Technical field

The present invention relates to use of a plant enzyme gene for transformation. More specifically, the invention relates to use of a previously not described phospholipid acyl hydrolase gene in combination with a gene for an uncommon fatty acid for obtaining transgenic plants comprising both said genes.

Background of the invention

There is considerable interest world-wide in producing chemical feedstocks, such as fatty acids, for industrial use from renewable plant resources rather than non-renewable petrochemicals. This concept has broad appeal to manufactures and consumers on the basis of resource conservation and provides significant opportunity to develop new industrial crops for agriculture.

There is a diverse array of unusual fatty acids in oils from wild plant species and these have been well characterized (see e.g. Badami & Patil, 1981 ). Many of these acids have industrial potential and this has led to interest in domesticating relevant plant species to enable agricultural production of particular fatty acids.

Development in genetic engineering technologies combined with greater understanding of the biosynthesis of unusual fatty acids, now makes it possible to transfer genes coding for key enzymes involved in the synthesis of a particular fatty acid from a wild species into domesticated oilseed crops. In this way individual fatty acids can be produced in high purity and quantities at moderate costs.

Prior art

Within prior art it is known that plant tissues accumulating triacylglycerols with high amount of medium chain (fatty acids shorter than 16 carbon atoms), hydroxy fatty acids, epoxy fatty acids and acetylenic acids have low concentration of these acids in their membrane lipids (phospholipids). (Stymne et al 1990; Bafor et al., 1990, 1991 , 1993; Kohn et al., 1994).

Furthermore it is known that diacylglycerols is a common precursor for both phospholipids and triacylglycerols in plant tissues accumulating triacylglycerols (see Stymne, 1993a for review).

There is also known that CDP-choline choline phosphotransferase in plant tissues accumulating high amounts of medium chain and hydroxy fatty acids in their triacylglycerols do not discriminate against diacylglycerols containing these fatty acids in the synthesis of phosphatidylcholine (Vogel & Browse, 1995).

Prior art also describes that tissues naturally accumulating triacylglycerols with medium chain fatty acids, epoxygenated fatty acids and hydroxylated fatty acids have membrane associated acyl hydrolase activities with high specifities towards phospholipids containing the particular uncommon fatty acid this tissue is accumulating, but low activity for common membrane fatty acids (Stymne, 1993, Stahl et al., 1995).

Furthermore, prior art describes that rape seed genetically engineered to produce dodecanoic (lauric) acid in their seeds have much higher content of that acid in seed phospholipids than two plant species naturally accumulating lauric acids to approximately the same relative level (Wiberg et al, 1995).

Finally, there exists prior art concerning an anonymous expressed cDNA sequences from young shoots of rice (ID's: D49050, D47724, D47653, D47320) deposited by the Rice genome Research Program in the GenBank.

Summary of the invention

Many of the unusual fatty acids of interest, e.g. medium chain fatty acids, hydroxy fatty acids, epoxy fatty acids and acetylenic fatty acids, have physical properties that are distinctly different from the common plant fatty acids. The present inventors have found that, in plant species naturally accumulating these uncommon fatty acids in their seed oil (triacylglycerols), these acids are absent, or present in very low amounts, in the membrane (phospho) lipids of the seed. The low concentration of these acids in the membrane lipids is most likely a prerequisite for proper membrane function and thereby for proper cell functions. The idea underlying the invention is that uncommon fatty acids can be made to accumulate to high amounts in seeds of transgenic crops if these uncommon fatty acids are, more or less, excluded from the membrane lipids of the seeds.

The present invention relates to genetically engineering of oil seeds, oleogeneous yeast and moulds to accommodate high amounts of uncommon fatty acids in their triacylglycerols by introducing genes coding for phospholipid hydrolases, below also

called phospholipases, that specifically removes these fatty acids from the membrane lipids of the cell.

The inventors have identified phospholipase (phospholipase A2) enzymes respon¬ sible for the removal of medium chain fatty acids from phospholipids in plants.

Thus, in a first aspect the present invention relates to cDNA or genomic DNA coding for a phospholipid acyl hydrolase comprising a nucleotide sequence coding for an amino acid sequence with homology to Ulmus glabra phospholipase A2 as presented in Fig. 7 or amino acid sequences homologous to those encoded by the rice cDNA clones D49050, D47724, D47653 as presented in Fig. 6 and 7.

In a second aspect, the invention relates to the use of a plant phospholipid hydrolase gene (cDNA or genomic DNA coding for a phospholipid hydrolase) in combination with a gene for an uncommon fatty acid for obtaining transgenic plants comprising both said genes.

Preferably, the enzyme encoded by said phospholipid acyl hydrolase gene, or cDNA, is coding for a low molecular weight phospholipase A2 with distinct acyl specificity for uncommon fatty acids, such as medium chain, long chain (>Cι ₈ ), hydroxy, epoxy and acetylenic acids.

In a third aspect, the invention relates to transgenic oil accumulating organisms comprising, in their genome, a plant phospholipid hydrolase gene having specificity for a particular uncommon fatty acid and the gene for said uncommon fatty acid.

Preferably said organisms are selected from the group consisting of oil crops, yeasts, and moulds.

In a fourth aspect, the invention also relates to oils from such organisms.

Detailed description of the invention

Studies by the inventors on the biosynthesis and metabolism of uncommon fatty acids (i.e. medium chain fatty acids, hydroxy fatty acids, epoxy fatty acids) in different oil seeds (Bafor et al., 1991 , 1993; Banas et al., 1993, Stymne, 1993, Stahl et al., 1995), led the inventors to conclude that microsomal phospholipid acyl hydrolases (phospholipases) with specifities towards uncommon acyl groups was, at

least in part, responsible for the removal of these acids from the phospholipids in the developing oil seeds.

It was also shown that the acyl specificities of the phospholipid acyl hydrolases from different plant species were correlated with the type of accumulated fatty acid in the plants.

Elm (Ulmus glabra) seed triacylglycerols are mainly composed of octanoic (8:0) and decanoic (10:0) acids, but these acids are very low in concentrations in the phospholipids of the seeds (Stahl et al., 1995). Membrane fractions (microsomal preparations) from developing Ulmus glabra seeds had high phospholipase A2 (PLA2) activity towards phosphatidylcholine with medium chain fatty acids in position sn-2 (octanoic, decanoic and dodecanoic (12:0) acids but very low activity towards phosphatidylcholine with octadeca-9-enoic acid (oleic acid - a common fatty acids) (Stymne, 1993, Stahl et al. 1995). Microsomal preparations from developing rape seed did not have such phospholipase A2 activity towards medium chain fatty acids (Stahl et al. 1995).

If a gene coding for plant phospholipase A2 with specificities for a particular uncommon fatty acids is expressed in transgenic oil producing organisms engineered to produce that uncommon fatty acids, the recombinant phospholipase A2 will remove the uncommon fatty acids from the phospholipids of the cell and thereby prevent deleterious effects on cell metabolism caused by the presence of this acid in the membrane lipids. This invention describes how such phospholipase A2 genes will be isolated and what uses they will have in commercial applications.

The invention will be described more closely below in association with an experimental part and the enclosed drawings. The drawings show:

Fig.1 shows a SDS-polyacrylamid electrophoresis of soluble developing elm seed PLA2 purification fractions, followed by colloidal Coomassie staining. Lane A and I contain 100 ng of MW standards (Pharmacia low MW); lane B, 100 OOOg supernatant, 50 μg; lane C, ammonium sulphate pellet, 50 μg; lane D, acetone supernatant, 50 μg; lane E, Q Sepharose, 40 μg; lane F, Superose 12, 25 μg; lane G, C4-HPLC, 2 μg; lane H C2C18-SMART, 100 ng; and lane J, commercial Naja naja kouthia PLA2, 100 ng. All samples were reduced with DTT in sample buffer.

Fig.2 shows PLA2 activity measurements of gel pieces from whole lanes (5-94 KD) of a SDS-PAGE 8-18% gradient gel. Lane A contains 50 ng of Naja naja kouthia PLA2 (Sigma) and lane B 50 ng of developing elm seed soluble PLA2. PLA2 activity recovered from gel pieces of similar lanes are shown on each side.

Fig. 3 shows molecular weight data of the purified soluble PLA2, from MS-Malditof.

Fig. 4 shows molecular weight data of the purified soluble PLA2 that has been reduced and iodoacetamide alkylated, from MS-Malditof.

Fig. 5 shows a SDS-polyacrylamid electrophoresis of purified microsomal peak II PLA2 from developing elm seeds, with recovered PLA2 activity. Lane A contain about 20 ng of purified peak II PLA2, non reduced; lane B contain 25 ng of MW standard (Pharmacia low MW). The gel was silver stained.

Fig. 6 Alignment of deduced amino acid sequence of the full length rice cDNA clone

GenBank ID: D49050 with 10 different low molecular weight phospholipase A2 from animal tissues. Conserved amino acid sequences are boxed. Spaces introduced to optimize alignment are indicated by a dash. The different sequences represent phospholipase A2 from the following species:

D00035: Canis sp.

D 10070: Trimeresurus flavoiridis

M21054: Homo sapiens

X12605: Notechis scutatus scutatus

X53406: Bungarus multicinctus

X53471 : Vipera ammodytes

X76289: Bothrops jararacussυ

YO0120: Bos taurus

Y00377: Laticauda laticaudata

Fig. 7 Alignment of the N-terminal sequence of the purified soluble PLA2 from elm seeds with deduced amino acid sequences from three EST-clones from rice green shoots, including the cDNA clone D49050 fully sequenced by the inventors. The EST-sequences are denoted by their GenBank accession number. Conserved amino acid positions betweeen the elm and rice proteins as well as the regions with homology to the Ca2 ⁺ - binding and the active site in animal low molecular weight

PLA2 ^' s are boxed A fourth rice clone (GenBank ID D47320) with high homology to the three above was found in the EST database, but excluded from the alignment due to lower quality of the DNA sequence

Experimental part

Proteins with phospholipase A2 (PLA2) activities towards

1 -palmιtoyl-2-decanoyl-sn-glycerol-3-phosphocholιne were purified from a soluble and microsomal fraction of developing elm seeds according to the following protocols

Assays of phospholipase A2 activity

Membrane associated PLA2 activity was assayed according to Stahl et al (1995) using s/7-1 -palmιtoyl-s7-2-[ ^' ' ⁴ C]decanoyl-SA7-glycerol-3-phosphocholιne as substrate In standard assays of the solubilized microsomal activity and of the soluble activity 1-palmιtoyl-2-[ ¹⁴ C]palmιtoyl-glycerol-s/?-3-phosphocholιne was used as substrate and was presented as mixed micelles with the non-ionic detergent lubrol PX, in a PC/detergent molar relation of 1 10 Samples, 0 5-10 μl, were assayed for PLA2 activity by incubation at 30°C for 5-30 mm with 5 nmol of 1 Re¬ labelled phosphatidylcholine (10 000 dpm/nmol) in a total volume of 50 μl of 50 mM Tπs/HCI, pH 8 0 containing 10 mM CaCl2 and 0 06% (w/v) lubrol PX The reaction was stopped by the addition of 400 μl of chloroform/methanol/acetic acid 50 50 1 followed by 150 μl of H2O The samples were mixed thoroughly and centrifuged 10 OOOg for 1 mm Chloroform phases containing extracted lipids were passed through mini-columns of silica gel, followed by a wash with 400 μl of chloroform The eluates from the silica columns, containing released [14C]palmιtιc acid, were collected and analysed by scintillation counting

Assays of PLA2 activity from SDS-gels was performed according to following protocol, A whole or part of a SDS-PAGE (Pharmacia Exelgel 8-18%) lain, not fixed containing purified PLA2 were divided in 2-3 mm wide pieces and placed in eppendorf tubes together with 400 μl of 20 mM Tris, pH 8 0 containing SDS 0 5% (w/v) The tubes were rotated end over end at 37° C for 16 h, in order to eluate proteins from the gel pieces Fractions were concentrated to 100 μl in a Speed-Vac concentrator (Savant) and then precipitated with ethanol/chloroform (Wessel and Flugge 1984) to remove SDS Air dried pellets were solubilized in 150 μl of 50 mM

Tris/HCI, pH 8.0 containing 10 mM CaCl2 and 0.06% (w/v) lubrol PX and activity measurements were started by adding 5 nmol of s/7-1-palmitoyl-s7-2-[ ¹⁴ C]palmitoyl- sn-glycerol-sn-3-phosphocholine (10 000 dpm/nmol) solubilized in 50 μl of 50 mM Tris/HCI, pH 8.0 containing 10 mM CaCl2 and 0.06% (w/v) lubrol PX. The samples were incubated at 30° C for 2-4 h and stopped by adding 400 μl of CHCl3/MeOH/Hac, 50:50:1.

SDS-gel electrophoresis

Protein fractions were if necessary concentrated in Sped-Vac and precipitated with ethanol/chloroform according to Wessel and Flϋgge (1984). Samples with a final volume of 20 μl in 50 mM Tris/acetat pH 7.5 with 1 % (w/v) SDS, with or without 10 mM of dithiothreitol, were heated to 95° C for 5 min, centrifugated 5 min 13 000 g and loaded on to a horizontal 8-18% gradient polyacrylamid gel (Pharmacia ExelGel SDS) with a 33 mm stacking zone and a 77 mm separating zone. The gel was chromatographed on a Pharmacia Multiphor II unit at 15° C and stained either with colloidal Coomassie (Neuhoff et al 1988) over night or with silver staining.

Material.

Developing seeds of elm (Ulmus glabra) were harvested and peeled. The white endosperms were immediately frozen in liquid nitrogen and stored in -80° C.

Purification of soluble phospholipase A2.

60 g of liquid nitrogen frozen elm endosperm was homogenized with a Ultraturrrax® in 600 ml of ice cold 100 mM potassium phosphate buffer, pH 7.2 containing 0.33 M sucrose. The homogenate was filtered through two layers of Miracloth® and centrifuged 10 000 g for 12 min. The supernatant was filtered through one layer of Miracloth and centrifuged a second time, at 100 000 g for 90 min. The final supernatant, which contained about 80% of the total PLA2 activity, was brought to 55% (w/v) of ammonium sulphate and left with steering at 4 °C for 1 h. Precipitated proteins were pelleted by centrifugation 10 000 g for 10 min and resuspended in 130 ml of 50 mM dietanolamin buffer pH 8.5. Ice cold acetone was added to a final concentration of 45 % (v/v) and the extract was left at 4°C for 30 min. Precipitated proteins were removed by centrifugation for 10 min at 10 000 g and the resulting supernatant was dialysed against 20 volumes of 20 mM piperidin, pH 11.0, with one change over night. The dialysed extract was applied to a Q-Sepharose Fast Flow 7 ml column (1.0 x 10.0 cm) equilibrated in 20 mM piperidin, pH 11.0. The column was

eluted with a linear salt gradient from 100 to 500 mM NaCl in 20 mM piperidin, pH 11.0 at a flow rate of 2 ml/min. 3 ml fractions were collected and assayed for PLA2 activity. A single broad peak of activity was eluted at a salt concentration of 200 to 300 mM NaCl. Peak fractions were pooled, concentrated on Centricon-10 to 0.6 ml and chromatographed in three separate runs on a Pharmacia Superose 12 (1.0 x 30.0 cm) gel filtration column (0.4 ml/min) in 20 mM Tris/HCI, 150 mM NaCl, pH 8.0. Fractions (0.5 ml ) were collected and tested for PLA2 activity. Peak fractions from all three runs were pooled and the PLA2 was further purified using a C4 reversed- phase HPLC column (Vydac 0.46 x 10.0 cm) that was equilibrated in 0.1% trifluoroacetic acid (TFA). The column was developed at 0.4 ml/min with a 30 min gradient (20-45% of acetonitrile in 0.1 % TFA) and peaks monitored at 280 nm were collected manually. Collected fractions from four separate runs were assayed for PLA2 activity. Peak fractions were pooled and the acetonitrile content was reduced by evaporation in a Speed-Vac concentrator (Savant). The PLA2 was finally purified to apparent homogeneity on a C2C18 reversed-phase HPLC column (0.21 x 10.0 cm) equilibrated in 0.1 % TFA and developed at 100 μl/min with a 60 min gradient (30-60% acetonitrile in 0.1 % TFA) using a SMART system (Pharmacia). Peaks monitored at 280 nm were automatically collected and then subjected to PLA2 assay. The PLA2 elutes as a discrete peak in the gradient at about 50% acetonitrile.

The PLA2 was purified about 180 000 times from the developing elm seed extract of soluble proteins, to a final specific activity of 44 μmol/min x mg protein (see table I).

Tabel I, Purification of soluble PLA2- from developing elm seeds

Specific

Protein Total activity Purificatio Yield

(mg) activity (nmol/min x n (%) (nmol/ml) mg protein)

100 000g sup 3 340 833 0.25 1 100

Am. sulphate pell 1 060 563 0.53 2 68

Acetone sup 150 780 5.2 21 94

Q Seph F.F. 24 420 17.5 70 50

Gel filtration 3.8 263 69.2 277 32

C4-HPLC 0.09 173 1 922 7 690 21

C2C18-SMART 0.003 133 44 330 177 300 16

The final extract showed one major band when subjected to SDS-PAGE on a 8-18% gel and stained with colloidal Coomassie, with a molecular mass of 14 kDa (se Fig 1 ). Recovered PLA2 activity from SDS-PAGE gels coincide with the 14 kDa band (see fig. 2). When subjected to Malditof-MS, the PLA2 gave two major peaks with the masses 13220 and 13890 and a minor with a molecular mass of 12680 (see fig 3). When alkylated with N-isopropyl all three peaks changed masses with about the same amount, 1150, which would correspond to 12 cystein residues in each of the three proteins (fig. ).

N-terminal sequence analysis

About 1 μg of purified PLA2 was reduced, by incubation in 0.1 M Tris/HCI, pH 8.5 containing 8 M guanidinhydrocloride, 10 mM EDTA and 20 mM DTT at 56° C for 30 min followed by alkylation in 20 mM 4-vinylpyridin for 60 min at room temperature. The reduced and alkylated PLA2 was desalted, applied on a C2C18 reversed- phase HPLC column (0.21 x 10.0 cm) equilibrated in 0.1 % TFA and eluted at 100 μl/min with a 30 min gradient (30-60% acetonitrile in 0.1 % TFA) using a SMART system (Pharmacia). The protein was then subjected to amino-terminal sequence determination by automated Edman degradation using an Aplied Biosystems 476A gase phase protein sequenator. The amino-terminal sequence was manually determined to be:

XNVGVQATGTSISVGKGCF(S)RKCE(P)P(K)F(Y,L)FCYGPXFLR(L)Y(S) (when signals for several amino acids were obtained the minor amino acid signal(s) is denoted in brackets after the main signal). When using the amino-terminal sequence as query for the Basic local alignment search tool at NCBI with the blastp search program against the Non-redundant GenBank CDS translations+PDB+SwissProt+SPupdate+ PIR, the tblastn search program against Non-redundant GenBank+EMBL+DDBJ+ PDB sequences and Non-redundant Database of GenBank EST Division the best aligned sequences are three EST's (GenBank accession number D47724, D47653 and D47320) derived from green rice shoots. Fig. 7 shows an alignment of the amino-terminal sequence with the deduced amino-acid sequence from two of these EST-clones and the D49050 rice EST- clone. The amino-terminal sequence show significant homology with the rice sequences, notably the positions of the three cysteine-residues are conserved. In addition, predictions of leader peptide cleavage site of rice clones D47724 and D47653 suggest cleavage between G 24 and L 25. This supports the alignment of the amino-terminal sequence to the mature part of the rice sequences. Regions with

high homology to the conserved Ca ²⁺ binding- and active sites of secretory PLA2's (see fig 6) are both found in all three alined rice EST's.

Purification of microsomal phospholipase A2.

60 g of liquid nitrogen frozen elm endosperm was homogenized with a ultraturrax® in 600 ml of ice cold 100 mM potassium phosphate buffer, pH 7.2 containing 0.33 M sucrose. The homogenate was filtered through two layers of Miracloth® and centrifugated 10 000 g for 12 min. The supernatant was filtered through one layer of Miracloth® and centrifugated a second time, at 100 000 g for 90 min. The microsomal pellets were resuspended in 90 ml of 100 mM potasium phosphate, pH 7.2 with a glass homogenizer and, if not used immediately, stored at -80° C. The microsomal membranes were diluted to 150 ml with 100 mM potasium phosphate, pH 7.2 and solubilized by the addition of 150 ml of 200 mM potasium phosphate, pH 7.2 containing glycerol 17% (v/v), lubrol PX 0.6% (w/v) and EGTA 1 mM. The mixture was incubated 15 min at 4° C followed by a centrifugation 100 000 g at 4° C for 90 min. The supernatant was dialysed against two changes of 5 liter of 20mM dietanolamin, pH 8.5 containing glycerol 8.7% (v/v) and lubrol PX 0.06% (w/v). The dialysed supernatant was passed through a 200 ml Q-Sepharose column (50 x 100 mm) at a flow rate of 3 ml/min. Non-retained material with about 30-50 % of the PLA2 activity was collected and pH adjusted to 5.7 by adding requiring amount of 1 M MES buffer. This fraction was applied to a 7 ml SP-Sepharose column (10x 100 mm) equilibrated in 50 mM MES, pH 5.7 with glycerol 8.7% (v/v) and lubrol PX 0.06%(w/v) at a flow rate of 3 ml/min. After the sample had passed through, the column was washed with several column volumes of equilibrating buffer and then eluted with a 100 ml linear gradient from 0 to 480 mM of NaCl in the same buffer. Eluated fractions containing PLA2 activity were pooled and concentrated to a volume of 200 μl by centrifugation on Centricon-50 and vacuum evaporation in a Speed-Vac concentrator (Savant). The sample was applied to a Superose 12 (10 x 300 mm) Pharmacia column equilibrated in 20 mM Tris, pH 8.0 with glycerol 4.3% (v/v), lubrol PX 0.06% (w/v) and 50 mM NaCl. Fractions with PLA2 activity were pooled and further purified using a C4 reversed-phase HPLC column (Vydac 0.46 x 10.0 cm) that was equilibrated in 0.1 % trifluoroacetic acid (TFA). The column was developed at 0.4 ml/min with a 30 min gradient (20-45% of acetonitrile in 0.1 % TFA) and peaks monitored at 280 nm were collected manually. Collected fractions were assayed for PLA2 activity, and found to be divided into to aktivity peaks, one (peak I) which eluted at about 35% acetonitrile and the second (peak II) which eluted at

about 47% acetonitrile Peak fractions were pooled and lubrol PX to a final concentration of 0 5% (w/v) was added befor the acetonitrile content was reduced by evaporation in a Speed-Vac concentrator (Savant) The two PLA2 fractions were both finally purified to near homogeneity on a C2C18 reversed-phase HPLC column (0 21 x 10 0 cm) equilibrated in 0 1 % TFA and developed at 30 μl/min with a 60 mm gradient (30-60% acetonitrile in 0 1% TFA) using a SMART system (Pharmacia) Peaks monitored at 280 nm were automatically collected and then subjected to PLA2 assay The purified peak I PLA2 gave a very sharp band on SDS-PAGE 8- 18% gradient gel with a molecular mass around 17 KD and the peak II PLA2 gave a 14 KD band Both bands coincided with recovered PLA2 activity from gel pieces Fig 5 shows the purified peak II PLA2 separated on a SDS-PAGE followed by silverstaming After Coomassie staining only the 14 KD band was visible However, upon silverstaming some minor contaminants show up

The microsomal PLA2 activity was purified from the microsomal fraction with a specific activity of 0 28 nmol/min x mg protein to a specific activity of about 50 μmol/min x mg protein which gives a purification factor of about 100 000

Properties of punfied PLA2S

The purified soluble and microsomal PLA2S have very similar properties They have a pH optimum between 7 and 9, an absolute requirement for Ca2+ for activity with several mM for optimal activity The activities are extremly stable both to extreme pH values, heat and organic solvents The activities are, however, sensitive to reducing agents like DTT and mercaptoethanol (see Table II)

Table II Effects of reduction, EGTA and heat on developing elm seed soluble PLA2 Treatment PLA2 activity Released [ ¹⁴ C]fatty acids (dpm)

Control 4740

95° C for 5 mm 5190

Mercaptoethanol 1 % (v/v) 170

EGTA I O mM 170

The purified PLA2's hydrolyses the sn-2 position of phospholipids (Table III), and does not show any activity towards diacylglycerols or lysophosphatidylcholine (Table IV).

Table III. Position specificity of soluble developing elm seed PLA2. Incubations were done according to the PLA2 assay described above with a partly purified soluble

PLA2 fraction. (PC=phosphatidylcholine, LPC=lysophosphatidylcholine) Recovered ¹⁴ C activity (% of total recovery)

PC fatty acid PC ^LPC substrate

Table IV. Substrate specificity of microsomal PLA2. Incubations were done according to the

PLA2 assay described above. (PC=phosphatidylcholine, LPC=sn-1- lysophosphatidylcholine, DAG=Diacylglycerol)

Recovered ¹⁴ C-activity (% of total recovery) Substrate fatty ^{PC PC} DAG acid sn-1-16:0-sn-2-[ ¹⁴ C]16:0- 34 65 0.8

PC di-[ ¹⁴ C]16:0-PC 27 40 33 sn-1-16:0-sn-2-[ ¹ C]10:0- 44 54 0.6

PC

[ ¹ C]10:0-LPC 0.8 _ 98 sn-1-16:0-sn-2-[ ¹⁴ C]10:0- 0.8 . _— 99

DAG

The molecular weight and the biochemical characteristics of both the soluble and microsomal elm PLA2 suggest that they are related to the well described low MW "secretory" PLA2S from animal sources. This is further supported by the amino- terminal sequence data and alignments. The secretory PLA2's have all conserved amino acid sequences at the Ca ²⁺ bindning site and at the active site as well as cyein residues.

When searching databases for deposited expressed sequences from plants with homology to low molecular weight animal phospholipases in Ca ²⁺ binding site and active site, the inventors found three anonymous partially sequenced cDNA clones from green shots of (GenBank ID: D49050, D47724, D47653). The cDNA clone D49050 was received upon request from Dr. Yoshiaka Nagamura, DNA Materials Management group, Rice Genome Project, NIAR/STAFF, STAFF Institute, 446-1 , Ippaizuka, Kamiyokoba Tsukuba, Ibaraki 305 Japan. The entire cDNA was sequenced and was shown to contain an open reading frame encoding a full length protein of an estimated molecular weight of 15 kDa. An alignment of the deduced amino acid squence of D49050 with a number of animal low molecular weight PLA2s is presented in Fig. 6.

The D49050, D47724, D47653 clones coded for proteins with the same amino acid sequences as in thee Ca ² + binding site and active site in the animal low molecular weight PLA2s and similar to these enzymes they contained several cystein residues (see Fig. 7). The cDNA clones also coded for amino acid sequences with significant homologies with the N-terminal sequence of the purified phospholipasee A2 from elm seeds where the positions of the three cystein residues of the elm enzyme was totally conserved in all three cDNAs (see Fig.7). Thus with all probability these rice cDNAs were coding for a plant PLA2 similar to the enzyme purified from developing elm seeds according to the invention.

By expressing this cDNA in suitable organism, like bacteria, for example E.coli, yeast or plants, a recombinant PLA2 protein will be obtained and PLA2 activities can be demonstrated. Although the physiological function of the rice enzyme is unknown, a function in rice shoots could be removal of oxygenated fatty acids from membrane lipids, as has been shown to take place in e.g. wheat roots (Banas et al, 1992).

By constructing degenerated nucleotide primers based on suitable amino acid sequences of the soluble elm PLA2 and rice cDNA clones amplification of elm fragments containing the corresponding sequences will be done from cDNA or genomic DNA from elm seeds by PCR. These fragments will be used as probes for screening for the elm cDNA PLA2 from a cDNA library from developing elm seeds.

Since phospholipid acyl hydrolases with high specificities towards epoxy and hydroxy fatty acids have been described in membrane preparations from other plant

species (Stahl et al. 1995) homologous cDNA coding for PLA2 with other acyl specificities than the elm enzyme can be isolated from other plant species with the aid of the cDNA encoding for the elm PLA2 and/or the rice cDNA clones as probes Alternatively suitable amino acid sequences of these enzymes can be used to construct degenerated nucleotide primers and amplify cDNA fragments derived from the other plant species. These fragments will be used as probes for screening for the cDNA coding for PLA2 from a cDNA library from other plant species.

When a cDNA clone containing a full length cDNA or genomic DNA coding for a PLA2 have beeen obtained this cDNA can be used for transformation.

According to the invention, the PLA2 gene, i.e. the PLA2 cDNA or genomic clone, is used in combination with a gene for an uncommon fatty acid for obtaining transgenic plants comprising both said genes. The transgenic plants are obtained by using said plant phospholipid acyl hydrolase gene for transforming transgenic oil accumulating organisms engineered to produce said uncommon fatty acid. Alternatively, transgenic plants are obtained by using the plant phospholipid acyl hydrolase gene for transforming oil accumulating organisms, which are crossed with other oil accumulating organisms engineered to produce said uncommon fatty acid.

REFERENCES

Badami, R.C, and Patil, K B. (1981). Structure and occurrence of unusual fatty acids in minor seed oils. Progress in Lipid Research, 19, 119-53.

Bafor, M., Smith, M.A., Jonsson, L, Stobart, K. & Stymne, S (1990) Regulation of triacylglycerol biosynthesis in embryos and microsomal fractions from the developing seeds of Cuphea lanceolata. Biochem. J. 272, 31-38

Bafor, M., Smith, M.A., Jonsson, L, Stobart, K. & Stymne, S. (1993) Biosynthesis of vernoleate (cis-12-epoxyoctadeca-cis-9-enoate) in microsomal preparations from developing endosperm of Euphorbia lagascae. Arch. Biochem. Biophys. 303, 145-151

Banas, A., Johansson, I. & Stymne, S. (1992) Plant microsomal phospholipases exhibit preference for phosphatidylcholine with oxygenated acyl groups. Plant Science 84, 137-144

Kohn, G., Hartmann, E., Stymne, S. & Beutelmann, P. (1994) Biosynthesis of acetylenic acids in the moss Ceratodon purpureus. J. Plant Physiol. 144, 265-271

Neuhoff, V, Arold, N., Taube, D. and Ehrhart, W. (1988) Improved staining in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity useing Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9, 255-262

Stymne, S (1993a) Biosynthesis of storage fat in oil crops - today and tomorrow. ln:Phytochemistry and Agriculture (Eds. T.A. van Beek & H. Breteler. pp. 288-312 Clarendon Press, Oxford)

Stymne, S. (1993b) Biosynthesis of uncommon fatty acids and their incorporation into triacylglycerols. In: Biochemistry and Molecular Biology of membrane and Storage Lipids of Plants (eds. N. Murata & C. Somerville), pp. 150-158. American Society of Plant Physiology, Rockville.

Stymne, S., Bafor, M., Jonsson, L, Wiberg, E., and Stobart, A.K. (1990). Triacylglycerol assembly. In Plant Lipid Biochemistry, Structure and Utilization, (eds. P J. Quinn and J.L. Harwood), pp. 191-97. Portland Press, London

Stahl, U., Banas, A. & Stymne, S. (1995) Plant microsomal phospholipid acyl hydrolases have selectivities for uncommon fatty acids. Plant Physiol. 107, 953-962

Vogel, G. & Browse, J. (1995) Role of choline phosphotransferase and diacylglycerol acyltransferase in channelling unusual fatty acids into the triacylglycerol pool during oil seed development. In: Plant lipid metabolism (eds. J-C. Kader & P. Mazliak), pp. 506-508. Kluwer Academic Press, Dordrecht

Wessel, D. and Flϋgge, U. I. (1984) A method for quantitative recovery of protein in diluted solution in the presence of detergents and lipids. Biochemistry 138, 141 -143

Wiberg, E., Banas, A. & Stymne, S. (1995) Partitioning of medium chain fatty acids between membrane and storage lipids in laureate producing rape. Abstract 039 in abstract book from the symposium: Biochemistry and Molecular Biology of Plant Fatty acid and Glycerolipid. June 1-4, South Lake Tahoe, California