-339, DE Vance and JE Vance (ed.), Elsevier Science BV, Amsterdam (1996)). Various structural modifications of the ceramide are possible; however, few genes for its enzymatic variation have been cloned to date. It was not until late 1997 that the genes for the delta-4-hydroxylase of the long-chain base (D. Haak et al., "Hydroxylation of Saccharomyces cerevisiae Ceramide Requires Sur2p and Scs7p", J. Biol. Chem. 272: 29704- 29710 (1997)) and for the alpha-hydroxylase of the long-chain fatty acid (AG Mitchell et al., "Fahlp, a Saccharomyces cerevisiae Cytochrome b ₅ Fusion Protein, and Its Arabidopsis thaliana Homolog that Lacks the Cytochrome b ₅ Domain Both Function in the α - hydroxylation of sphingolipid-associated Very Long Chain Fatty Acids ", J. Biol. Chem. 272: 28281-28288 (1997)). However, no gene or DNA sequence that codes for a sphingolipid desaturase is known.

A number of enzymes and nucleic acids are known, the amino acid sequences or DNA sequences of which are somewhat similar to the sequences according to the invention. However, these are either a) stereoselective glycerolipid desaturases or DNA sequences coding for such glycerolipid desaturases that introduce a c / s double bond into acyl residues (J. Shanklin et al., “Eight Histidine Residues Are Catalytically Essential in a Membrane-Associated Iron Enzyme, Stearoyl-CoA Desaturase, and Are Conserved in Alkane Hydroxylase and Xylene Monooxygenase ", Biochemistry 33: 12787-12794 (1994); O. Sayanova et al.," Expression of a borage desaturase cDNA containing an N-terminal cytochrome b ₅ domain results in the accumulation of high levels of Δ ⁶ -desaturated fatty acids in transgenic tobaco ", Proc. Natl. Acad. Sei. USA 94: 4211-4216 (1997); and WO 96 / 21022) and not sphingoiipid desaturases, or b) DNA sequences in which the enzymatic function of the protein encoded by it is unknown (P. Sperling et al., “A cytochrome-b ₅ - containing fusion protein similar to plant acyl lipid desaturases ", Eur. J. Biochem 232: 798-805 (1995)). Furthermore, DNA sequences of Arabidopsis thaliana have been published as EST clones in gene banks (EMBL Accession Numbers: T42569, N37558, F13728, T42806 and F13717).

Sphingolipids are complex natural products with high heterogeneity that are found in plants, animals, insects, fungi and some bacteria. An inexpensive The production of homogeneous compounds on an industrial scale has so far not been possible, since corresponding genes have not yet been cloned. Since it was found that sphingolipids, as an essential component of the epidermal lipids, play a decisive role in stabilizing the barrier functions of the skin, they have meanwhile been used widely in cosmetics. However, since natural sphingolipids have to be extracted and purified by complex processes, they are very expensive, so that efforts are not lacking to even look for analog structures that can be easily synthesized from inexpensive starting materials (H. Möller et al., “Neue Pseudoceramides by Linking Special Fats with Carbohydrate Derivatives ", abstract and short presentation presented at the conference of the German Society for Fat Science eV (DGF) 'Innovation through combination: fat carbohydrate proteins', Bremen, October 7th, 1996).

The problem underlying the present invention was to provide the enzymes required for the biochemical synthesis of unsaturated sphingolipids or for the modification of sphingolipids.

Surprisingly, it was found that certain cDNA sequences isolated from the plants Arabidopsis thaliana and Brassica napus code for sphingoiipid desaturases with which sphingolipids with different regio- and stereoisomerism of the double bond in the sphingobase are accessible.

The invention thus relates to (1) sphingoiipid desaturases which selectively introduce a double bond into the sphingobase of the ceramide residue of sphingolipids and of capnoids. The sphingolipid desaturase according to the invention preferably introduces a Δ-6 double bond into the sphingobase. In the present application, “Δ-6” is to be understood as the relative position of the double bond introduced (a double bond in position C ₆ -C after the last oxygen-activated C atom (Ci); see FIG. 14). In addition, “delta- 4 "," delta-8 "," delta-9 "and" delta-10 "hereinafter denote the absolute position of the double bond. The enzyme according to the invention differs from the above-mentioned glycerolipid desaturases with regard to its substrate specificity (long-chain base of the sphingolipids), its regioselectivity (position of the introduced double bond delta-8 in plants or delta-8 and delta-9 in transgenic yeast) and its stereoselectivity ( ice and trans).

In a preferred embodiment, the amino acid sequence comprises one or more of the fragments a) - c) or c) - e), sphingoiipid desaturases whose amino acid sequence comprising all three fragments a) - c) or c) - e) being particularly preferred : a) HDAGH, b) HN [A, S] HH (ie HNAHH or HNSHH); c) QLEHH; d) AYX _a X _b HD [A, S] GH, where X _a and X _{b are} any amino acid residues, and e) THNAHH.

The fragment d) preferably has the sequence AYX _a GHDAGH, wherein X _a is any amino acid residue. In addition to fragments a) - e), the desaturases according to the invention can also contain the fragments AWWKWT and / or FGGLQF.

The desaturases according to the invention do not include a desaturase with the amino acid sequence shown in FIG. 15. Particularly preferred sphingoiipid desaturases of the present invention include the amino acid sequences shown in Fig. 2 or 4.

The invention further relates to (2) DNA sequences which code for the above desaturases defined in (1), the DNA sequence shown in FIG. 15 not being included. The DNA sequence is preferably obtainable from Arabidopsis thaliana and Brassica napus, with DNA sequences comprising the sequences shown in FIG. 1 or 3 being particularly preferred.

Furthermore, (3) vectors comprising the DNA sequences defined in (2) are disclosed. The vectors according to the invention can be used in addition to the DNA Sequences still functional DNA sequences promoting the expression of these DNA sequences such. B. promoters, a termination signal and / or secretion sequences that are functionally linked to the DNA sequences (2).

The invention further relates to

(4) cells transformed with one of the vectors defined in (3), these cells preferably being plant, fungal, bacterial or animal cells;

(5) transgenic organisms which comprise a) a DNA sequence as defined above under (2), b) a vector as defined above under (3) and / or c) from a cell as above under ( 4) defined, can be generated;

(6) plants or their progeny which can be generated from the plant cell as defined under (5) above, the plants preferably being useful plants and / or having a modified delta-8 base pattern; and

(7) a process for the production of plants which has an increased or reduced proportion or a modified cis / trans ratio of delta-8-unsaturated long-chain bases, comprising a) transforming a plant cell with a DNA sequence, as described above under ( 2), including the DNA sequences from FIG. 15, and b) the regeneration of a plant with an altered delta-8 base pattern in its sphingolipids.

The method according to (7) is a) to compensate for a lack of delta-8-unsaturated long-chain bases in an organism b) to switch off the production of delta-8-unsaturated bases, c) to produce plants with improved tolerance and resistance against salinization and / or ionic stress or toxicity and / or drought and / or moisture and / or cold or frost and plant pathogenic microorganisms and d) suitable for the production of plants with changed size growth and flowering time. For the production of transgenic plants with a greatly reduced content of deIta-8 unsaturated long-chain bases in their sphingolipids, e.g. B. the above-mentioned gene from A. thaliana in antisense orientation and the above-mentioned gene from B. napus in sense orientation (co-suppression) can be cloned into a constitutive plant expression vector. The transformation of A. thaliana then takes place e.g. B. with the help of Agrobacterium fcvme ac / ens-mediated in planta vacuum infiltration (N. Bechthold et al., "In Planta Agrobacterium mediated gene transfer by infiltration of adult Arabidopsis thaliana plants", CR Acad. Sei., Paris, Life Science 316: 1194-1199 (1993); AF Bent et al., "RPS2 of Arabidopsis thaliana: A Leucine-Rich Repeat Class of Plant Disease Resistance Genes", Science 265: 1856-1860 (1994)). Successful suppression of delta-8 sphingolipid desaturation can be demonstrated in the transgenic plants by analyzing the long-chain bases isolated from the sphingolipids.

The invention further relates to

(8) a method for producing sphingolipids and capnoids with unsaturated sphingobase, comprising culturing cells, organisms or plants as defined in (4) - (6) above. The method further includes isolating the sphingolipids from the culture. The sphingolipid produced is preferably a compound of the formula (I)

wherein

R ^{1 is} -OH, -OPO ₃ HR ⁶ , a carbohydrate residue or -S0 ₃ H;

R ^{2 is} a dC ₃₃ alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more hydroxyl radicals;

R ³ and R ^{5 are} independently -H or -OH, or R ³ and R ⁵ together form a trans double bond; R ^{4 is} -H or a C 1 -C ₁₅ -alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more methyl radicals; R ^{δ is} -H, a polyhydroxyalkyl group, an aminoalkyl group or a carbohydrate group; and either a, b or c is a double bond,

(a) when R ³ and R ^{5 are} -H, a is a double bond, and

(b) when R ^{3 is} -OH and R ^{5 is} -H, a or b is a double bond.

In the sphingolipids produced, R ² is preferably a C ₁ -C ₂ -alkyl radical and R ^{4 is} a C ₃ -Ci2-alkyl radical, in particular a C ₇ -, Cg- or Cn-alkyl radical. A carbohydrate residue in the sense of the present invention consists of one or more glycosidically linked aldopentoses, aldohexoses, ketohexoses and the amino, urone and deoxy derivatives derived therefrom. Polyhydroxy compounds are to be understood as radicals derived from alkane polyols such as glycol, glycerol and mannitol. Aminoalkyl radicals for the purposes of the present invention are 2-aminoethyl and 2- (trimethylammonium) alkyl radicals. In particular, the sphingolipid produced is a compound of the formula (II)

wherein R ¹ , R ² , R ³ , a and b have the meaning given above.

In the cloning and expression according to the invention of the gene s / / 1 coding for a sphingolipid desaturase from cDNA of the plants Arabidopsis thaliana and Brassica napus in the yeast Saccharomyces, novel products could be detected in the transgenic yeasts, the chemical structure of which was clearly identified. These are different sphingolipids whose spingobases have different regio- and stereoisomerism of the double bond: (i) D-eryf 7ro-2-amino-1, 3,4-trihydroxy- (8Z) -octadecene (t18: 18 ^ö c _λ )

(ii) D-etyf / Vo-2-amino-1, 3,4-trihydroxy- (8E) octadecene

9cx

D-etyf / 7ro-2-amino-1, 3,4-trihydroxy- (9Z) -octadecene (t18: 1 ^ao ) (iv) De / yf 7ro-2-amino-1, 3,4-trihydroxy- ( 9E) octadecene (t18: 1 ^9t )

Surprisingly, not only the plant-typical delta-8 stereoisomers (i) and (ii), as described, for. B. from H. Imai et al., "Sphingoid Base Composition of Cerebrosides from Plant Leaves", Biosci. Biotech. Biochem. 61: 351-353 (1997), but also delta-9, which does not occur in nature Regioisomer (iii) and (iv) of the phytosphinine formed.

The invention thus also relates to (9) spingobases and capnoids with unsaturated sphingobase with the formula

(D.

wherein

R ^{1 is} -OH, -OPO ₃ HR ⁶ , a carbohydrate residue or -S0 ₃ H;

R ^{2 is} a -CC ₃₃ alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more hydroxyl radicals;

R ³ and R ^{5 are} independently -H or -OH, or R ³ and R ⁵ together form a trans double bond;

R ^{4 is} -H or a Ci-Cis-alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more methyl radicals;

R ^{6 is} -H, a polyhydroxyalkyl group, an aminoalkyl group or a carbohydrate group; and either a, b or c is a double bond,

(a) when R ³ and R ^{5 are} -H, a is a double bond, (b) when R ^{3 is} -OH and R ^{5 is} -H, a or b is a double bond, and (b) when R ^{4 is} a C ₇ -, C ₉ - or Cn-alkyl radical, a is not a double bond, and Derivatives of the same.

Preferred embodiments are mentioned above under (8). “Derivatives” are to be understood as products from chemical or enzymatic reactions, with this reaction the molecular weight of the sphingobase can both be increased (eg by O-acylation and hydroxylation of double bonds) and reduced by cleavage reactions (eg by splitting off the fatty acid residue of the ceramide residue). “Derivative” is therefore also to be understood as the free sphingobase.

Sphingolipids with the following formula are preferred:

(St.).

wherein R ¹ , R ² , R ³ , R ⁴ , R ⁵ , b and c have the meaning given above.

Finally, the invention relates to (10) cosmetics, pharmaceuticals or foodstuffs or chemical raw materials which comprise a compound as produced according to method (8) or as defined in (9). Medicines include agents for regulating blood pressure and heart rhythm, for modulating insulin secretion, agents for wound healing, agents with anti-tumor activity, protective action against radioactivity, stimulation of cell growth, influencing leukocyte differentiation, immunoreceptors for viruses, bacterial and mycotoxins. The numerous human sphingolipid storage diseases, such as. B. Niemann-Pick and Tay-Sachs disease, also put an application as therapeutic effective fine chemical close to humans, with a high pharmacological, toxicological and nutritional potential.

Further possibilities to use these substances economically are to manipulate their content and / or composition in certain organisms and thus to change the economically relevant properties of these organisms positively. We see market opportunities in the area of changes with regard to growth behavior (length increase and leaf blade size) as well as influencing the start of flowering in transgenic crop plants. Heterologous expression and / or homologous overexpression and / or antisense elimination of the sphingolipid desaturase genes could result in crop plants with improved tolerance or resistance 1.) to harmful environmental influences such as soil salinity, ion stress and toxicity, dryness, moisture and cold / frost (M . Uemura et.al., "Cold Accimation of Arabidopsis thaliana", Plant Physiol. 109: 15-30 (1995) or 2.) against plant-pathogenic microorganisms (for example fungus resistance). Most plants which are resistant to coolers are distinguished by a higher one Proportion of delta-8-c / s-phytosphingenin from (H. Imai et al., See above) Furthermore, it has been shown that vegetable and fungal cerebrosides with delta-8-unsaturated long-chain bases can induce or increase the fruiting of mushrooms (Kawai et al., "Stimulatory effect of certain plant sphingolipids on fruiting of Schyzophyllum commune", J. Biol. Chem. 261: 779-784 (1986)).

By antisense switching off the corresponding Δ6-sphingolipid desaturase gene in microorganisms, such as. B. yeasts and fungi, their phyto- and / or human pathogenicity could be reduced. By heterologous expression of the plant sphingolipid desaturase in microorganisms such as. B. Baker's, brewer's yeast and their mutants (RC Dickson et al., "Isolation of mutant Saccharomyces cerevisiae strains that survive without sphingolipids", Mol. Cell Biol. 10: 2176-2181 (1990); WJ Pinto et al., " Characterization of Enzymatic Synthesis of Sphingolipid Long-Chain Bases in Saccharomyces cerevisiae: Mutant Strains Exhibiting Long-Chain-Base Auxotrophy Are Deficient in Serine palmitoyltransferase Activity ", J. Bacteriol. 174: 2575-2581 (1992)) could include these in their Growth properties, e.g. B. temperature tolerance, be improved and serve as bioreactors for natural and artificial spingolipids and their derivatives.

Expression of the plant sphingolipid desaturase can also lead to modified sulfonolipids and derived products in prokaryotes, which on the one hand change their mobility and on the other hand could be economically interesting as new chemical compounds.

The invention is illustrated by the following figures and examples.

Figure description

1: Nucleotide sequence (1594 bp) of the sphingolipid desaturase (s / 1) from Brassica napus with 5'- and 3'-untranslated regions and an open reading frame.

Fig. 2: Reduced amino acid sequence (449 aa) of the sphingolipid desaturase from β. napus.

3: Nucleotide sequence (1678 bp) of the sphingolipid desaturase (s / cf1) from Arabidopsis thaliana with 5'- and 3'-untranslated regions and an open reading frame.

Fig. 4: Reduced amino acid sequence (449 aa) of the sphingolipid desaturase from A. thaliana.

Fig. 5: Protein sequence comparisons of the Sphingoiipid desaturases from β. napus (BnDESδ, Fig. 2), A. thaliana (AtDES8, Fig. 4) and sunflower (HaDES ?, (Sperling et al., 1995), Fig. 15) with the delta-6 glycerolipid desaturase from Borage (B0DES6 (Sayanova et al., 1997)). The N-terminal domain up to position 121 of the sunflower sequence is homologous to the hydrophilic part of cytochrome b ₅ , the conserved amino acids of which are underlined. The three highly conserved histidine boxes (Shanklin et al., 1994), which are characteristic of lipid desaturases, are shown. Identical amino acids are highlighted by gray shading.

6: Vector construct for the expression of the s / crt gene from β. napus (pYES2Bn) in the yeast Saccharomyces cerevisiae.

7: Vector construct for the expression of the s / cd gene from A. thaliana (pYES2At) in the yeast Saccharomyces cerevisiae.

Fig. 8: Formation of phytosphine genes in yeast cells by heterologous expression of plant sphingoiipid desaturases. The long-chain bases (LCB) of the sphingolipids were obtained from whole cells by strong alkaline hydrolysis (Morrison and Hay, 1970), converted into their dinitrophenyl derivatives (Karisson, 1970) and analyzed by RP-HPLC. (A) Separation of the reference substances (Sigma) phytosphinganine (t18: 0), 4-rrans-sphingenine (d18: 1) and sphinganine (d18: 0). (B) LCB pattern of the wild-type yeast cells (INVSd) or transformed with the empty vector pYES2. Formation of eis- and frans-phytosphingenin (t18: 1) in yeast cells which contain either (C) the A.thaliana fusion protein (pYES2At) or (D) the ß. express napus fusion protein (pYES2Bn). (E) LCB pattern of A. thaliana (wild type) plants.

9: GC-MS elution profile of fatty acid methyl esters which are obtained by KMn0 oxidation of an unsaturated phytosphinganine isomer mixture (Example 5).

Fig. 10: Partial ¹ H-NMR spectra (600 MHz) of (A) trans-Δ ⁸ ' ⁹ - and (B) cis-Δ ⁸ ' ⁹ - phytosphingenin. The signals were assigned using COZY experiments.

11: Vector construct for heterologous overexpression in sense orientation of the s / c / 1 gene from β. napus (pBIB5Bn) in A. thaliana.

Fig. 12: Vector construct for homologous antisense suppression of the s / c 1 gene from A. thaliana (pBIB5At) in A. thaliana. Fig. 13: Suppression of delta-8 sphingolipid desaturation in transgenic A. thaliana plants. The long-chain bases (LCB) of the sphingolipids were obtained from mixed samples (350 mg fresh weight) of whole T2 plants, 34 d after germination, by strong alkaline hydrolysis (Morrison and Hay, 1970) and converted into their dinitrophenyl derivatives (Karisson, 1970) and analyzed by RP-HPLC. (A) Separation of the reference substances (Sigma) phytosphinganine (t18: 0), 4-trans-sphingenine (d18: 1) and sphinganine (d18: 0). (B) LBC pattern of the wild-type plants or transformed with the empty vector pBI121. Reduction of ice and trans-A8 phytosphingenin (t18: 1) in T2 plants (C) by homologous antisense expression of the s / cd gene from A. thaliana (pBIB5At) and (D) by heterologous sense expression ( Co-suppression) of the s / c / 1 gene from ß. napus (pBIB5Bn).

Fig. 14: Regioselectivity of the plant sphingolipid desaturase. The Δ-6-sphingolipid desaturases introduce a delta-8 and -9 double bond in the eis and frans configuration into long-chain bases, which leads to the formation of phytosphingenin or phytosphinginin-containing sphingolipids (t18: 1) in the transformed yeast cells (R represents a variety of possible fatty acid residues).

15: nucleotide and amino acid sequence derived therefrom from P. Sperling et al. (1995), p. O.

Fig. 16: Extended protein sequence comparison of sphingoiipid desaturases (the top 5 sequences) with Δ6 fatty acid desaturases (BnDES8, AtDES8, HaDES? And B0DES6 see Fig. Δ; dδlpu, bδlbo and dδ2pu are unpublished partial sequences of the applicant, 2bδca bδpp have the EMBC Accession Numbers Z70271, Z81122 and AJ222980).

Examples

The following abbreviations are used: cDNA copy-deoxyribonucleic acid FAME fatty acid methyl ester

GC-MS gas chromatography mass spectrometry

HPLC high pressure liquid chromatography

LCB long chain bases (of the sphingolipids) mRNA messenger ribonucleic acid

NMR 'Nuclear Magnetic Resonance' spectroscopy

PCR polymerase chain reaction

TLC thin layer chromatography

Example 1: Isolation and cloning of S / Q * 1 from B.napus

Unless otherwise stated, restriction endonucleases and DNA-modifying enzymes were obtained from New England Biolabs and Boehringer Mannheim and used according to the manufacturer's instructions. E. coli XL1 blue, MRF '(Stratagene) was at 37 ° C in LB medium (J. Sambrook et al., "Molecular Cloning: A Laboratory Manual", 2nd edn., Cold Spring Harbor Laboratory Press, USA (19δ9 For the selection of E. colis transformed with plasmids, the antibiotics ampicillin (100 μg / ml) and kanamycin (30 μg / ml) were added to the medium, and competent E. coli were transformed according to H. Inoue et al. , "High efficiency transformation of Escherichia coli with plasmids", Gene 96: 23-2δ (1990) and the DNA mini-preparation according to MG Riggs et al., "A simplified screening procedure for large numbers of plasmid mini-preparation", BioTechniques 4 : 310-313 (19δ6).

In order to isolate a homologous DNA sequence from Brassica napus, degenerate oligonucleotide primers were derived from the functionally unknown DNA sequence from Helianthus annuus (Sperling et al. (199δ), see above). A δ71 bp long DNA fragment was extracted from a lambda-ZAP cDNA library from maturing pods of Brassica napus cv. Ascari (M. Fulda et al., "Brassica napus cDNAs encoding fatty acyl-CoA synthetase", Plant Mol. Biol. 33: 911-922 (1997)) was isolated by means of PCR (polymerase chain reaction). For this, the degenerate primers were used

BN1 (δ'-G [GC] [ATGC] TGGTGGAA [AG] TGG-3 ') and BN2 (δ'-GG [AG] AA [ATGC] A [AG] [AG] TG [AG] TG-3' ) used, which correspond to the amino acid sequences [GA] WWKW and HHLFP from H.annuus.

The following program was used for the amplification: δ min at 96 ° C, followed by 30 cycles of 20 s at 96 ° C, 1 min at 40 ° C (binding temperature, T _m ) and 2 min at 72 ° C, 1 cycle of 10 min at 72 ° C. Unless otherwise stated, Gibco BRL used the TaG / DNA polymerase for the amplification. The δ 'end of the cDNA was treated with the Stratagene T3 primer (20 mer) and the specific back primer

BN3 (δ'-TTATGCGTCCATTTCCACCA-3 ') amplified by PCR as described, except that the Tm was δδ ° C.

In order to amplify the 3 'end of the DNA, cDNA was made from mRNA maturing embryos of ß. napus cv. Drakkar used. For this purpose, mRNA was extracted from ß maturing embryos using oligo (dT) Dynabeads according to the instructions for use from Dynal GmbH (Hamburg). napus cv. Drakkar isolated. This mRNA was converted into a single-strand cDNA synthesis using the reverse transcriptase reaction using the SuperScript preamplification system (Gibco BRL) and a synthetic oligo (dT) primer

(δ'-GA TCTAGA CTCGAG GTCGAC T ₁₄ -3 '), which carries the three restriction sites for Xbal, Xhol and Sall (underlined). The synthesized cDNA was purified on SizeSep 400 columns from Pharmacia according to their instructions. Then the 3 'end of the DNA was with the specific forward primer

BN4 (δ'-TTCTTTGGCGGGTTGCAGTT-3 ') and a synthetic oligo (dT) primer

(5'-GA TCTAGA CTCGAG GTCGAC T ₄ -3 ') amplified using the above-mentioned PCR conditions at Tm 54 ° C. The cloning and sequencing of the double-stranded cDNA fragments from the different PCR amplifications (δ71 bp, δ 'and 3' end) was carried out as described (P. Sperling et al. (1996), see above): The DNA fragments were alloyed with the SureClone Ligation Kit (Pharmacia, Freiburg) into the plasmid vector pUC1δ / Smal, transformed into E. coli XLIblue (Stratagene) and sequenced on both strands with the ABI PRISM Dye Primer Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, Weiterstadt). The length of the complete clone derived from the overlapping partial sequences is 1δ94 bp (FIG. 1), which codes for an open reading frame of 449 amino acids (FIG. 2). Specific primers

BNδ (δ'-AACCATCTCTGTTTCAAC-3 ^* ) and BN6 (δ'-CAA GTGATGATGAGTTAC-3 '),

which were derived from the δ'- and 3'-untranslated regions were analyzed in a further PCR (T _m δδ ° C) with cDNA from ß. napus cv. Drakkar used. This reaction was carried out with the P u-DNA polymerase (Stratagene). A complete clone of 1δ02 bp (PCR clone 1) was isolated, alloyed with the SureClone Ligation Kit (Pharmacia) in pUC1δ / Smal and transformed into E. coli XLIblue (Stratagene). The clone was sequenced with the ABI PRISM Dye Terminator Kit (Perkin-Elmer) and matches the nucleotide sequence in FIG. 1 and the amino acid sequence in FIG. 2.

For later expression in yeast, a complete clone (1347 bp), which contained the coding sequence, was PCR-derived with the derived primers

BN7, forward δ'-CCGGTACCATGTCGGAGCAGACAAAG-3 'and BN8, reverse δ'-CCGAATTC CTAGCCATGAGTATTCAGA-3'

amplified at T _m δ6 ° C. The start and stop codons are printed in bold and the restriction interfaces Kpnl and EcoRI are underlined. A complete clone of 1363 bp (PCR clone 2) was isolated and using the pGEM-T vector system cloned by Promega in E coli XLIblue. To check this clone was sequenced with the ABI PRISM Dye Primer Kit (Perkin-Elmer).

Example 2: Isolation and cloning of s / dl from A. thaliana In order to isolate the s / c / 1 gene from Arabidopsis thaliana, a TBLASTN search (SF Altschul et al., "Basic local alignment search tool", J. Mol. Biol. 215: 403-410 (1990)) in databases with the derived protein sequence from ß.napus (Fig. 2.) Parts of the protein sequence from ß.napus showed a significant similarity to the following EST clones from A. thaliana whose EMBL accession numbers are: T42569 (27-93), N37δδ8 (33-196), F13726 (113-201), T42306 (366-396), F13717 (421-449). The numbers in brackets refer to to the amino acid positions of the rapeseed sequence (Fig. 2) covered by these EST clones.

Specific oligonucleotide primers were derived from the nucleotide sequences of the EST clones

T42569: (AT1, forward) δ'-GCGATTCAAGGCAAGGTCTAC-S 'and F13717: (AT2, reverse) δ'-TTAGCCATGAGTATTCAAAGC-3 ^*

derived, with the stop codon printed in bold. These primers were used for the PCR amplification (2 min at 9δ ° C, 30 cycles with 20 s at 9δ ° C, 30 s at T _m δ9 ° C, 1 min at 72 ° C, 1 cycle 10 min at 72 ° C ) a 1272 bp DNA fragment from a lambda-ZAP cDNA library from A. thaliana. This cDNA bank was obtained from the Genetic Stock Center of the Max Planck Institute in Cologne, which was mainly isolated from green parts of the plant (leaves, some flowers, few roots). The δ 'or 3' end was amplified by means of the above-mentioned PCR (T _m 60 ° C.) with the primer combination T3 primer (Stratagene) and AT2 or T7 primer and AT1. The 1272 bp DNA fragment (PCR clone 3) was converted into pUC 19 / Smal and the 5 'and 3' DNA fragments were analyzed with the pGEM-T Vector System (Promega) in E. coli XLIblue (MRF \ Stratagene ) cloned. The sequencing was carried out using the ABI PRISM Dye Primer Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, Weiterstadt). From these three overlapping partial sequences a complete sequence of 167δ bp (Fig. 3) could be derived, which for an open reading frame of 449 amino acids coded (Fig. 4). A complete clone, which contained the coding sequence, was PCR-derived with the derived primers

AT3, forward: δ'-CCGGTACCATGGCGGAAGAGACGGAG-3 'and AT4, reverse: δ'-CCGAATTC TTAGCCATGAGTATTCAAAGC-3'

amplified at T _m 66 ° C. The start and stop codons are printed in bold and the restriction sites Kpnl and EcoRI added for later cloning into a yeast expression vector are underlined. A complete clone of 1363 bp (PCR clone 4) was isolated, cloned with the pGEM-T vector from Promega and transformed into E. coli XLIblue. This clone was sequenced on both strands with the ABI PRISM Dye Terminator Kit (Perkin-Elmer) and corresponds to the nucleotide sequence in FIG. 3 and the amino acid sequence in FIG. 4.

The s / c / 1 nucleotide sequences from β. napus and A. thaliana both have an open reading frame of the same length, which is 9 amino acids shorter at the N-terminus than the corresponding protein from the sunflower (P. Sperling et al. (199δ), see above). The two derived polypeptides of approx. 61-62 kD have 76% identical amino acids to each other and 6δ% and 64% to the sunflower sequence. The greatest similarity between the s / c 1 proteins and the delta-6 glycerolipid desaturase from Borage (O. Sayanova et al. (1997), see above), whose protein sequence is one amino acid shorter, 60% and 66% respectively the amino acids are identical (Fig. δ).

Example 3: Cloning and expression of s / c / 1 in Saccharomyces cerevisiae: The cloning and expression of s / c 1 was based on the successful expression method of the fac / 2 gene from A. thaliana in yeast (S. Kajiwara et al ., "Polyunsaturated fatty acid biosynthesis in Saccaromyces cerecisiae: Expression of Ethanol Tolerance and the FAD2 Gene from Arabidopsis thaliana", Appl. Environ. Microbiol. 62: 4309-4313 (1996)). The amplified s / c / 1 DNA from ß napus and A. thaliana (PCR clones 2 and 4), which had been cloned into the multiple cloning site of the vector pGEM-T vector (Promega) (see above), were cut with the restriction endonucleases Kpnl and EcoRI Kpnl / EcoRI- Fragments were inserted into the Kpnl / EcoRI site of the zipped yeast E. coli shuttle expression vector pYES2 alloyed (Invitrogen Co.). The resulting new plasmids, pYES2Bn and pYES2At, are shown in Fig. 6 and Fig. 7 and were transformed into E. coli TOP10F '(Invitrogen Co.).

The DNA maxi preparation was carried out with the plasmid Maxi Kit (Qiagen). The isolated plasmids pYES2Bn and pYES2At were checked again with the ABI PRISM Dye Primer Ready Reaction Kit (Perkin-Elmer) and then in S. cerevisiae INVSd (Invitrogen Co.) with the polyethylene glycol method (M. von Pein, dissertation, Heinrich Heine University Düsseldorf (1992)) transformed.

The selection was carried out on agar plates from complete minimal dropout uracil (CM) medium with 2% glucose (FM Asubel et al., "Current Protocols in Molecular Biology", John Wiley & Sons, New York 1996). The review of the transgenic Yeast colonies were carried out by means of a mini preparation of yeast plasmid DNA (Asubel et al. (1996), see above) whose retransformation in E. coli TOP10F ', subsequent mini preparation and restriction analysis of the E.co// ^' plasmid DNA with Kpnl / EcoRI the analytical comparison of the transformants with the wild-type yeast also transformed the uncut pYES2 vector into INVSd (control).

The preculture was carried out by inoculating 2-δ ml of CM- 2% raffinose medium with a transgenic yeast colony and then incubating for 2 d at 30 ° C. in a roller up to an optical density at 600 nm (ODβoo) of 4-δ. For the main culture, CM-2% raffinose medium was inoculated with an aliquot of the preculture (600-fold dilution) to OD ₆₀ or 0.01-0.02 and incubated in the shaker at 30 ° C., 2δ0 rpm for 24 h. The test cultures were induced in the logarithmic growth phase (OD ₆ oo 0.1-0.6) by adding galactose ad 1.8%. The induced cells were harvested after a further 24 h of aerobic growth at 30 ° C. at OD ₆₀ o 4-δ.

Example 4: Analysis of the long-chain bases from the sphingolipids of the yeast transformants The induced yeast cells were harvested by centrifugation at 2000 g for 10 min, twice with distilled water. washed, 20 min in distilled water boiled, and sedimented again after cooling on ice.

Since S. cerevisiae mainly contains only the hard-to-extract sphingophosphoglycolipids, the long-chain bases (LCB) were obtained by strong alkaline hydrolysis (24 h at 110 ° C) of whole cells (350 mg fresh weight) in 10% Ba (OH) ₂ / dioxane (1 : 1, v / v) released (WR Morrison et al., "Polar lipids in bovine milk II. Long-chain bases, normal and 2-hydroxy fatty acids, and isomeric eis and trans monoenoic fatty acids in the sphingolipids", Biochim Biophys Acta 202: 460-467 (1970)) The LCB were extracted from the hydrolyzate ad CHCI ₃ / dioxane / H ₂ 0 (6: 1: 5), the CHC phase was distilled off, and the free LCB of the hydrolyzate were removed converted directly into their dinitrophenyl derivatives (K.-A. Karisson, "On the occurrence of sphingolipid long-chain bases", Chem. Phys. Lipids 5: 6-43 (1970)). The dinitrophenyl LCBs were extracted with CHCI ₃ / methanol / H ₂ 0 (δ: 4: 3), purified by thin layer chromatography in CHCI ₃ / methanol (90:10) (K.-A. Karisson (1970), see above) and with Methanol extracted from the silica gel.

A reverse phase high performance liquid chromatography (RP-HPLC) method was developed for the analysis of dinitrophenyl LCB. The different species were separated using an ODS Hypersil ^® RP1δ column (5 μm, 26 x 4.6 cm) from Bischoff. Elution was carried out using a linear gradient at a flow rate of 1 ml / min: from 80% methanol / acetonitrile / 2-propanol (10: 3: 1, v / v / v) and 20% distilled water. to 0% aqua dest. in 40 min. The elution of the dinitrophenyl derivatives was detected at 360 nm in proportion to the mass (FIG. 8). The reference LCB were obtained from Sigma (FIG. Δ A). The HPLC analysis showed that transgenic yeast cells, which contain the plasmid pYES2At or pYES2Bn, produce at least two novel LCB (Fig. Δ C, D) which do not contain in wild-type and pYES2-transformed yeast (Fig. Δ B) are. Calculated from the HPLC data, the ratio of the novel eis and trans-phytosphingenins (t1δ: 1) was 1: 6.7 for transgenic yeasts with pYES2Bn and 1: 3 with pYES2At. The stereoisomers delta-δ and delta-9 were not separated by HPLC.

Example δ: Analysis of the novel long-chain sphingobases using NMR and MS Isolation and derivatization of the phytosphine genes: For a chemical structure analysis, the LCB were isolated from transgenic yeast cells (δ, 7 g fresh weight) by means of alkaline hydrolysis (see Example 4). The fraction of the free C1δ trihydroxy bases (phytosphinganin and phytosphingenin, R _f 0.34) was separated from the remaining bases of the hydrolyzate by thin layer chromatography in CHCI ₃ / methanol / NH ₄ (40: 40: 1) (M. Ohnishi et al. , Biochim. Biophys. Acta 752: 416-422 (1933)), extracted with CHCI ₃ / methanol / 1N KOH (2: 1: 0.75) and extracted with acetic anhydride, 4- (dimethylamino) pyridine in pyridine. O-acetylated (CM Gupta et al., "Glycerophospholipid synthesis: improved general method and new analogs containing photoactivable groups", Proc. Natl. Acad. Sci. USA 74: 4315-4319 (1977)). The per-O-acetylated C13-trihydroxy bases were extracted in n-hexane against water (1: 1) and purified by thin layer chromatography in 100% diethyl ether, detection with 0.2% methanolic δ-anilinonaphthalene-1-sulfonic acid under UV light and extraction with n -Hexane: The separation of the per-O-acetylated base fraction (0.73 mg) into phytosphinganine (125 μg t1δ: 0, R _f 0 , 46), in trans- (300 μg t18: 1 ^l , R _f 0.44) and in c / s-phytosphingenin (100 μg t1δ: 1 ^c , R _f 0.44), thin-layer chromatography was carried out using 10% silver nitrate Silica gel plates (O. Renkonen et al., "Structure of plasma sphingadienine", J. Lipid Res. 10: 6δ7-693 (1969)) in CHCI ₃ / methanol (95: 5) at 4 ° C. The bands were detected at 0.2 % methanolic dichlorofluorescein under UV light and the extraction was carried out with n-hexane.

Λ // W /? - SpeWros / op / ^' e: ¹ H-NMR spectra were recorded with a 600 MHz Bruker Avance DRX 600 spectrometer, equipped with a micro probe head (PH TXI 600SB), with capillary tubes of 2.5 mm OD ( Wilmad, Buena, NJ, USA). The per-O-acetylated samples (60-300 μg) were dissolved in 100 μl CDCI ₃ (Cambridge Isotope Lab., MA., USA) and their spectra recorded at 300 K with TMS as internal standard (5H = 0.000 ppm). One- and two-dimensional homonuclear ¹ H, ¹ H COZY experiments were shown with the standard Bruker software (XWINNNMR, version 1.3). GC-MS analysis: The gas chromatographic

Mass spectroscopy (GC-MS) was carried out with a Hewlett-Packard model 5989 spectrometer with an HP-5 capillary column (30 m x 0.2δ mm). The following temperature gradients were used for the analysis of the various derivatives: Per-O-acetylated phytosphingenins: 3 min at 230 ° C, gradient from δ ° C / min to 330 ° C and medium-chain fatty acid methyl ester: 3 min at 100 ° C, gradient from 10 ° C / min to 330 ° C. GC-MS analysis was carried out in Cl (with ammonia) and El mode (70 eV).

Derivatization of phytosphine genes for GC-MS

A) Lead tetraacetate oxidation: An aliquot of the unfractionated, peracetylated C18-trihydroxy bases (0.36 mg) were desacyated with sodium methylate and then oxidized with Pb (OAc) ₄ in chloroform. The resulting aldehydes were extracted into chloroform, evaporated to dryness and reduced to the corresponding alcohols with LiAIH in ether. 1/10 of the alcohols were converted into their trimethylsilyl derivatives with BSTFA (Sigma) and analyzed directly by GC-MS. The remaining alcohols (9/10) were converted to nicotinates with nicotinic acid chloride in pyridine, extracted into water in n-hexane and analyzed by GC-MS.

B) Potassium permanganate oxidation: An aliquot of the peracetylated C1δ-trihydroxy bases was oxidized under alkaline conditions with KMn0 ₄ in methanol / water (2: 1). After acidification, the resulting fatty acids were extracted and converted into their methyl esters with ethereal diazomethane and analyzed by GC-MS.

The fraction of the per-O-acetylated C1δ-trihydroxy bases from the yeast transformed with pYES2At was analyzed by GC-MS, whereby three distinct peaks were obtained. A peak eluting at 15.72 min was identified as 2-N-acetamido-1, 3,4-tri-O-acetyl-octadecane (phytosphinganine, t1δ: 0). The additional two peaks eluted at 15.30 and 15.46 min were considered as 2-N-acetamido-1, 3,4-tri-O-acetyl-octadecene (m / z ^" 144, 423 [M-60] ⁺ , 433 [M] ⁺ ) (phytosphingenin, t1δ: 1). The t1δ: 0 and the two t1δ: 1 isomers were in a ratio of 3: 2: 1. The two nicotinate derivatives obtained after lead tetraacetate oxidation from t1δ: 1 resulted in two peaks (15, δO and 15.9δ min) with the expected molecular mass ion (M = 331) in the GC-MS analysis. Based on the observed fragmentation pattern of the nicotinates, the position of the double bond could only be approximated in the first peak between C6 and C7 (penta-Δ6-decene, m / z = 137, 151, 165, 173, 192, 206, 21 δ, 232) and located in the second peak between C7 and Cδ (penta-Δ5-decene, m / z = 137, 151, 165, 17δ, 192, 204, 21 δ, 232).

For a clear localization of the double bond, the fraction of the C13 trihydroxy bases was oxidized with potassium permanganate and converted into its fatty acid methyl ester (FAME). A methyl pentadecanoate (C15: 0) was obtained from the per-O-acetylated t1δ: 0 due to the oxidation of the diol groups between C3 and C4. In addition, two medium-chain FAME were obtained and identified by GC-MS as nonanoic acid methyl ester (C9: 0, 7.5 min) and decanoic acid methyl ester (C10: 0, 9.0 min). Both fatty acids were present in a ratio of 1: 3, from which it follows that in the original sample there were two phytosphingenin isomers (t1δ: 1 ⁸ and t1δ: 1 ⁹ ) with different positions of the double bond in positions delta-δ and -9. Furthermore, the acetylated per-O-C1δ-Trihydroxybasen on AgN0 ₃ were - Dünnschichchromatographie (TLC) in 125 ug t1δ: 0, 300 ug t18: 1 ^{trans 8} ^'9 and 100 ug _{t18: 1} ^{cis 8,9} _separated although the regioisomers t1δ : 1 ^{trans 8} and t1δ: 1 ^{trans 9} or t1δ: 1 ^cis8 and t1δ: 1 ^{cιs9 were} not separated from each other by the AgN0 ₃ -TLC, the determination of the eis (Fig. 10B) and / rans stereoisomers (Fig 10A) by means of NMR. One (1D) and two-dimensional (2D) ¹ H-NMR experiments (COZY) of the purified t1δ: 1 ^traπs and t1δ: 1 ^cιs stereoisomers were carried out. The t18: 1 ^trans fraction showed non-distinct but characteristic multiplets for the olefinic protons

(-CH = CH-) at 5,036 ppm, which in the COZY experiment show cross signals to the α-methylene protons (1.δ79 ppm), which are characteristic of rraπs-olefinic protons (DJ Frost et al., The PMR analysis of non-conjugated alkenoic and alkynoic acids and esters, Chem. Phys. Lipis 15: 53-δö (1976)). In the t1δ: 1 ^cιs fraction these signals were ^shifted slightly into the higher field for the olefinic protons (δ, 27δ ppm), whereas the α-methylene protons were shifted into the lower field (1, 927 ppm), which in turn was characteristic of is the cis isomer (DJ Frost et al. (1976), see above). The ratio of all unsaturated phytosphingenin isomers can be determined from the relative proportions of the four derivatives. In the GC-MS analysis of the FAME (C10: 0 and C9: 0), obtained after oxidation of potassium permanganate, the two regioisomers had a ratio of 3: 1, which means that the delta-δ- is three times stronger than that of delta-9 -Desaturase activity speaks. The relative mass distributions of the stereoisomers in the AgN0 ₃ -TLC indicate a three-fold higher dominance of the rans isomer over the c / s desaturase activity. The heterologous expression of the s / d1 gene from A. thaliana in yeast led to the new formation of four phytosphingenin isomers t1δ: 1 ^8trans , t1δ: 1 ^9trans , t1δ: 1 ^8cis and t1δ: 1 ^9cis in a ratio of 9: 3: 3 :1.

The MS and NMR spectra are shown in Figs. 9 and 10.

Example 6: Cloning and transformation of s / d1 in Arabidopsis thaliana For heterologous overexpression of the s / d1 gene from ß. The 1δ02 bp PCR clone 1 was cloned into puc19 / Smal (see Example 1) and cut with the restriction enzymes BamHI and SacI. The resulting 1621 bp DNA fragment was alloyed in a sense orientation in the BamBI and Sacl-cut plant expression vector pBI121 from CLONTECH behind the constitutive 35S promoter (CaMV) instead of the GUS gene. The resulting plasmid pBIBδBn is shown in Fig. 11.

For a homologous antisense elimination of the s / d1 gene, the 1272 bp PCR clone 3 from A. thaliana was cloned into pudδ / Smal (see Example 2) and digested with BamHI and Sacl. The resulting 12δ3 bp DNA fragment was alloyed in the antisense orientation in the BamHI / SacI sites of the vector pBI121. The resulting plasmid pBIBδAt is shown in Fig. 12.

The plasmids pBIBδBn and pBIBδAt were transformed into E. coli XLIblue and a DNA maxipreparation was carried out with the plasmid Maxi Kit (Qiagen). The completeness and correct orientation of the PCR clones in pBIBδBn and pBIBδAt was checked on the one hand by various restriction digests (BamHI + Sacl, BamHI + EcoRI; Xbal + Sacl) and on the other hand by repeated partial sequencing the ABI PRISM Dye Terminator Kit (Perkin-Elmer) at T _m δO ° C and the following specific primers:

3δS promoter primer PR01: δ'-CACAATCCCACTATCCTT-3 '(forward) BN9: δ'-GGAGCTTTGTAAGAAGCAT-3' (reverse) and ATδ: 5'-GGAATCTCAATCGCGTGG-3 '(reverse).

The Agrobacterium tumefaciens strain GV3101 with the Ti helper plasmid pMP90 (C. Koncz et al., "The promotor of T _L -DNA gene δ controls the tissue-specific expression of chimaeric genes by a novel type of Agrobacterium binary vector", Mol. Gen Genet. 204: 363-396 (1986)) was each with one of the binary vectors pBI121 (control), pBIBδBn and pBIBδAt according to the method of H. Chen et al., "Enhanced Recovery of Transformation of Agrobacterium tumefaciens after Freeze-Thaw Transformation and Drug Selection ", BioTechniques 16: 664-669 (1994). The transformed bacteria were grown for 24 h at 28 ° C. in Luria-Bertani (LB) selective medium with 100 μg / ml rifampicin, 40 μg / ml gentamycin and 50 μg / ml kanamycin (Sigma). After the antibiotic selection on LB agar plates, positive single colonies were checked by means of PCR at T _m 50 ° C. with the following primer combinations: PR01 and BN6 for pBIB5Bn; PR01 and AT1 for pBIBδAt. The transformation of whole A. thaliana plants, ecotype Columbia (C24) was carried out in planta by / Agroöacter / wm-mediated vacuum infiltration (N. Bechthold et al. (1993), and modified by AF Bent et al. (1994), see above) )

Example 7: Cultivation and selection of the transgenic rafe / dojPS / s plants The A / rofeacter / um-infiltrated TO plants were grown in pots with uniform earth / sand (2: 1) in the climatic chamber under long-day conditions (16 h light) at 22 ° C day and 17 ° C night temperature and 60% humidity grown to maturity and the T1 seeds harvested together. The dried seeds were surface-sterilized with 4% NaOCI in 0.02% Triton ^® X-100, on selective nutrient medium (1x MS salts incl. Gamborg's Vit. Bδ, Sigma M6619, 0.0δ% MES-KOH pH δ, 8 , 1% sucrose, 0, δ% agar) plated with 60 μg / ml kanamycin. After a 2 d cold treatment at 4 ° C to synchronize the germination, the plates were transferred to the climatic chamber (see above). After 14-20 d, the kanamycin-resistant T1 plants were grown in Repotted soil, self-grown as single plants and the T2 seeds harvested. Again, as described above (V. Katavic et al., "In planta transformation of Arabidopsis thaliana", Mol. Gen Genet. 245: 363-370 (1994)), antibiotic-resistant T2 plants were selected, which were found in a Split the ratio of hetero- to homozygous plants from 2: 1.

Example δ: Analysis of the LCB from the sphingolipids of transgenic Arabidopsis plants

For the subsequent analyzes of the fatty acids and long-chain bases of the sphingolipids, mixed samples of whole T2 plants (leaves and roots) were used directly from the selective plates, which resulted from the self-cultivation of independent T1 plants.

After lipid extraction of whole T2 plants in CHCI ₃ / methanol / water (2: 1: 0.75), the total fatty acids as bromophenacyl ester derivatives were analyzed by RP-HPLC (HP Haschke et al. (1990), see above). Analysis of transgenic T2 plants that had been transformed with pBIBδBn and pBIBδAt, 19 d and 36 d after germination, showed no significant changes in the fatty acid pattern compared to wild type and control plants (+ pBI121) and corresponded to the fatty acid composition in Arabidopsis wild-type leaves (B. Lemieux et al., "Mutants of Arabidopsis with alterations in seed lipid fatty acid composition", Theor. Appl. Genet. 80: 234-240 (1990)). A thin layer chromatographic separation of the lipid extract in acetone / toluene / water (91: 30 : 6) also showed no significant changes in the ceramide and cerebroside content in T2 plants 37 d after germination.

The analysis of the long-chain bases (LCB) from sphingolipids was carried out as described in Example 4 with whole T2 plants (total sphingolipids) 19 d, 23 d and 34 d after germination, with lipid extracts (cerebroside and ceramides) 21 d and 23 d after germination and with lipid-extracted T2 plants (non-extractable sphingophosphoglycolipids) 26 d after germination. The LCB analyzes of the mixed samples showed that the content of eis- and frans-phytosphingenin (t1δ: 1) in the total sphingolipids from δO-δδ% in wild type or in T2 control plants (+ pBI121) to up to δ% in T2 plants with pBIBδAt (antisense) or pBIBδBn (sense) was reduced (Fig. 13). While the antisense effect in T2 plants with pBIBδBn (sense) is due to co-suppression effects (Review by J. Finnegan et al., "Transgenic Inactivation: Plants Fight Back!", BIO / TECHNOLOGY 12: 363-δδδ (1994)), an increase in the phytosphingenin content in A. thaliana by heterologous overexpression of the s / d7 gene from ß.napus was only possible by a maximum of 6-9% with the high initial values in the wild type (see above) the content and ratio of eis- and rrans-phytosphingenin varies, for example, with the plant species (Imai et al., see above), a heterologous overexpression of the s / d1 gene in other organisms would lead to a greater modification of the base composition Analyzes of the above subtractions showed that cerebrosides contain more c / s-phytosphingenin and sphingophosphoglycolipids contain significantly more arans-phytosphingenin

If the phytosphinine content in the cerebrosides is kept relatively constant, the antisense effect leads to a drastic reduction of these LCB species in the sphingophosphoglycolipids.

The chemical structure analysis of the phytosphinine in Arabidopsis thaliana was carried out by direct Ba (OH) ₂ hydrolysis of wild-type leaves and extraction of the free LCB from the hydrolyzate (see Example 4). The C1δ trihydroxy bases were isolated as described in Example 5 and analyzed accordingly. The NMR and MS analyzes showed that

in contrast to the transgenic yeast cells only contain the delta-δ-phytosphingenin (t1δ: 1 ^8traπs and t1δ: 1 ^8cis ). The HPLC analyzes of the LCB from \ rac »/ dops / s plants (FIG. Δ E) speak for a similar ratio of the stereoisomers (t1δ: 1 ^trans and t1δ: 1 ^cis ) of 3: 1, as is the case with the heterologous Expression of the s / d1 gene from A. thaliana is present in yeast.

Example 9: Substrate specificity

Enzymatic studies have shown that not the free long-chain bases, but the N-acylated bases (ceramides and complex sphingolipids) are desaturated (C. Michel et al., "Characterization of Ceramide Synthesis. A Dihydroceramide Desaturase introduces the 4,5-frat ? s-double bond of sphingosine at the level of dihydroceramide ", J. Biol. Chem. 272: 22432-22437 (1997); JK Kok et al.," Dihydroceramide Biology. Structure-specific metabolism and intracellular localization ", J. Biol. Chem. 272: 21126- 21136 (1997)).

Our results and the composition of the LCB species in spingolipids from plants (Imai et al., See above) show that the nucleic acids isolated here (s / d1) code for a sphingolipid desaturase which is responsible for the formation of the following eis (Z) - and trans (E) -delta-δ-unsaturated bases from D-ety / 7ro-2-amino-1,3-dihydroxy-octadecane (sphinganine, d1δ: 0), which for the most part also delta-4-desaturates or - is hydroxylated, results in plants:

D-etyf 7ro-2-amino-1,3-dihydroxy-δ-octadecene (δ-sphingenin, d1δ: 1 ⁸ )

De / yf / 7ro-2-amino-1,3-dihydroxy-4 (E), δ-octadecadiene (4t, δ-sphingadienine, d1δ: 1 ^4t ' ⁸ )

- D-erytf) ro-2-amino-1, 3,4-trihydroxy-3-octadecen (δ-phytosphingenin, t1δ: 1 ⁸ )

The expression of the plant s / d1 nucleic acids from Arabidopsis thaliana and Brassica napus in the yeast Saccharomyces cerevisiae leads to the production of novel ice (Z) and trans (E) unsaturated bases in this organism:

D-eAy # 7ro-2-amino-1, 3,4-trihydroxy-3-octadecen (δ-phytosphingenin, t1δ: 1 ⁸ )

De / yf] ro-2-amino-1, 3,4-trihydroxy-9-octadecen (9-phytosphingenin, t1δ: 1 ⁹ )

Desaturation of D-eryf /? Ro-2-amino-1, 3-dihydroxy-octadecane (sphinganine, d1δ: 0) and subsequent delta-4-hydroxylation (Haak et al. (1997), see above) leads to the transgenic Yeasts for the plant-typical δ-phytosphingenin, while the desaturation of the delta-4-hydroxylated D-eryf / 7ro-2-amino-1, 3,4-trihydroxy-octadecane (phytosphinganine, t1δ: 0) is also possible and thus for formation of 9-phytosphinine, which does not occur in nature. It follows that the plant enzyme introduces the new double bond at a distance of Δ-6 from the last oxygen-activated carbon atom of the LCB (hydroxy group at C3 of sphinganine or at C4 of phytosphinganine) (Fig. 14). Transgenic T2 plants that had been transformed with pBIBδBn, pBIBδAt or pBI121 showed the unchanged fatty acid pattern of the wild type, which has no γ-linolenic acid and therefore also no delta-6-desaturase activity (Lemieux et al. (1990) , see above). As for the transgenic plants under example δ. described, total fatty acid analyzes were carried out on induced yeast cells which had been transformed with pYES2Bn, pYES2At or pYES2 (see Example 3). The analyzes of the three cultures showed that the fatty acid pattern was identical to that of the wild type, whereby delta-12 desaturation to hexadecadiene and linoleic acid could be excluded. Transformed and wild-type yeasts that incorporate exogenously applied 9, 12-linoleic acid into their glycerolipids showed no delta-1δ- or delta-6-desaturase activity that would have resulted in the formation of α- or γ-linolenic acid (Girke et al. "Identification of a novel Δ6-acyl-lipid desaturase by targeted gene disruption in Physcomitrella patens", Plant J., in press). The sphingolipid desaturase identified here clearly differs from a delta-6 c / s glycerolipid -Desaturase, with regard to its regional, stereo and substrate specificity.

Claims

claims

1. Sphingolipid desaturase which selectively introduces a double bond into the sphingobase of the ceramide residue of sphingolipids and of capnoids.

2. Sphingolipid desaturase according to claim 1, which introduces a double bond in position C ₆ -C after the last oxygen-activated carbon atom (Ci) of the sphingobase.

3. Sphingolipid desaturase according to claim 1 or 2, wherein the amino acid sequence of the desaturase comprises one or more of the fragments a) - c): a) HD [A, S] GH, b) HNAHH and c) QLEHH.

4. Sphingolipid desaturase according to claim 3, wherein the amino acid sequence of the desaturase comprises one or more of the fragments c) - e): c) QLEHH d) AYX _a X _b HD [A, S] GH, where X _a and X _b any Are amino acid residues, and e) THNAHH.

δ. Sphingolipid desaturase of claim 4, wherein the fragment d) comprises the amino acid sequence AYX _a GHDAGH, wherein X _a is any amino acid residue.

6. sphingolipid desaturase according to claim 3, 4 or 5, wherein the amino acid sequence of the desaturase comprises all three fragments a) - c) or c) - e).

7. sphingolipid desaturase according to one or more of claims 3 to 6, wherein the amino acid sequence of the desaturase further comprises the amino acid sequence AWWKWT and / or FGGLQF.

δ. Sphingolipid desaturase according to one or more of claims 1 to 7, wherein the desaturase comprises one of the amino acid sequences shown in Fig. 2 or 4.

9. DNA sequence coding for a sphingolipid desaturase according to one or more of claims 1 to δ.

10. DNA sequence according to claim 9, which is obtainable from Arabidopsis thaliana, Brassica napus.

11. DNA sequence according to claim 9 or 10, which comprises the nucleotide sequences shown in Fig. 1 or 3.

12. Vector comprising a DNA sequence according to one or more of claims 9 to 11.

13. Vector according to claim 12, in which the DNA sequence is functionally linked to a promoter and optionally to a termination signal.

14. A cell transformed with a vector according to claim 12 or 13.

15. A cell according to claim 14 which is a plant, fungal, bacterial or animal cell.

16. A transgenic organism which comprises a) a DNA sequence according to one or more of claims 9 to 11, b) a vector according to claim 12 or 13 and / or c) can be generated from a cell according to claim 14 or 15.

17. A transgenic organism according to claim 16 which is a plant, fungus, bacterium or animal.

16. Plant or its progeny that can be generated from the plant cell according to claim 14 or 1δ.

19. The plant of claim 16 which is a crop.

20. A plant according to claim 1δ or 19, which has an altered delta-δ base pattern in its sphingolipids.

21. A method for producing plants which have an increased or reduced proportion or a changed cis / trans ratio of delta-δ-unsaturated long-chain bases, comprising a) transforming a plant cell with a DNA sequence according to one or more of claims 9 11 or with a vector according to claim 12 or 13 and b) the regeneration of a plant with an altered delta-δ base pattern in its sphingolipids.

22. The method of claim 21, wherein the plant is a crop.

23. The method according to claim 21 or 22, which a) to compensate for a lack of delta-δ-unsaturated long-chain bases in an organism b) to switch off the production of delta-δ-unsaturated bases, c) to produce plants with improved tolerance and resistance to salinity and / or ion stress or toxicity and / or drought and / or moisture and / or cold or frost and phytopathogenic microorganisms and d) for the production of plants with changed size growth and flowering time suitable is.

24. A method for producing sphingolipids and capnoids with unsaturated sphingobase, comprising culturing cells according to claim 14 or 16, transgenic organism according to claim 16 or 17 or plants according to one or more of claims 1δ to 21.

2δ. The method of claim 24, wherein the sphingolipid or capnoid produced has the formula (I)

has, wherein

R ^{1 is} -OH, -OPO ₃ HR ⁶ , a carbohydrate residue or -S0 ₃ H;

R ^{4 is} -H or a -CC ₅ -alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more methyl radicals;

(a) when R ³ and R ^{5 are} -H, a is a double bond, and

(b) when R ^{3 is} -OH and R ^{5 is} -H, a or b is a double bond.

26. The method according to claim 2δ, wherein R ^{3 is} a Cιo-C ₂ alkyl radical and R ^{4 is} a C ₃ -C ₁ alkyl radical.

27. The method according to claim 2δ, wherein the sphingolipid or capnoid produced has the formula (II)

has, wherein R ¹ , R ² , R ³ , a and b have the meaning given in claim 2δ.

2δ. The method of claim 2δ, wherein the sphingolipid or capnoid produced is a sphingobase selected from:

D-e / y # 7ro-2-amino-1, 3,4-trihydroxy- (δZ) -octadecene (t18: 1 8cv)

De / y? Ro-2-amino-1, 3,4-trihydroxy- (8E) -octadecene (t1δ: 1 8 ° t ^l χ)

De / y / - 7ro-2-amino-1, 3,4-trihydroxy- (9Z) -octadecen (t1 δ: 1 9 ^a c ^{ς <} )

D-eryr7 / "θ-2-amino-1, 3,4-trihydroxy- (9E) octadecene

contains.

29. Sphinglipid or capnoid with unsaturated sphingobase with the formula (I):

(l).

wherein R ^{1 is} -OH, -OP0 ₃ HR ⁶ , a carbohydrate residue or -S0 ₃ H;

R ^{2 is} a C 1 -C ₃₃ -alkyl radical which can be mono- or polyunsaturated and / or substituted with one or more hydroxyl radicals;

(a) when R ³ and R ^{5 are} -H, a is a double bond,

(b) when R ^{3 is} -OH and R ^{5 is} -H, a or b is a double bond, and

(c) when R ^{4 is} C ₇ , C ₉ or Cn alkyl, a is not a double bond, and derivatives thereof.

30. Sphingolipid or capnoid according to claim 29 with the formula (III):

(III),

wherein R ¹ , R ² , R ³ , R ⁴ , R ⁵ , b and c have the meaning given in claim 29.

31. A cosmetic, pharmaceutical or food or chemical raw material comprising

(a) a sphingolipid or capnoid or produced according to claims 24 to 23

(b) a sphingolipid or capnoid according to claim 29 or 30.