CA2374015A1 - Gene isolated from ricinus communis encoding a new protein that interacts with the oleate 12-hydroxylase enzyme - Google Patents

Abstract

A description is provided of the isolation and characterization of a gene isolated from Ricinus communis which encodes for a protein capable of interacting with the oleate 12-hydroxylase enzyme that catalyzes the introduction of a hydroxyl group in the molecule of oleic acid (18:1.DELTA.9 ) transforming it into ricinoleic acid (12-OH, 18:1.DELTA.9).

Description

GENE ISOLATED FROM RICINUS COMMUNIS ENCODING A NEW PRO-ZYME.
The present invention relates to the identifica-tion and characterization of a gene of Ricinus communis (R.communis) which encodes for a protein capable of in teracting with the oleate 12-hydroxylase enzyme that catalyzes the introduction of a hydroxyl group into the molecule of oleic acid transforming it into ricinoleic acid.
The invention also relates to means and methods for producing transgenic plants with a modified compo-sition of fatty acids.
Ricinoleic (12-hydroxy-9-octadecenoic) acid is a monohydroxylated fatty acid whose only commercial source is seed oil synthesized in the endosperm of ripe seeds of R.communis, where it represents about 90% of hydroxylated fatty acids.
Studies in vivo with radioactive tracers indicate that, in the endosperm of unripe seeds of R.communis, ricinoleic acid (also known with the term ricinoleate) is synthesized by the direct substitution of a double bond of oleic acid with a hydroxyl group (Morris, L.J.
1967, Biochem. Biophys. Res. Commun. 29, 311-315). This reaction is catalyzed by the oleate 12-hydroxylase en-zyme whose activity seems to be associated with the en-doplasmic reticulum.
Enzymatic tests indicate that the substrate for oleate 12-hydroxylase is the oleic acid esterified with lecithin or with another phospholipid; the esterified ricinoleate is released from the lipid complex owing to the intervention of a phospholipase A, specific for fatty acids oxygenated in the presence of molecular oxygen, NAD(P)H and cytochrome b5. NAD(P)H is required for reducing the cytochrome b5. intermediate electron donor for the hydroxylase reaction (Bafor M. Et al., (1991), Biochem J., 280, 507-514; Smith M.A. et al., (1992), Biochem J, 287, 141-144). The hydroxylated fatty acid is then transferred, by means of the Kennedy pathway, to the pathway of triacylglycerols where it accumulates.
Ricinoleic acid, owing to the presence of the hy-droxyl group, is one of the most versatile natural products and has numerous industrial and food applica-tions. In particular, ricinoleic acid can be used in the production of paints, polymers such as nylon-11, drugs, lubricants, cosmetics, resins and other materi-als.

The production of ricinoleic acid however is lim-ited by the high susceptibility to climatic variations of the cultivations of R.communis plants and by the toxicity of ricin, an allergen present in castor beans.
The possibility of producing ricinoleic acid in vegetable species which are more tolerant towards cli-matic variations and which do not contain toxic sub-stances would allow a larger and simpler production and application of the acid itself.
For this purpose, the gene that encodes the oleate 12-hydroxylase enzyme has recently been isolated and used to transform plants such as Nicotiana tabacum, Arabidopsis thaliana, Linum usitatissimum and Brassica napus.
In all these cases, however, although a modified content of fatty acids was observed, the increase in ricinoleic acid was low if not zero (Broun P. and Somerville C., 1997, Plant Physiol, 113, 933-942).
It is known that in many biological processes, such as replication, transcription or metabolism, enzy matic complexes, whose action is correlated to the co operation of various proteic subunits, intervene.
For example, evidence of the possible interaction of various proteins in desaturation processes was ob-tamed from studies on the activity of the toluene 2-mono-oxygenase enzyme isolated from the bacterium Burk-holderia cepacia (Newman L.M, et al., 1995, Biochemis-try, 34, 14066-14076).
It was therefore assumed that the hydroxylase of vegetable fatty acids can also form part of a multicom ponent system and that the hydroxylase activity of the oleate 12-hydroxylase enzyme of R.communis consequently requires the intervention of further co-factors or pro teins.
A gene of R.communis which encodes for a new pro-tein capable of interacting with the oleate 12-hydroxylase enzyme, has now been identified and charac-terized. This gene can be used in genetic transforma-tion programs of plants containing the oleate 12-hydroxylase enzyme to favour the production of ricino-leic acid.
In accordance with this, an objective of the pres ent invention is the cloned and sequenced gene which encodes a protein capable of interacting with oleate 12-hydroxylase.
A second objective of the present invention is an expression recombinant vector in host cells comprising said gene.

A further objective of the present invention is a host microorganism transformed with said vector.
Yet another objective of the present invention re-lates to transgenic plants transformed with said vec-5 tor.
Additional objectives of the present invention will appear evident upon reading the description and examples.
Brief description of the figures Figure 1: Southern Blot of the genomic DNA of different species digested with the restriction enzyme EcoRI and hybridized with the 762 by fragment of the plasmid pTargl marked with 32P. The hybridization signal corre-sponds to a gene in a single copy only evident in R.communis.
Figure 2: Northern blot of messenger RNA extracted at various development stages of the seed (10, 20, 30, 35 and 40 DAP), from the leaves, stem and roots of the R.communis plant. The filter was hybridized with the 762 by fragment of the plasmid pTargl marked with 32P.
The presence of an mRNA with a molecular weight of about 1.0 Kb, is observed in unripe seeds at 10 DAP and 20 DAP and a transcript with larger dimensions in the leaves.

Figure 3: Northern blot of messenger RNA extracted at different development stages of the seed (10, 20, 30, 35 and 40 DAP) , from the leaves, stem and roots of the R.communis plant. The filter was hybridized with the 1216 by fragment of oleate 12-hydroxylase marked with s2P. The hybridization signal, of about 1.6 Kb, is pres-ent in the unripe seeds at 20, 30, 35 and 40 DAP.
Detailed description of the invention The isolation of nucleotide sequences which encode for proteins of interest can be carried out by known techniques.
In particular, to isolate new proteins that inter-act with oleate 12-hydroxylase in unripe seeds of R.communis, the "HybridZap two hybrid vector" of Stratagene was used, a eukaryotic system (Saccharomy-ces cerevisiae) which enables new genes to be identi-fled in vivo, that encode proteins which interact with a known protein (Fields S. et al., 1989, Nature, 340, 245-246). This system exploits the characteristics of the transcriptional activator GAL4 of S.cerevisiae, which regulates the expression of genes that encode en-zymes involved in the galactose metabolism.
GAL4 consists of two domains separable and func-tionally essential for its activity; an N-terminal do-main (Binding Domain, BD), which is linked to specific sequences of the DNA (UAS: upstream activating se quences), and a C-terminal domain, containing acid re gions (Activation Domain, AD), which is necessary for the transcriptional activation.
The system used allows two hybrid proteins to be generated, containing the functional domains of GAL4, i.e. the Binding Domain fused with a known protein which acts as bait, and the Activation Domain fused with unknown proteins (target) from an expression li-brary.
If the known protein interacts with a target pro-tein forming a protein-protein complex, the two func-tional domains of GAL4 are brought under optimum condi-tions and activate the transcription of the reporter gene lac-Z, whose product is shown by means of colori-metric reaction.
The total RNA was extracted from a pool of unripe seeds of R.communis using the "Hot-Phenol" method and the polyadenilate messenger RNA (mRNA) was isolated with oligo-dT columns (Pharmacia). The cDNA encoding the oleate 12-hydroxylase enzyme was subsequently pre-pared by applying the polymerase chain reaction (PCR) technique on the mRNA using, as primers, a pair of oli-gonucleotides having the sequences that flank the en-coding region comprising the start and stop translation codon of said gene.
As the interaction of target proteins can take place with the whole protein under examination or with parts of it, it was decided to clone both the whole gene encoding for oleate 12-hydroxylase and its termi nal regions 5' and 3', in the plasmid pBD-GAL4, which contains the sequence that encodes for the Binding Do main.
The three DNA inserts were first amplified with the appropriate primers having restriction sites EcoRI
for the Forward primer and SalI for the Reverse primer, digested with the above enzymes, directionally inserted into the vector pBD-GALA4 predigested with the same en-zymes and introduced into the competent cells Epicurean coli XL1-Blue.
The recombinant clones, containing the expected fragments, were characterized by means of restriction analysis and their identity was confirmed with the se quence reactions carried out using the Taq Dye Deoxy Terminator Cycle Sequencing kit (Perkin Eimer) and ana-lyzed with the automatic sequencer. A cDNA library was then prepared from unripe seeds of R.communis in the HybridZap phage vector which expresses consistent hy-brid proteins of the activation domain of GALA4 and proteins of R. communis .
In practice, the polyadenilate mRNA of R.communis was used for the synthesis of double filament cDNA op erating according to the protocols suggested by the Kit distributor (Stratagene).
The molecules of cDNA having a high molecular weight, useful for the construction of the library, were then separated from those having a low molecular weight, which represent the fraction of molecules in which the synthesis is incomplete.
The fraction of high molecular weight cDNA was in-serted into the ~. HybridZap phage vector and packed with the packaging extracts containing proteins for the head and tail of the phage. The dimensions of the in-serts present in the library produced were checked by means of PCR and amplified with a specific pair of primers for the vector pAD-GAL4. The results obtained demonstrated that the fragments of the cDNA library had an average dimension of 1.4 Kb.
After amplification, the library in the lambda phage was converted to a plasmid library by excision in vivo.

The library was subsequently multiplied in the strain of E.coli XLOR and the plasmid DNA was extracted and co-transformed, in separate co-transformation proc-esses, with the DNA extracted from the bait plasmids 5 containing the whole gene and parts of oleate 12-hydroxylase, using yeast cells (YRG-2 strain), having reporter genes his3 and lacZ.
From the test of the three sequences of oleate 12 hydroxylase with the expression library, colonies of 10 yeast were identified which had the typical blue colour of the lacZ gene activity, thus demonstrating the prob-able interaction of the bait protein with an unknown target protein.
Subsequent analyses on these colonies enabled a "positive" co-transformed yeast clone to be identified, which activated the transcription of both reporter genes. This indicated the complete interaction between the N-terminal region of oleate 12-hydroxylase and an unknown Target protein.
The plasmid isolated from this positive yeast clone was indicated with the abbreviation pTargl.
The interaction specificity between oleate 12-hydroxylase and the new protein identified, was con-firmed by co-transformation experiments in S.cerevisiae YRG-2 yeast cells with bait plasmids containing the whole oleate 12-hydroxylase gene and the 5' terminal portion of this gene, respectively. Two yeast colonies were identified from tests of the two sequences, one for each co-transformation process, which had the typi-cal blue colour of the lacZ gene activity.
To confirm the presence and dimensions of the bait proteins and to verify the presence and dimensions of the gene encoding the protein Targl2 identified, PCR
analyses were carried out on the DNA extracted from the yeast clones resulting positive from the expression test of the reporter gene lacZ, using specific primers for the Binding Domain region and for the Activation Domain region.
The authenticity of the amplification products ob-tamed was demonstrated not only by the size of the ex-pected fragments, but also by the hybridization analy-sis carried out on the latter.
In addition, the plasmid DNA of pTargl was iso lated from the yeast colony and the cDNA insert was pu rified with the "Double GeneClean" kit (BIO 101 Inc., U.S.A.). The purified fragment was subsequently cloned in the vector pGEM-T (Promega) and then introduced into competent cells of E.coli DHSa.

The plasmid DNA isolated from the recombinant clones was subjected to sequence analysis with a Taq Dye Deoxy Terminator Cycle Sequencing Kit (Perkin Elmer), using the automatic sequencer ABI 373A (Perkin Elmer).
From the sequence analyses carried out on the DNA
of the insert of the plasmid pTargl, it can be observed that the fragment isolated has a dimension of 762 by and contains an Open Reading Frame (ORF) of 540 by pre-ceded by 75 by at 5' and followed by 147 by at 3'. The ORF encodes a protein of 180 aminoacids with a molecu-lar weight of 19.8 KDa.
The nucleotide and aminoacid sequences were com pared with the sequences available on data banks by means of FASTA and BLAST analyses, and homologous se quences were not found, indicating the uniqueness of the DNA tract of R.communis and the uniqueness of the protein identified.
To verify the identity of the protein capable of interacting with oleate 12-hydroxylase, analyses were carried out on the genomic DNA of different species.
The results showed the presence of a signal only in R.communis. This demonstrates that the gene isolated is specific of the genome of R.communis and is not an out come of the system adopted for its identification.
In addition, expression analyses were carried out on the messenger RNA extracted at different development stages of the seed (10, 20, 30, 35 and 40 days after pollination), from the leaves, stem and roots of the R. communis plant .
The results of the hybridization of the Northern blot with the fragment of pTargl marked with 32P showed the gene expression in the leaves and unripe seeds 10 and 20 days after pollination.
The same filter was hybridized with the fragment of oleate 12-hydroxylase which begins to be expressed in the unripe seeds at 20 DAP, where the signal is very weak, subsequently increasing its expression in the stages at 30, 35 and 40 DAP. As expected, there was no hybridization signal in the samples of RNA correspond-ing to leaves, stem and roots.
On the basis of these results, it can be concluded that the new protein most probably intervenes in the first development stages of the R.communis seed at the beginning of the synthesis of ricinoleic acid, carrying out a regulating action. The gene of the present inven-tion can be cloned in an expression vector in plants, by putting it under the control of suitable regulation sequences (promoter and terminator).
Vectors suitable for the purposes of the present invention are for example those deriving from the Ti plasmid of Agrobacterium tumefaciens as described by Bevan M., (1984), Nucleic Acid Research 12: 8711-8721.
These vectors are used to transform the plants by means of conventional methods. The method described by G. An et al. (Binary vectors, Plant Molecular Biology Manual A3, Kluwer Academic, Dordrecht, pages 1-19, 1988), which is based on the capacity of Agrobacterium tumefaciens to transfer part of its own DNA to vegeta-ble cells, is preferably used.
The plasmid pTargl containing the gene of the pre-sent invention was deposited as E.coli DHSa/MA292 at the CentraalBureau Voor Schimmelcultures where it re ceived the deposit number CBS 101642.
The following examples, whose sole purpose is to provide a more detailed description of the present in vention, should in no way be considered as limiting the scope of the invention itself.
Example 1 RNA Isolation The total RNA was extracted from a pool of unripe seeds (10-40 days after pollination) of Ricinus commu-nis by means of the "Hot-Phenol" method described by Shirzadegan M. et al. (1991), Nucl. Acids Res.: 19, 6055, to which several modifications were made.
In short, 2 g of vegetable material were crushed 5 in liquid nitrogen and then suspended in 6 ml of ex traction buffer (0.1 M LiCl, 0.1 M Tris-HC1 pH 7.6, 0.01 M EDTA, to Sodium dodecylsulfate (SDS), phenol) preheated to 80°C. The suspension was incubated at 80°C
for 5 minutes and 3 ml of a mixture of chloro 10 form/isoamyl (24:1, v/v) were added. The sample was vortex mixed and subsequently centrifuged at 12,000 rpm, at 4°C for 15 minutes. The aqueous phase was re-covered, subjected to an additional extraction cycle with phenol/chloroform/isoamyl (25:24:1) and centri-15 fuged under the same conditions specified above. The supernatant was recovered and the RNA precipitated by the addition of a volume of LiCl 4 M. The sample was incubated at -20°C for a night and then centrifuged at 13,000 rpm, at 4°C for 30 minutes.
The RNA pellet was re-suspended in water, trans-ferred to microcentrifuge tubes and precipitated by the addition of LiCl 4 M and 0.2 volumes of EDTA 0.5 M. Af-ter centrifugation at 13,000 rpm, at 4°C for 30 min-utes, the pellet was suspended again in water and pre-cipitated with 0.1 volumes of NaCl 5 M and 2.5 volumes of ethanol 1000. After centrifugation at 15,000 rpm, at 4°C for 30 minutes, the pellet was recovered, washed twice with ethanol 700, dried and re-suspended in wa-ter.
Example 2 Isolation of the cDNA encoding for oleate 12-hydroxylase The polyadenilate messenger RNA (poliA-RNA) was prepared from the total RNA obtained in example 1, us ing oligo-dT columns (Pharmacia) according to the in structions of the distributor.
3.5 ~g of polyadenilate messenger RNA were then used for the synthesis of double filament cDNA using the Kit distributed by PHARMACIA, operating under the experimental conditions suggested by the supplier of the kit.
On the basis of the sequence of oleate 12-hydroxylase deposited in the data bank (GenBank, AC
U22378) the following oligonucleotides were synthe-sized:
(1) 5'GGA TCC CTC AGG AAA GTG CTT A 3' (FORWARD) (2) 5' TCT AGA CAT TCC TTC TTG TTC TAA TT3' (REVERSE) These oligonucleotides, which correspond to the regions that flank the portion encoding the enzyme com prising the start and stop translation codon, were used as primers for the isolation of the fragment corre sponding to oleate 12-hydroxylase by means of the poly merase chain reaction (PCR) technique.
The amplification was effected in a DNA Thermal Cycler 480 apparatus (Perkin Elmer Cetus) using a reac-tion mixture (25 ~1) containing 6 ~1 of double filament cDNA, 10 mM Tris HC1 pH 8.3, 1.5 mM MgClz, 50 mM KCl, 2 . 5 ~M of each primer, 0 . 1 mM of dNTP and ?_ . 5 polymer-ase Taq Units (Boheringer).
After a first denaturation cycle for 5 minutes at 95°C, the reaction was continued with the following cy-cles:
1 minute at 94°C (denaturation) 1 minute at 56°C (pairing) 2 minutes at 72°C (lengthening) for a total of 35 cycles, followed by 10 minutes at 72°C (final extension).
The amplification product, corresponding to a fragment of about 1200 base couples, was separated on agarose gel at 1.0o, the DNA band of interest was re-covered and purified with the GeneCleanTM kit (BIO 101 Inc, U.S.A.). About 100 ng of the DNA thus isolated were ligated to 50 ng of pGEM-T plasmid (Promega) in 10 ~1 of reaction mixture, in the presence of 2 units of T4 DNA ligase, at 4°C for a night.
~1 of this mixture were used to transform compe-5 tent cells of E.coli DH5a (BRL). The transforming agents were selected on plates of LB medium (NaCl 10 g/1, Yeast extract 5 g/1, Bacto-triptone 10 g/1 and agar 20 g/1) containing 50 ~tg/ml of ampicillin.
The plasmid DNA extracted from 6 positive clones was subjected to sequence analysis to verify the nu-cleotidic correspondence with the gene of oleate 12-hydroxylase isolated by Van de Loo F. Et al., 1995, PNAS, 92, 6743-6747. The reactions and sequence analy-ses were carried out with the Taq Dye Deoxy Terminator Cycle SequencingTM kit (AB-PEC) using an ABI Prism 373A
DNA Sequencer (AB-PEC).
One of the plasmids analyzed, containing a frag-ment of DNA analogous to the published sequence SEQ:ID
Nr. l, was called pCl8-MA.
Example 3 Construction of the "bait" vectors (pBD-GALS
As the interaction of target proteins can take place with the whole protein under examination (bait) or parts of this, it was decided to clone both the fragment corresponding to the whole gene encoding for oleate 12-hydroxylase, and also the fragments corre-sponding to the regions 5' and 3' of said gene, in the plasmid pBD-GAL4 (Stratagene).
For this purpose four oligonucleotides were syn-thesized, of which the Forward primers (abbreviated as F) have the restriction site EcoRI and the Reverse primers (abbreviated as R) the site SalI. The nucleo-tide sequences of the primers are as follows:
(a) 5'GAA TTC CGC ATG TCT ACT GTC 3' (Forward, HydGal-F) (b) 5'GTC GAC CAT TCC TTC TTG TTC 3' (Reverse, HydGal-R) (c) 5'GTC GAC GCG ATC GTA AGG 3' (GalHydi-R) and (d) 5'GAA TTC AAT GTC TCT GGT AGA C 3' (GalHydi-F) The primers HydGal-F and HydGal-R were used to am-plify the whole gene of oleate 12-hydroxylase.
The primers HydGal-F/GalHydi-R and HydGal R/GalHydi-F were used to amplify the region 5' of 624 by (SEQ: ID Nr:2) and the region 3' of 633 by (SEQ: ID
Nr:3) of the gene of oleate 12-hydroxylase, respec-tively.
The amplifications with the above primers were carried out on 20 ng of the fragment of oleate 12-hydroxylase previously cloned and sequenced.
The amplification products having the expected di-mensions were digested with 10 units of restriction en-zymes EcoRI and SalI (Boheringher), separated on aga-5 rose gel to and the fragments of DNA of interest were then recovered and purified with the GeneCleanTM kit (BIO 101 Inc.).
About 100 ng of each fragment were ligated, sepa rately, with the plasmid pBD-GAL4 linearized with the 10 enzymes EcoRI and SalI, in 10 ml of reaction mixture, in the presence of 2 units of T4 DNA ligase, at 4°C for a night. The ligase mixtures were used to transform competent cells of Epicurian coli XL1 Blue (Strata-gene). The recombinant clones were selected on plates 15 of LB medium to which 30 ~g/ml of chloramphenicol had been added. The following recombinant "bait" plasmids were identified, which contain a fragment which con-sists of the Binding Domain of Gal4 condensed with:
(a) the whole gene of oleate 12-hydroxylase (pBD-20 GALA4/C18)~
(b) the 5-terminal region of the gene of oleate 12-hydroxylase (pBD-GALA4/C18-5); and (c) the 3-terminal region of the gene of oleate 12-hydroxylase (pBD-GALA4/C18-3).

The plasmids were characterized by restriction analysis and their identity was confirmed by effecting sequence reactions with the Taq Dye Deoxy Terminator Cycle sequencing kit (Perkin Elmer) and analyzed with the automatic sequencer ABI 373A (Perkin Elmer).
Example 4 Construction of the target librar For the construction of a "target" library made up of hybrid proteins consisting of the activation domain of GALA4 and proteins of unripe seeds of R.communis, the HybridZap phage vector was used. The experimental conditions adopted were those suggested by the supplier of the kit (Stratagene, HybridZapTM Two-Hybrid cDNA
gapack cloning kit, catalogue Nr.: 235612).
About 5 ~g of polyadenilate mRNA of R.communis were used for the synthesis of the double filament cDNA.
The ends of the cDNA molecules, to which 3.6 ~.g of a linker having the restriction site EcoRI had been added, were then blunted by the action of the polymerase DNA
Pfu (Stratagene), and subjected to digestion with the enzyme XhoI (120 units), whose site is present in the primer polydT used for the synthesis of the first cDNA

filament. This gave rise to molecules having the EcoRI
site at one end and the XhoI site at the other end.
The sample of cDNA was then passed on a Sephacryl S-500 column equilibrated in 20 mM Tris-HC1 pH 7.5, 10 mM EDTA, 100 mM NaCl, and subjected to centrifugation for 2 minutes at 400 revs.
Three fractions were recovered, whose molecular weight was verified by means of electrophoresis on non-denaturing polyacrylamide gel at 50 (Sambrook, J. Et.
A1., (1989), Cold Spring Harbor Laboratory Press).
About 100 ng of the first fraction of cDNA, corre-sponding to that with a high molecular weight, were 1i-gated with 1 ~g of the ~ HybridZap phage vector predi-Bested with the enzymes EcoRI and XhoI and packed with packaging extracts containing proteins for the head and tail of the phage. The total quantity of phage parti-cles obtained by the packaging in vitro was determined by plating aliquots with the strain XL1-Blue MRF'.
The primary library obtained contains a total of 1.4 x 106 plaque forming units (pfu) per ~tg of ligated vector arms and 97~ of these contains the DNA insert.
The dimensions of the library produced were verified by subjecting to PCR reaction, 20 phage plaques, selected at random, and amplified with the specific pair of primers for the vector pAD-GAL4:
a) 5'AD primer: 5' AGG GAT GTT TAA TAC CAC TAC3' b) 3'AD primer: 5' GCA CAG TTG AAG TGA ACT TGC3' The results obtained showed that the inserts of the cDNA library had an average dimension of 1.4 kb.
Example 5 Conversion of the HybridZap phagic library to the pAD-GALA4 plasmid library The primary library constructed in the ~. HybridZap phage was converted to the pAD-GAL4 plasmid library by means of total excision in vivo according to the method described by the supplier of the kit adopted (Strata-gene ) .
For the amplification of the primary library, 1x106 phages were used to infect the host cells XL1 Blue MRF', which, by enabling replication inside the phages, after their lysis, allowed a library consisting of about 1x109 phagic particles to be recovered.
An aliquot of the amplified library (1x108 pfu) was then incubated with 1x101° pfu of ExAssist helper phage and 1x109 of XL1 Blue MRF' cells to generate particles of phagemid containing the plasmid vector excised from the phage vector.
The excess number of phage helpers and bacterial cells with respect to the number of phages of the li-brary was used to ensure that each cell was infected both by the phage helpers and by the lambda phage, thus obtaining an effective and representative excision in vivo.
The incubation of the lambda phages with the phage helpers and XL1 Blue MRF' cells took place at 37 °C for minutes, after which 20 ml of LB medium were added and the incubation was continued at 37°C for a further 10 3 hours.
In order to lyse the phage particles and enable the particles of phagemid to be released and recovered, the suspension was incubated at 70°C for 20 minutes and then centrifuged for 10 minutes at 500 revs. The super-15 natant (phagemid stock) was recovered and conserved at 4°C.
Cells of E.coli XLOR were incubated with 1x10$ of phagemid in a ratio of 10:1, at 37°C for 15 minutes.
500 ml of LB medium containing 50 ~g/ml of ampicillin were then added and the incubation was continued at 37°C for 3 hours.
This operation allowed cells to be obtained, con-taming the target library in a plasmid vector.
The suspension was centrifuged for 10 minutes at 500 revs and the plasmidic DNA was isolated from the pellet according to the protocol of alkaline lysis sug-Bested by Sambrook, J. Et al., 1989, Cold Spring Harbor laboratory Press.
5 Example 6 Screening of the library To identify the "target" proteins which interact with the "bait" protein, the yeast strain S.cerevisiae YRG-2, containing the reporter genes his3 and lacZ, was 10 co-transformed with the DNA prepared from the bait plasmids and the DNA isolated from the target plasmid library.
About 5 ~.g of DNA extracted from the plasmids pBD-GALA4/C18, pBD-GALA4/C18-5' and pBD-GALA4/C18-3', were 15 co-transformed, in separate co-transformation proc-esses, with 10 ~.g of DNA extracted from the target li-brary. The co-transformations were plated on SD selec-tive medium (yeast nitrogenated base without aminoacids 6.7 g/1, D-sorbitol 182.2 g/1, agar 20 g/1, 100 ml of 20 the appropriate aminoacidic solution concentrated lOx and glucose 2%) without leucine aminoacids (Leu ), tryp-tophan (Trp ) and histidine (His ), which allows only the growth of the yeast colonies containing both of the recombinant plasmids pBD-GALA4(Trp ) and pAD-GAL4 (Leu ), and incubated at 30°C for 5 days. The yeast colonies thus obtained were transferred to Watman 3MM filter pa-per and subjected to the expression test for the re-porter gene lacZ, by soaking the filters in a solution containing the chromogenic substrate 5-bromo-4-chloro-indolyl ~3-D-galactoside (X-Gal) and incubating at 30°C
for a night.
From the test of the 3 sequences of oleate 12-hydroxylase with the library, 7 yeast colonies were identified, having the typical blue colour of the ac-tivity of the gene lacZ, thus indicating the probable interaction of the bait protein with an unknown target protein. To verify the authenticity of the above colo-nies, these were re-plated on selective medium and the X-Gal test was repeated.
Of the 7 colonies identified, only one reconfirmed the previous phenotype, whereas the others proved to be false positives.
The positive yeast colony proved to come from the co-transformation process in which the plasmid pBD-GAL4/C18-5' was used, containing the 5' region of the oleate 12-hydroxylase gene.
Transformation control experiments were parallelly carried out with the 4 plasmids pGal4, p53, pSV40 and pLaminC (Stratagene). These plasmids were used singly or in pairs as indicated in table 1 and, on the basis of the combinations, they acted as positive or negative controls. Table 1 also indicates the results of the transformation processes in which the bait protein in-teracts with an unknown target protein.
Table 1 Transformations SD1 SD2 SD3 SD4 GROWTH

pGAL4 blue p53 white pSV40 white pLaminC white p53 and pSV40 blue blue pLaminC and pSV40 white 0 pBDGAL4-Bait white pBDGAL4-Target white pBDGAL4-Target pBDGAL4-Bait white white blue blue wherein SD1 - SD medium without Leu; SD2 - SD medium without Trp; SD3 - SD medium without Leu and Trp and SD4 - SD medium without Leu, Trp and HIs.
Example 7 Verification of the interaction specificit To verify the interaction specificity between ole-ate 12-hydroxylase and the new protein identified, co-transformation experiments were carried out in yeast in which the DNA of the target plasmid, called pTargl, was tested in combination with the whole gene of oleate 12-hydroxylase, with the 5' region and with the genes of the bait proteins p53 and pLaminC.
To isolate the plasmid DNA pTargl from the yeast colony resulting positive from the X-GAL test, the lat-ter was inoculated into 2 ml of YPAD growth medium (peptone 20 g/1, yeast extract 10 g/1, adenine sulfate 40 mg/1 and glucose 20) and incubated at 30°C for 2 days.
To recover the plasmid DNA the cellular wall of the yeast was mechanically broken using glass balls and the nucleic acids were precipitated following the pro-cedure described by Sambrook, J. Et al., 1989, Cold Spring Harbor Laboratory Press.
About 5 ~g of the plasmid pTargl were used in a co-transformation experiment with 5 ~g of DNA of the plasmid containing the whole oleate 12-hydroxylase gene (pBD-GAL4/C18), and in another experiment with 5 ~g of DNA of the plasmid containing 5-terminal portion of the oleate 12-hydroxylase gene (pBD-GAL4/C18-5'). 2 yeast colonies were identified from the tests of the two se quences, one for each co-transformation process, which had the typical blue colour of the activity of the lacZ
gene and thus indicating the interaction of the bait protein with an unknown target protein.
The above colonies were plated on selective SD me-dium (Leu , Trp , His ) and tested again with X-Gal. The two colonies identified reconfirmed the previous pheno-type.
When, on the other hand, the target protein was tested with the bait proteins p53 and pLaminC, as ex pected, none of the colonies had the blue colour with the X-Gal test. These results indicated that the new protein identified interacted specifically with oleate 12-hydroxylase.
To confirm the presence and dimensions of the pro-teins used as bait and to verify the presence and di-mensions of the fragment encoding for the target pro-tein, called TargHl2, PCR analyses were effected on the DNA extracted from the two yeast clones resulting posi-tive from the expression test of the reporter gene lacZ.
Specific primers for the Binding Domain (BD) re-gion and for the Activation Domain (AD) region were used for the amplification under the conditions sug-Bested by the kit supplier.
The sequences of the specific primers for the Ac-tivation Domain were specified in example 4, whereas 5 the sequences for the Binding Domain are the following:
a) 5'BD: 5'GTG CGA CAT CAT CAT CGG AAG3' b) 3'BD: 5'CCT AAG AGT CAC TTT AAA ATT3' The authenticity of the amplification products was demonstrated by the size of the expected fragments and 10 also by the hybridization analysis carried out on these.
Example 8 Sequence analysis of the plasmid Targ1 The plasmid DNA of pTargl was subjected to ampli-15 fication reaction with the use of specific primers of the plasmid pAD-Gal4, 5'AD primer and 3'AD primer. The fragment produced was purified with the GeneCleanTM Kit (BIO 101 Inc. U.S.A.). About 100 ng of the DNA thus pu-rified were ligated with 50 ng of the plasmid pGEM-T
20 (Promega) in 10 ~1 of reaction mixture, in the presence of 2 units of T4 DNA ligase, at 4°C for a night.
5 ~g of the ligase mixture were used to transform competent cells of E.coli DHSa (BRL). The transforming agents were selected on LB medium to which 50 ~g/ml of ampicillin had been added.
The plasmid DNA was extracted from 6 positive clones, i.e. showing a white colour, and sequenced with the Taq Dye Deoxy T Cycle SequencingTM Kit (AB-PEC), us ing an ABI Prism 373A DNA Sequencer (AB-PEC).
From the sequence analyses, it can be observed that the isolated gene of 762 by (SEQ: ID Nr:4) con-tains an Open reading frame (ORF) of 540 bp, preceded by 75 by at 5' and followed by 147 by at 3' where the poly (A) tail is present. The ORF encodes a protein of 180 aminoacids (SEQ: ID Nr:5), indicated as TargHl2, with a molecular weight of 19.8 kilodaltons.
Example 9 Southern Blot analysis To verify the identity of the protein capable of interacting with oleate 12-hydroxylase and the number of ricin copies of the gene corresponding to the insert of the plasmid pTargl, analyses were carried out on the genomic DNA of different species, isolated with the method of Dellaporta et al., 1983, Plant Mol. Biol.: 1, 19-21.
About 6 ~g of genomic DNA isolated respectively from:
Ricinus communis, Lesquerella fendleri, Linum usitatis-simum, Brassica napus, Helianthus annus, Limnantes douglasii, Lycopersicon esculentum, Beta vulgaris, Zea mays, Nicotiana tabacum and Saccharomyces cerevisiae, were digested with 100 units of the enzyme EcoRI (Boe-hringer) in 100 ~1 of reaction mixture, at 37°C for 1 hour.
The digestion mixtures were charged onto agarose gel at 0.8o and subjected to horizontal electrophore-sis. The DNA was transferred onto nitrocellulose filter (Hybond-N(+)R, Hamersham) with the Southern method (Sam-brook, J. Et al., Cold Spring Harbor Laboratory Press).
This filter was hybridized for a night at 65°C, after a pre-hybridization of about 4 hours at 65°C, in a solution containing 6XSSC, to SDS, lOX Denhardt's and tRNA at a concentration of 10 ~g per ml of hybridiza-tion solution used. The fragment of 762 by isolated from the plasmid pTargl, marked with 3'P (Sambrook and Fritsch, E.F. and Maniatis, T. (1989), Molecular Clon-ing: A Laboratory Manual, (Cold Spring Harbor Lab., Cold Spring Harbor, NY) 2nd Ed. 1989), was used as probe.
The filter was washed twice at 65°C in 2XSSC, 0.20 SDS for 30 minutes and once in 0.2XSSC, 0.020 SDS for 20 minutes, then exposed to X-rays by autoradiography.

The hybridization showed a single band only on the genomic DNA isolated from R.communis. This demonstrated that the gene isolated was specific of the genome of this plant, which was present in a single copy and that the result obtained was not an outcome of the system used for its identification (figure 1).
Example 10 Expression analysis To effect the expression analysis of the gene en-coding the protein TargHl2, the messenger RNA was pre-pared from different organs (leaves, stem and roots) of R.communis and at different development stages of the seed (10-20-30-35-40 days after pollination, DAP).
About 7 ~g of each sample were charged onto aga-rose gel at 1.4o containing formaldehyde and subjected to an electrophoretic run. The RNA were transferred onto a nitrocellulose filter (Hybond-N(+)R Hamersham) using the standard Northern blot procedure. The filter was hybridized with a probe corresponding to the frag-ment of 762 by of the plasmid pTargl marked with 32P.
The reaction was carried out in a hybridization solu-tion containing 6XSSC, loSDS, lOX Denhardt's and 10 ~.g of tRNA, at 65°C for a night, after a pre-hybridization of 5 hours in the same solution at 65°C.

The filter was washed twice at 65°C in 2xSSC, 0.2oSDS for 20 minutes and once in 0.2XSSC, 0.02oSDS at 65°C for 20 minutes and then exposed to X-rays by auto-radiography.
The results showed the presence of a signal, lo-calized in a band of about 1 Kb (figure 2), in the sam-ples corresponding to unripe seeds of 10 and 20 DAP. A
similar signal was observed in the sample of RNA ex-tracted from the leaves.
The same Northern filter was hybridized with the fragment of oleate 12-hydroxylase which begins to ex-press itself in the unripe seeds at 20 DAP, where the signal is very weak, subsequently increasing its ex-pression in the stages at 30, 35 and 40 DAP (figure 3).
No hybridization signal was observed in the samples of RNA corresponding to the leaves, stem and roots.
From these results it can be deduced that the pro-tein isolated with the double hybrid system most proba-bly intervenes in the first stages of development of the seed of R.communis, at the beginning of the synthe-sis of ricinoleic acid.

EUDAPEST TREATY ON THE INTERNATIONAL
RECOGNITION OF THE DEPOSIT OF MICROORGANISMS
FOR THE PURPOSES OF PATENT PROCEDURE
INTERNATIONAL FORM
RECEIPT IN THE CAS OF AN ORIGINAL DEPOSIT
Metapontum Agrobios S.$.iOnICa 106 km. 448,2 I issued pursuant to Rule 7.1 by the I 7~JO1O Mi_TAPONTO MT ~ INTER"IATIONAL DEPCSITARY AUTHORITY
identified at the bottom of this page ~ta~l~
name and address of depOSltor I. IDENTIFICATION OF T~ M=CROORGANISM
Identification reference given by the ~ Accession number given by the DEPOSITOR: ~ INTERNATIONAL DEPOSITARY AUTHORITY:

II. SCIENTIFIC DESG~PTION AND/OR PROPOSED TAXONOMIC DESIQ~1TION I
The microorganism identified under I above was accompanied by:
I I
I ~ a scientific description I ~ a proposed taxonomic designation (mark with a cross where applicable) III. RECEIPT AND ACCEPTANCB

j This International Depositary accepts the microorganism identified under I above, which I
received by it on ~3..~ (date dd-mm-yy of the original deposit) IV. RECEIPT OF REQUEST FOR CONVERSION

I The microorganism identified was under I above received by this International Depositary Authority on not applicable (datedd-mm-yy of the original deposit) and a I
request to convert the original a deposit to deposit under the Budapest Treaty was received by it on . notappliCable (date dd-mm-yy of receipt of request for conversion) I V. ?NTBRt~ITIONAL DEPOSITARY

Name : Centraalbureau voo~ Schimmelcuttures signature ( s ) of person ( s ) having the power j to represent the International Depositary Authority authorized offi 'al Address : Oosterstraat 1 P.O. Box 273 3740 AG BAARN Mrs F.B. Snippe-Claus JA S rs The Netherlands Date (dd-mm-yy): 1 3-99 I

1 Where Rule 6.9(d) applies, such date is the date on which the status of ftlteznational depositary authority was acquired.
Form BP/9 (sole page) CBS/9107 BUDAPEST TREATY ON THE INTERNATIONAL
RECOGNITION OF THE DEPOSIT OF MICROORGANISMS
FOR THE PURPOSES OF PATENT PROCEDURE
INTERNATIONAL FORM
Metapontum Agrobios vIAeiLITY STAT~ENT
S.S.IOnlca 106 km. 448,2 issued pursuant to Rule 10.2 by the 7~JO1O METAPONTO MT INTERNATIONAL DEPOSITARY A~)THORITY
identified on the follcwing page Itaii~
name and acdress of the party tc whom the viabi_'ity statement is issued I I. DEPOSITOR I II. IDENTIFICATION OF TF~ MZCROORC,ANISM

Name: MetapOntum AgrOblOS Accession number given by the INTERNATIONAL DEPOSIT.ARY AUTHORITY:

I

Address: S.S. lonica 106 km. 448,2 75010 METAPONTO MT Date (dd-mm-yy) of the deposit or of the Italie3 t r ans f a r : 1 III. VIABILITY STATEDtENT

The viability of the microorganism identified under II above was tested on 08-03-99 2 On that date(dd-mm-yy), the said microorganism was ~3 Xu viable ~3 no longer viable lIndicate the date of the original deposit oz, where a new deposit or a transfer has been made, the most recent relevant date (date of the new deposit or date of the transfer).
2 In the cases referred to in Rule 10.2(a)(ii) and (iii), refer to the most recent viability test.

Mark with a cross the applicable box.
Form BP/9 (first page) IV. COI~ITIONS BNDER WHICH THE VIABILITY HAS BEEN PERFORI~D
I V. IZ7TERNATIONAL DEPOSITARY ADTHORITY
Name: CentraalbureauvOOrSCtlimmeICUltures Signature (s) of persons) having the power to represent the International Cepositary I
A~hc~i~p-ox~of-authorized official(s): I
Address: posterstraat 1 ~ ~' .
PØ Box 273 l 3740 AG BAARN Mrs F.B. Snippe-Claus S pers The Netherlands Date (dd-mm-y.y) : 11-03- "
9 Fill in if the information has been requested and if the results of the test were negative.
Fozm HP/9 (second and last page) SEQUENCE LISTING

NUMBER OF SEQUENCES: 5 (1) INFORMATION FOR Q ID NO:1:
SE

(i) SEQUENCE CHARACTER ISTICS:

(A) LENGHT: 1216 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Sing le (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: A (genomic) DN

(xi) SEQUENCE DESCRIPTION:

CACAAGGCGT TTTCTGGTAC CGGAACAAGT ATTAAAP~AAG 1160 (1) INFORM ATION FOR
SEQ ID N0:2:

(i) SEQUEN CE CHARACTERISTICS:

(A) LENGHT : 624 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS:
Single (D) TOPOLOGY:
Linear (ii) MOLEC ULE TYPE: rimer p (xi) SEQUENCE
DESCRIPTION:

(1) INFORMATION EQ ID N0:3:
FOR S

(i) SEQUEN CE CHARACTERISTICS:

(A) LENGHT : 633 base pairs (B) TYPE: Nucleic d aci (C) STRAND EDNESS: gle Sin (D) TOPOLO GY: Linear (ii) MOLEC ULE TYPE: ) DNA (genomic (xi) SEQUENCE
DESCRIPTION:

(1) INFORMATION
FOR SEQ
ID N0:4:

(i) SEQUEN CE CHARACTERISTICS:

(A) LENGHT : 762 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS:
Single (D) TOPOLOGY:
Linear (ii) MOLEC ULE TYPE:
cDNA

(xi) SEQUENCE
DESCRIPTION:

ATCTCTCTCA
GTTTTCTTTC

GTTTATTTAT ATG
TTTTCTTGCT GCA

Met Ala Glu Thr Pro Ser Lys Arg Gln Arg Ser Asn Glu Thr Tyr Ala Glu Asp Val Glu Leu Leu Leu Leu Leu Leu Pro Thr Gln Asp Ile Ser Ser Leu Ile Leu Leu Leu Leu Leu Leu Pro Thr Gln Asp Ile Ser Ser Leu Ile Thr Thr Leu Gln Gln Glu Leu Asp Asp Pro Leu Ser Cys Pro Ser Thr Glu Thr Gly Arg Glu Ser Pro Phe Ala Thr Ile Thr Ser Ala Ile Glu Asp Tyr Pro Ser Ser Ser Ser Ser SerSer Ser Ser Ser Ser Ser Met Leu Leu Lys Glu AspGlu Glu Asp Asp Lys Asp Arg Val Ile Arg His LeuLeu Glu Ala Ser Asp Asp Glu Leu Gly Ile Pro AspThr Glu Thr Gly Ser Gly Phe Asp Asp Gly Tyr GluGly Phe Val Ser Ser Ala Asn Gly Phe Ser Gly ValAsp Gly Phe Ser Leu Cys Asp Gly Leu Trp Glu IleGlu Asp Ala Asn Ala Asn Tyr Tyr Ala Leu Leu Gln Ser Glu Leu Ser Phe Met Leu TTAGGGGGTG
TAGATGGAAA

(1) INFORMATION FOR SEQ ID N0:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 180 amino acids (B) TYPE: amino acid (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION:

Met Ala Glu Thr Pro Ser Lys Arg Arg Ser Glu Thr Tyr Gln Asn Ala Glu Asp Val Glu Leu Leu Leu Leu Leu Thr Gln Asp Leu Pro Ile Ser Ser Leu Ile Leu Leu Leu Leu Leu Thr Gln Asp Leu Pro Ile Ser Ser Leu Ile Thr Thr Leu Gln Glu Asp Asp Pro Gln Leu Leu Ser Cys Pro Ser Thr Glu Thr Arg Glu Pro Phe Ala Gly Ser Thr Ile Thr Ser Ala Ile Glu Asp Pro Ser Ser Ser Ser Tyr Ser Ser Ser Ser Ser Ser Ser Ser Met Leu Leu Lys Glu Asp Glu Glu Asp Asp Lys Asp Arg Val Ile Arg His Leu Leu Glu Ala Ser Asp Asp Glu Leu Gly Ile Pro Asp Thr Thr Ser Gly Phe Glu Gly Asp Asp Gly Tyr Glu Gly Phe Val Ser Ala Gly Phe Ser Ser Asn Gly Val Asp Gly Phe Ser Leu Cys Asp Leu Glu Ile Glu Gly Trp Asp Ala Asn Ala Asn Tyr Tyr Ala Leu Gln Glu Leu Phe Leu Ser Met (1) INFORMATION FOR SEQ ID N0:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 22 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii ) MOLECULE TYPE: primer (xi ) SEQUENCE DESCRIPTION:

GGATCCCTCA GGAAAGTGCT TA

(1) INFORMATION FOR SEQ ID N0:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 26 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

TCTAGACATT CCTTCTTGTT CTAATT

(1) INFORMATION FOR SEQ ID N0:8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

GAATTCCGCA TGTCTACTGT C

(1) INFORMATION FOR SEQ ID N0:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

GTCGACCATT CCTTCTTGTT C

(1) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 18 base pairs 5 (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

(1) INFORMATION FOR SEQ ID N0:11:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 22 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

GAATTCAATG TCTCTGGTAG AC

(1) INFORMATION FOR SEQ ID N0:12:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

AGGGATGTTT AATACCACTA C

(1) INFORMATION FOR SEQ ID N0:13:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

GCACAGTTGA AGTGAACTTG C

(1) INFORMATION FOR SEQ ID N0:14:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:

GTGCGACATC ATCATCGGAA G

(1) INFORMATION FOR SEQ ID N0:15:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGHT: 21 base pairs (B) TYPE: Nucleic acid (C) STRANDEDNESS: Single (D) TOPOLOGY: Linear (ii) MOLECULE TYPE: primer (xi) SEQUENCE DESCRIPTION:
CCTAAGAGTC ACTTTAAAAT T

Claims (8)

1. A gene isolated from the genomic DNA of Ricinus communis characterized by the nucleotide sequence SEQ ID Nr: 4.

2. A recombinant expression vector comprising the gene having the nucleotide sequence SEQ ID Nr: 4.

3. The vector according to claim 2, deposited as E.coli DH5.alpha. MA292 with the deposit number CBS
101642.

4. A microorganism transformed with the recombinant expression vector according to claim 2.

5. Transgenic plants comprising in their cells the gene having the nucleotide sequence SEQ ID Nr: 4.

6. The transgenic plants according to claim 5, se-lected from Arabidopsis thaliana, Linum usitatis-simum, Helianthus annus and Brassica napus.

7. Seeds obtained from the transgenic plants accord-ing to claim 5.

8. A protein characterized by the aminoacid sequence having the sequence SEQ ID Nr: 5.