FIELD OF INVENTION The invention relates to methods enabling to prepare improved commodities by nanotechnology in the fields of food, feed, cosmetics, and medicine. In particular, the invention relates to products, which benefit from improved stability of water-insoluble compounds like fats and oils or compounds dissolved in such hydrophobic materials. According to the present invention the substances are protected by their encapsulation by proteins, which assemble to make a biphasic membrane layer between hydrophobic and hydrophilic phases.

BACKGROUND OF THE INVENTION

Storage of lipids into plant seeds is essential not only for survival and subsequent growth and maturation of new plants but also to serve as an important food source for animals and humans. Seeds of many plant species store lipids in discrete organelles called oleosomes or oil bodies. Oil bodies contain approximately 95% triacylglycerols (TAG) and 5 % of proteins and phospholipids. Majority of these oil body-proteins are oleosins (Napier et al., 1996, Plant MoI. Biol. 31:945-956). Oil bodies are about 0.6-2 μm in diameter and contain a core of TAGs surrounded by an outer coat of phospholipid (PL) monolayer and a single class of proteins termed oleosins. Current evidence suggests that oleosin synthesis occurs via co-translational integration into endoplasmic reticulum (ER) membranes. Biogenesis of oil bodies parallels the localisation of oleosins even though the signal sequence for targeting of oleosin to the sites where oil bodies are formed is not known. This indicates dynamic interaction between sites of oil body formation and oleosin synthesis as described (Sarmiento et al. 1997, Plant J 11:783-796). Oleosin structure possesses three different functional domains. It is known in the prior art that the hydrophobic central domain is the key element for localisation of oleosin into oil bodies; lack of it totally abolishes localisation of oleosin into oil bodies (Abell BM, et all 997, Plant Cell 9:1481-1493). Hydrophobic central domain is flanked by the N- and C-terminal domains, which are variable in length (50-70 and 55-98 amino acids, respectively). Short amino- (N)-terminal deletions (20 [N20] and 40 [N40] amino acids; respectively) do not disturb the transportation of oleosin into oil bodies, but longer N-terminal deletions (N66 and N90) reduce the localisation rate or fully abolish it. Neither the N- nor the C-terminal hydrophilic domains part are needed for targeting or localisation of oleosin into oil bodies. Previous observations suggest that the hydrophobic central domain in intact oil bodies is predominantly α-helical in structure, and it can complement the functions of some other lipid associated proteins (Hope et al. J. Biol. Chem. 277, 4261-4270). The N- and C-terminal domains of oleosins have been predicted to contain amphipathic α-helix and some random-coil structure. However, the structure of oleosin in the association of oil bodies and its functional properties are not known in the prior art, and there are no data about the structure of N- and C-terminal domains in the water-lipid interphase.

Oleosins accumulate with oil bodies, and oil seeds subjected to an aqueous extraction result in intact oil bodies (van Rooijen and Moloney, Bio/technol. 1995, 13:72-77). They can be separated with the accumulated oleosins by centrifugation from the rest of cellular debris as a floating oily "scum". Preparation of oleosomes and oleosome coats have been described (Bergfeld et al., 1978, Planta, 143:297-307). Mustard seeds were crushed with mortar and pestle followed by sucrose gradient centrifugation to get the oil body fraction. It was then extracted with chloroform or chloroform/methanol 2:1 and centrifuged with low speed to obtain "coat fraction". Tzen and Huang (J. Cell Biol. 1992, 117:327-335) showed that reconstitution of oil bodies was successful if oleosins, triacyl glycerol (TAG) and phospholipids (PL) were present. It was also reported that either TAG and PL purified from maize or commercially available lipids (dioleoyl phosphatidylcholine and a 1 :2 molar mixture of triolein and trilinolein) could be used in the reconstitution.

WO 0226788 (Harada Takya et al.) describes the use of oleosin as a preferable safe natural stabilizer of emulsification to be applied with or without other emulsifying agents. Because of the generally known biphasic nature of oleosins, like with the detergents, the emulsifying function is evident. Normally emulsified or micellar fats and oils are more prone to chemical reactions, including oxidation, than their non-micellar counterparts due to the drastically higher surface area of the reacting species. Therefore the mode of usage and function of oleosin in WO 0226788 differs essentially from that of the present invention, which states that oleosin can be used for protection of hydrophobic compounds from chemical reactions. Provisional patent application JP2002-101820 describes the preparation of oil bodies from soya seeds by the centrifugation of water-soya suspensions by the methods essentially known from the prior art followed by a heat treatment. Said invention does not involve examples for the use of the prepared nanoparticles or does not describe their chemical or functional properties.

We found surprisingly that oleosin can stabilize oils and compounds dissolved in oils from chemical reactions. This finding was studied and found that oleosins form a tight three- dimensional network via intramolecular non-covalent linkages between oleosin molecules, which are partially embedded into lipids. This protein "coat" protects the lipid phase from the chemicals. It was especially surprising that the oleosin "coat" can protect lipids from molecular oxygen because molecular oxygen is a small molecule, which easily diffuses through pores of, for example, certain plastic films (PTFE and others) of even of high thicknesses. Thus, the oleosin coat forms a specific shield over lipid bodies. Comprehension of this oleosin coat structure enabled us to devise technology for protection of hydrophobic compounds by the oleosin, as well as developing new methods for producing oleosin.

BRIEF DESCRIPTION OF DRAWINGS

Figure 1. Stability of edible seed oil from B. rapa for oxidation caused by an intact or deactivated oil bodies. The amount of conjugated dienes (measured by UV absorption at 234 nm in methanol) served as the measure of the oxidation (a). The relative area of gas-solid chromato graphic peaks of propanal indicates the volatile oxidation products generated by the reaction of unsaturated lipids with oxygen (b). Heat treated oleosin was used as the control.

Figure 2. Size of oil bodies and transportation of wild-type and mutant oleosin-GFP fusions visualized by confocal laser scanning microscope, (a) Normal oil bodies (diameter less than 2 μm) visualized by full-length oleosin-GFP. (b) Enlarged oil bodies in cells expressing amino- terminal oleosin mutant fused to GFP (oleosin-66-GFP). The inserted frame in the middle of the figure is magnified in c, d and e. (c) Oil bodies stained with neutral, lipid-specific stain, Nile blue A. (d) Localisation of oleosin-66-GFP in the lumen of endoplasmic reticulum (ER) and co- localisation with oil bodies; oleosin-66-GFP is strongly retained in ER (solid light area surrounding oil bodies), (e) Merged image of figures Lc and Ld. Scale bar (c-e) is 10 μm.

Figure 3. Molecular model of oleosins residing on the surface of oil bodies in water-lipid environment, (a) Longitudinal section of an oil body (sphere) covered by oleosin molecules, (b) Organization of oleosin molecules on the surface of oil bodies (a view from the top). Black spheres symbolize hydrophobic transmembrane α-helices of oleosin molecules that anchor oleosins into oil phase. Open boxes symbolize α-helices laying along the lipid phase (yellow) allowing optimal formation of a continuum of "zipper" structure, a β-sheet formed by intermolecular contacts between N- and C-terminal domains of oleosin (black and lighter sic-sac, respectively).

Figure 4. Sequence of Arabidopsis thaliana oleosin gene (data bank ace. X62353).

DETAILED DESCRIPTION OF THE INVENTION

The basic embodiment of the present invention is the finding that edible, technical or medical fats and oils, or other such hydrophobic compounds, can be protected from penetration of harmful substances by coating with specific proteins. Equally well, compounds dissolved or suspended in such hydrophobic materials can be protected. The harmful substances can be chemically reactive or such which can spoil odour, taste, nutritive value, or other related properties.

A key methodological step leading to the present invention was the finding that a marker protein, Green Fluorescent Protein (GFP) can be introduced into a specific position of the oleosin recombinant protein without losing functions of these two proteins. By virtue of this, we were able to demonstrate with the real-time imaging techniques that deletion of the N-terminal part of oleosin prevented efficient lateral transportation of oleosin from rough endoplasmic reticulum membranes to oil bodies. Oil bodies covered with truncated oleosin molecules were enlarged and allowed to be stained with lipid-specific stain, Nile Blue A, whereas oil bodies covered with intact oleosins remained small and could not be stained with Nile Blue A. Thus, opposite to the previous considerations, we demonstrated that entire oleosin structure is essential for preservation of structural integrity of oil bodies. This statement tempted us to search for unexpected structural solutions for the oil bodies associated with proteins. We entered to a molecular model in which N- and C-termini of two oleosins bind intermolecularly to each other to form a continuum of β- sheet-mediated "zipper" surface, while the hydrophobic central domain of oleosins anchors the zipper surface into oil bodies. Since the surface area of a sphere is strongly dependent on its volume, by limiting the content of oleosin the respect of triacylglycerol, the sizes of the oil bodies can be controlled by oleosin explaining the observations that without correctly formed zipper structure, the size of oil bodies is enlarged. Thermodynamics of oleosins' lipophilic core unit combined with the (upper) hydrophilic zipper part forces tight packaging of oleosins on the surfaces. Depending on geometry of such zipper surfaces, also small and variable amounts of other filler proteins may take place in the formation of the protecting film. Based on these surprising principal mechanistic findings, we were able to develop new technology for preparing oil bodies and defined oleosins' technological utilization.

A number of cDNA and genomic clones encoding oleosins have been isolated from many oil seed plants (e.g. sun flower, soya, spring turnip rape). Oleosin structures seem to be enough conserved in all of them indicating that the zipper-structure mechanism for storing oil bodies is universal in the protection of oily substances. According to our molecular modelling studies these oleosins produce similar zipper structures. This list of zipper coat structures contains, but is not limited to, species like Arabidopsis thaliana, sunflower (Helianthus annum), rice (Oryza sativa), maize, safflower (Carthamus tinctorius L.), wheat, barley, rye, soybean, rape, false flax (Camelina sativa), oat, canola {Brassica napus), spring turnip rape {Brassica rapa), sesam seeds. It is also noted that some other oleosin-like proteins can be found in e.g. mammals and we assume that their mechanism of action is similar to zipper oleosins from plants. These proteins include adipophilin, perilipins, and caveolins. Moreover, viruses such as core proteins of Hepatitis C virus and GB virus-B contain related structures. These gene and amino acid sequences can be found from publicly available data banks and such data can be subjected to computer modeling procedures to find out the zipper structures.

However, these differences in sequences indicate that it is essential from what source is the oleosin to be used in the zipper formation, and the oleosins from different sources may not be mixed. Various oil seed plants such as Canola, which is an edible oil grown and consumed around the world as a superior edible oilseed, are good sources for oleosin. Each canola seed contains at least 40 percent of oil, which is extracted to produce margarines and oils. Oleosin can be directly extracted during the processes of seed oil production.

Oleosin, as other lipid or membrane-associated proteins, has structural features that are difficult to determine with x-ray crystallography, and therefore other indirect non-trivial and novel techniques had to be applied to solve complex structural bias of oleosin. It is known from previous art that oleosin has structure of three different functional domains. Hydrophobic central domain is essential for localisation of oleosin into oil bodies, and it can even complement the function of an heterologous organism; binding of core protein of Hepatitis virus C to lipid membranes was restored by central domain of oleosin protein (Hope RG, et al. 2002, J Biol Chem 277:4261-4270). The roles of amino- (N)- and carboxy- (C)-termini of oleosin have been poorly understood, since there are no molecular modelling theories encompassing the polar-non polar (i.e. water-lipid) interfaces. Previously, it has been shown that a truncated oleosin mutant in which the first 66 N-terminal amino acids were deleted (designated as N66) has disturbed localisation in various plant tissues. However, these results rely on biochemical protein analysis and not imaging in intact cells, and are thus not reliable at cellular level. In contrast, C-terminus of oleosin is suspected not to be essential for localisation of oleosin into oil bodies. However, the techniques used in previous studies cannot provide solid cell biological data on oleosin localisation and transport and structure of oil bodies. Therefore, by applying new, real-time imaging techniques combined with computer-aided molecular modeling and previous observations, oleosin localisation and oleosin- membrane topology could be explained in the present invention.

Native oleosins can be extracted from seed materials by added oil droplets micellized into water with or without detergents. However, defatting of seeds by organic solvents tend to denaturate oleosin. Renaturation is a known technique in protein chemistry in transforming biologically inactive proteins and their aggregates back to biologically active forms. This technique is commonly used, for example, in the context of the proteins in the inclusion bodies produced by recombinant Escherichia coli. Usually, the renaturation of denaturated or aggregated proteins are unfolded by chaotropic salts, detergents, or other organic compounds in reducing conditions and then the unfolding agent is removed. Oleosin is relatively small protein, which can refold whenever the conditions of the renaturation are met. According to the present invention, oleosin from defatted oil seeds can be extracted by oil bodies from aqueous solutions. Previously such renaturation of oleosin has not been considered possible. According to the present invention refolding of oleosin can be created at the interfacial junction of lipid-water. The basic embodiment of the present invention is the established and unique structure of oleosin in the lipid-water interphase, which enables to advise several useful functions and technologies for oleosins. Analysis of conjugated dienes and volatile oxidation products and the protective ability of oleosins against oxidative agents were shown in the present invention (Figure 1). Moreover, in contrast to previous studies, we observed that the deleted N-terminal part of oleosin prevents the efficient lateral transportation of the oleosin from rER to oil bodies (Figure 2) but then the oleosin membrane loses its capability of protecting oil bodies against chemical reactions. Moreover, the sizes of such oil bodies were enlarged compared to oil bodies covered with full- length oleosin proteins. Oil bodies covered with truncated oleosin molecules were easily stained with neutral, lipid-specific stain, Nile Blue A, whereas oil bodies covered with intact oleosins were not. As Nile Blue A is a molecule with small molecular size of 730 Da (Spiekermann et al. 1999. Arch Microbiol 171 :73-80), also this result shows that surface of oil bodies is covered by a tight net of oleosin molecules that will not allow small molecules to pass or diffuse through. Thus, intact oleosin structure is required for protection of the lipid phase and preservation of structural integrity of oil bodies.

Structure of central hydrophobic domain has been shown to be an α-helix with FT-IR or CD spectroscopy (Alexander et al. 2002, Planta 214:546-551), but also to be a β-strand with a conventional prediction methods and spectroscopic analysis (Li et al. 1992, J Biol Chem 267:8245-8253; Li et al. 2002, J Biol Chem 277:37888-37895). These divergent models describe the potential of anchoring of the oleosin into the membrane of oil body, but not the assembly of oleosin together with other oleosins. It is also noted that NMR spectroscopy is severely restricted for the study of membrane protein in the reconstitution conditions, because of non-isotropic motion of membrane proteins in lipid phase (Li et al. 1993, J Biol Chem 268:17504-17512). Thus, the present models of oleosin localisation, including all three (N-, central, and C-terminal) domains, are based on careful pH studies, in which charges of oil body and oleosin are calculated together resulting deduced conformational structure of oleosin (Li et al. 1992). Problems in membrane protein secondary structure determination in aqueous medium or as dry film may reflect incorrect structural properties of proteins, because they are not in natural lipid environment. Therefore, we noticed, that generation of an alternative model including folding aspects of lipid phase results not only in more plausible explanation for oleosin-lipid interaction, but also new insight between adjacent oleosin molecules. Similar models have not been created previously, due to lack of programs that could predict protein structure at lipid:water interphase. Therefore, together the biological observations made with truncated oleosin-GFP fusions, the structure of oleosin proteins was analyzed by an empirical approach followed by various protein structure prediction programs. Using computer modeling, a segment of the first 23-28 amino acids in the N-terminus of oleosin protein was identified, which could interact with the phospholipids in the core of the oil body, whereas the aspartic residues of the same segment could participate in the formation of the negatively charged surface of the organelle. As a result, the whole segment of Arg-Asp-Arg-Asp-Gln-Tyr-Gln-Met (amino acids 23-30 in the N-terminus) could be β -strand-structured at the lipid:water interface, and therefore essential for coverage of the surface of oil bodies. Similarly, two amphipathic α-helical structures were identified in the C- terminus of oleosin protein. Further analysis of this segment suggested that positively charged residues in these α-helices interact with phospholipids on the surface of oil bodies, and the negatively charged residues are oriented towards the exterior. The sequence of residues 162-167, which is located after two amphipathic α-helical structures, is very similar to sequence of residues 23-28 of N-terminal domain. Both of these segments are extremely rare among known proteins; besides known oleosin structures, homological sequences were identified only against segment 162-167 in synaptotagmin C, which is a synaptic vesicle protein and a putative trigger of exocytosis in animal systems (Zhang et al. 2002, Genesis 34:142-145). This implies indirectly that domains found in oleosin protein are important for protein function particularly in water- lipid environment.

According to the secondary structure predictions, segment of residues 162-167 of oleosin protein does not possess α-helical or β-strand structures. However, because of the distribution of the positively and negatively charged residues, we deduced that this sequence produces β -stranded conformation in lipid phase. Recently, it has been shown that such intermolecular formation of β- sheet may take place in lipid- water emulsions (Lefevre and Subirade 2003, J Colloid Interface Sci 263:59-67). Thus, N- and C-termini of oleosin can bind intermolecularly to each other forming a continuum of β-sheet-mediated "zipper" shell in the oil body surface. By analogy with the experiments obtained with the outer surface protein A (Koide et al. 2000, Nature 403:456-460), conformation of such shell will be highly stable. The spatial structure of oil bodies is, further, determined by interactions between oleosins and oil bodies. The hydrophobic central domain of oleosins modeled as an α-helical hairpin (Alexander et al. 2002, Planta 214:546-551) anchors the zipper tightly into oil bodies (Figure 3 a and b). In conclusion, the simplest stable structure to explain interactions obtained by computer analyses is the single-layer of antiparallel β -sheets, that are formed by the putative β-stranded segments of amino- and carboxy-terminal domains, and a single-layer parallel α-helical sheet, which is built by the amphipathic α-helices of carboxy-terminal domains (Figure 3). These results and those of expression of plant oleosins in yeast and animal cells (Ting et al. 1997, J Biol Chem 272:3699-3706; Hope et al. 2002, J Biol Chem 277:4261-4270 ), and the rare existence of oleosin domains show that oleosin-like protein structures are unique in their ability to function at water-lipid interface to cover oil bodies and to facilitate packing of lipids into oil bodies. Giving that oleosins form extremely tight net to cover lipids in oil bodies suggests that their structure may have key importance in preventing chemical alterations of lipids during their storage in seeds.

Whereas the central hydrophobic domain of oleosin may be essential in binding the zipper assembly onto lipid bodies, it is evident that the central part of oleosin can be readily replaced with another hydrophobic domain by the methods of gene engineering. Then the protein is to be considered different from any known oleosin and functionally different from recombinant oleosins described in the prior art. The zipper domains themselves, without the central domain have a weak affinity to the lipid interphase which can be adequate for protection from certain chemical reactions. The recombinant zipper proteins involving the central domain of desired properties can be produced by the known gene-engineering techniques.

The found zipper structure can have a variability of applications in biotechnology other than the protection of lipids from chemical reactions of oxygen and other chemicals. Not only lipid interfaces can be covered by these structures but also any hydrophobic - hydrophilic interface. In such cases the anchoring centre domain can be artificially designed by computer modelling. The counterparts of the zipper can be in the same molecule like in oleosin or both of them can carry different anchoring domains. The zipper cover can be used for prevention of molecular trafficking from or into hydrophobic phase. Usually such artificial covers are produced by gene- engineering methods, but also peptide synthesis is possible.

The invention is further illustrated by the following non-limiting examples. EXAMPLE 1 Cloning of oleosin and GFP gene sequences Full-length oleosin-GFP fusion was described earlier (Wahlroos et al. 2003, Genesis 35:125- 132). The oleosin sequence is shown in the figure 4. Arάbidopsis thaliona oleosin gene carrying N-terminal deletion of 66 amino acids was created with PCR and cloned in-frame with GFP. Primers for PCR of oleosins were: OLE66-forw: 5'-GCACCATGGTTGGAACTGTCATA OLE66-rev: 5 '-GAACTCGAGAAGTAGTGTTGCTG. The schematic construct is shown below:

EXAMPLE 2 Plant material and biolistic gene transfer Tobacco plants (Nicotiana benthamiana) were grown in soil in a greenhouse at 24 °C. Plants were illuminated using fluorescent lights (Lucalox®, type LU400/HO/T/40) operated on a 16 h day photoperiod. Mature, full-expanded, leaves were subjected to particle bombardment. Particle bombardment was performed by the rupture disk method with a high-pressure helium-based apparatus PDS-1000 (Bio-Rad). Tungsten particles were prepared by vortexing of 60 mg in 70 % ethanol for 3-5 minutes, followed by incubation on a bench for 15 min. Mixture was pelleted by short spinning and the supernatant was removed. Sterile water was added onto pellet and vortexed for 1 min. Particles were allowed to settle for 1 minute and spinned for 2 s. The supernatant was removed. This was repeated three times. Sterile 50 % glycerol was added to bring particle concentration to 60 mg/ml. For each particle bombardment, DNA was precipitated onto tungsten particles (M-20, 1.3μ) with calcium chloride and ethanol. 5-10 μg of plasmid DNA, 50 μl of CaCl2 (2.5 M) and 20 μl of spermidine (0.1 M) were mixed and vortexed for 2-3 min, allowed to settle and spinned for 2 s. Supernatant was removed and 140 μl of 70 % ethanol was added onto surface of pellet, removed and 100 % ethanol added and removed without disturbing the pellet, and repeated. 6 μl of suspension was pipetted onto macrocarrier and used for bombardment. A detached leaf oϊNicotiana benthamiana (15-30 mm size) was placed in the center of a plastic Petri dish and bombarded on a solid support at a target distance of 7 cm. Bombardment was done with a pulse of 1350 kPa helium gas in a vacuum chamber. Inoculated leaves were analyzed 24 hours after particle bombardment. GFP fluorescence was monitored using a confocal laser scanning imaging system MRC-1024 (Bio-Rad).

EXAMPLE 3 Staining of oleosin-GFP mutants localized in oil bodies with neutral lipid stain, Nile Blue A Tobacco leaf cells expressing oleosin-66-GFP, were infiltrated with an aqueous solution (100ng ml'1) of fluorescing, lipid-specific stain, Nile blue A (Sigma-Aldrich, Helsinki, Finland; Spiekerman et al. 1999, Arch Microbiol 171:73-80), 24 h after particle bombardment. GFP fluorescence and Nile blue A fluorescence were visualised using appropriate confocal microscope fluorescence filters as done earlier (Wahlroos et al. 2003, Genesis 35:125-132).

EXAMPLE 4 Computer analyses Proteomics tools used in this study are available in http://www.expasv.ch/tools/ web site. Following methods were used GORl, GOR3, G0R4, ScanProsite, P ATTINPROT, Jpred, rmPredict, SOPM, SOPMA, HNN, SSP, NNSSP, and SSPAL.

EXAMPLE 5 Analysis of conjugated dienes and volatile oxidation products Oxidation stability tests were done with intact and heat deactivated product. Heat-deactivation was done at 90 °C for 5 min under nitrogen atmosphere. Stability test was done in closed test tubes at 37 °C. Oxidation was followed three times per week analysing conjugated dienes and volatile oxidation products, such as propanal, which are typical lipid oxidation products (Chan and Coxon 1987, Academic Press Inc., London, pp 17-50). Conjugated dienes were measured spectrophotometrically at 234 nm in methanol. Volatile compounds were measured with headspace gas cliromatograph (Autosystem XL, Perkin Elmer, USA). Each measurement was repeated twice (Figure Ia and Ib).

EXAMPLE 6 Visual analysis of seed oil oxidation and protective effect of oleosin A commercial box of margarine ("Flora", Finland) was coated with oleosin-oil bodies prepared by the method described earlier (Parmenter et al. 1995, Plant MoI Biol 29:1167-1180). The surface of the margarine was sprayed evenly with the oil body suspension so that the surface looked wet. The margarine was then allowed to stand for 2 weeks at the room temperature. The control experiment was carried out using rape seed oil containing 5-10 % of bovine serum albumin (Sigma, fraction V) and 1 % Triton X-IOO5 the blend micellized into 3 volumes of water with strong agitation. The oleosin-oil body treated margarine looked less yellow by visual inspection. EXAMPLE 7 Protection of unsaturated oils with oleosin-oil bodies Fish liver oil from chemist shop was mixed 1 part fish oil and 5 parts of oil bodies dry volume (prepared as in Example 3) and 0.5 part of soya lecithin with a high speed stirrer ("Ultra Turrax") during 1-2 min to get a homogenous mixture. The control experiment was performed with a similar mixture but instead of oil bodies, the rape oil-bovine albumin, 1 % Triton X-IOO - mixture was employed. Organoleptic analysis showed that the smell of typical rancid fish oil was stronger within the control. Related experiments were also performed with ω-3-fatty acid containing fatty acids of the camilon oil stored at + 37 0C for 2 weeks. The 1H NMR analysis of the samples in CDCl3/ CD3 OD showed a higher loss of the saturated protons in the control experiment.

EXAMPLE 8 Preparation of oil bodies Several different oil seeds was used for the preparation of oil bodies, including soya, rape, false flax (Camelina sativά), oats, canola (Brassica napus), spring turnip rape (Brassica rapa), and sesam oil. The seeds were used directly or after cold processing removing the majority of the oil while leaving plant remnants and proteins in the pellet. Oleosin in defatted (by extraction with organic solvents) pellet could be regenerated by extraction with oil micelles.

The seeds (or the seed residue/pellet after pressing the oil) are crushed and made mechanically into a fine powder of less than 50-100 μm diameter on the average (controlled by microscopy). The powder (1 kg) is dispensed into water (10 liter) by strong mechanical agitation, which optionally can also involve ultrasonic oscillator treatment at a high energy. Temperature of the solution must stay below 6O0C during the process. Preferably the suspension is conditioned for 1- 5 h before and allowed to cool to +0-50C. The suspension is then filtrated using a standard drum filtration equipment operating at a reduced pressure. The optimal filter is of about the similar pore size as is used for the recovery of bakery yeast. The filter can be covered with diatomaceous earth for preventing clogging of the filter. The substance recovered from the blade of the drum filter can be returned to the process after homogenization and refiltrated to increase the yield of oil bodies/seed mass.

The filtrate contains remnants of oil bodies and soluble seed proteins including oleosin. For many technological purposes it is not necessary to purify the oil body fraction as described in the technologies described in the context of known centrifugation methods of preparing oil bodies. Usually it is advantageous to reduce the content of water from the oil bodies for the storage. This can be done with ultrafiltration, freeze-drying of, or with drying by the air cyclone technique at the flowing air temperature of 40-700C. The oil bodies can be stored for several months at temperatures near O0C. Preferably, the water content of the oil bodies is kept at 10-15 %. The ratio of oil/hydrophobic material per protein depends on the application of the preparation. In many applications it is not necessary to purify the protein fraction. However, if it is needed the content and purity of oleosin can be controlled by the temperature, strength, and time of treatment of the first extraction process. The minimum content of protein in oil bodies is 5 % (w/w, protein/ lipid).

EXAMPLE 9 Construction of an artificial recombinant zipper protein The N- and C-terminus of oleosin by PCR and the movement protein (MP) of Tobacco mosaic virus (TMV) was cloned in the middle of these two fragments in translatable fusion. TMV-MP is a known α-helical, hydrophobic membrane protein which binds to cellular membranes (Brill et al. 2000, Proc Natl Acad Sci USA 97, 7112-7117).

To clone 3OK gene from TMV genomic RNA, two specific oligonucleotide primers were chemically synthesized. These primers were: N30K S'-GCGGAATTCCCATGGCTCTAGTTGTTAAAGG) C3 OK 5'-AGACCTCGAGGAAACGAATCCGATTCGGCGAC. The N30K and C30K primers contained an EcoBl restriction site and Xhol restriction site, respectively. The first strand cDNA was synthesized from TMV genomic RNA using 20 nM of C30K primer, 1 mM dNTPs and 1 unit of AMV in IX AMV reverse transcriptase (RT) reaction buffer (50 mM Tris-HCl; 8.3, 50 mM KCl, 10 mM MgCl2, 0.5 mM spermidine, 10 mM DTT) for 45 min at 370C. After that 5 μl of reaction mix was subjected to PCR reaction in 1 x PCR buffer (10 x buffer: 500 mM KCl, 100 mM Tris-HCl; pH 9.0 at 250C and Triton X-100 ) with 1.5 mM MgCl2, 2μl of Taq polymerase (5U/μl), 0.2 mM of each dNTPs, primers at 0.4 μM in 25μl reaction volume. The template was denatured with heating for 3 min at 950C, and 28 cycles of PCR were carried out with iCycler (Bio-Rad) thermal cycler with denaturation at 950C for 1 min, primer annealing at 680C for 1.5 min, primer extension at 720C for 2 min, with a final elongation step after 30 cycles at 720C for 10 min. The resulting DNA product was cleaved by EcoRI and Xhol and cloned into iscoRI-Jttøl-digested pGΕM-7Zf(+) (Promega Corporation, USA; catalogue number P2251). Following restriction analysis of the recombinant clones and sequencing, plasmid pGEM-30K was selected for further work. In this plasmid, the termination codon of the TMV 30K gene was replaced by a XIwI site for subsequent fusion with N- and C- termini of oleosin. These were cloned in fusion to MP. Oleosion-specific primers were designed according to full-length oleosin (Fig. 4.). This Constructs N-MP-C-ole was used to constructs artificial oil body membranes.

EXAMPLE 10 Preparation of artificial oil body membrane Previously it has been shown that plant oleosin and domains of Hepatitis C virus have similarities and can be used to parallel complementation of vesicle structures (Hope et al. 2002, JBC 277; 4261-4270). Using this scheme as a model, we constructed a hypothetical N- and C- terminal oleosin molecule with central domain replaced by Tobacco mosaic virus movement protein. This protein is a known membrane-bound protein. This constructs was cloned in fusion to GFP (Wahlroos et al. 2003. Genesis 35: 125-132) into binary vector for agroinfiltration into plant tissues. After 5 days, the localization of the construct was visualized by confocal laser scanning microscope and found to be similar to that of oleosin-GFP (Wahlroos et al. 2003, Genesis 35: 125-132). Thus, we have created artificial oleosin oil body complexes that are independent of the native central domain but highly dependent in structural integrity of N- and C-termini of oleosin molecules in accordance to the present invention.

EXAMPLE 11 Protection of a drug from oxidation Vitamin A dissolved in olive oil (50 mg/ml) in the presence of lecitin (1209 mg/ml of olive oil) is micellized in ratio 1:20 (v/v) to water with a stirrer (20,000 rpm) on ice in nitrogen atmosphere. Two grams of oil bodies (Example 8) is added to the mixture in the same conditions. A control is made equally without oil bodies. The liquids were allowed to stand at the room temperature and were analyzed by UV at the absorption maximum of vitamin A. The liquid containing oil bodies were more stable for oxidation of vitamin A than the control.