WO2000018416A1

WO2000018416A1 - Antioxidant compositions and process for their preparation

Info

Publication number: WO2000018416A1
Application number: PCT/NZ1998/000146
Authority: WO
Inventors: Lai Yeap Foo
Original assignee: Industrial Research Ltd
Current assignee: Callaghan Innovation Research Ltd
Priority date: 1998-09-30
Filing date: 1998-09-30
Publication date: 2000-04-06
Anticipated expiration: 2001-03-30
Also published as: AU9369198A

Abstract

The invention relates to a food-, cosmetic- and pharmaceutically- acceptable extract of an Oenothera species containing antioxidant is mainly prepared using polar solvent extraction. The invention includes ingestible and topical compositions comprising the extracts. Particularly preferred are abstracts containing 1, 2, 3, 4, 6-pentagalloyl glucose.

Description

ANTIOXIDANT COMPOSITIONS AND PROCESS FOR THEIR PREPARATION

The invention relates to antioxidant compositions and processes for preparing them.

While the need for oxygen to sustain life is universally understood its less desirable and destructive nature is perhaps less recognised. Organic materials be they foods or biological systems are subjected to the ravaging effect of oxygen from the moment they are produced and effective anti-oxidative agents need to be in place to prevent their degradation. Two of the most susceptible components to oxidative degradation are protein, and lipids. In the natural state in living tissues such as plants and animals, lipids seem stable enough to perform their function but when they are extracted and processed into a food system they become exposed to the inevitable oxidation process leading to changes in the chemical composition and rancidity.

Oxidative spoilage of refined vegetable oils and foods containing them are associated almost exclusively with unsaturated fatty acids. The reactions may be catalysed by several factors including UV light and metal ions and involved free radical participation. The most susceptible portions of the fatty acids are in the vicinity of the carbon-carbon double bonds which readily react with free radical species. These reactive intermediates propagate and degrade to give rise to a host of volatile acids, ketones and aldehydes to give rise to the characteristic "rancid" odours. Antioxidants are widely employed to prevent these oxidative degradations and they represent one of the oldest established groups of modern food additives.

The oils in oil seeds are known to be stable even after prolonged storage of the seeds and researchers attempts to emulate nature in stabilising the oils have met with only limited success. One of the published records of scientists appreciation of the efficiency of nature's method in this respect can be traced to Hickman in his paper¹ on practical application of physiological antioxidants in which he made the following observation. "It was Professor Mattil and his various pupils who showed us that an unprotected oil that might go rancid in an hour (in the Swift Tester) could be, for instance, protected for 8 hours with α-tocopherol, for 50 hours with α- tocopherol plus phospholipid, for 200 hours with α-tocopherol, phospholipid and ascorbic acid and 300 hours with all the above plus tannins. Thus was offered an insight into how nature stabilised vegetable fats for relative vast periods with meagre supplies of antioxidants . . .". The remarkable synergetic effects between various natural components in giving enhanced stability against oxidative degradation have been noted since early time and recent investigations of this effect are centred mostly around α-tocopherol. The oxidative degradation reaction is highly complex involving free radicals and influence by a range of factors including exposure to UV light, enzymes, metal ions and pro-oxidants. Antioxidants can act by inhibiting any of these catalysing agents as well as interrupting the free radical chain reaction by serving as scavengers or terminators of free radicals. These compounds come in many chemical forms and in the natural state these compounds may act concurrently, serially, additively and synergistically to impart the kind of stability for example in whole or intact oil seeds.

The performance of an antioxidant is known to be highly specific. α-Tocopherol for instance serves as a better antioxidant in animal fats than vegetable oils and that the response of a specific substrate to any antioxidant can depend on the stereochemistry of the antioxidant molecule in biological systems and in processed foods. Evans and Reynhout² reported that certain antioxidants in model systems appear to be specific for one or more of the radicals formed in the oxidation process and radical formation varies with the degree of unsaturation, carbon chain length, enzyme activity, temperature and trace metals. Photoxidation affects the cis-trans configuration of breakdown products and the response to antioxidants and therefore oxidative reaction is more effectively inhibited by a combination of phenolic molecules that are specific for the substrate environments and the decomposition intermediate.

Recently there has been recognition that the life-giving gas, oxygen, is also a source of degeneration, disease and aging. This is due to the reactive nature of the oxygen molecule or associated radicals such as the superoxide, the highly reactive hydroxy radical, that cause damage in the human body, attacking DNA leading to dysfunction, mutation and cancer. They can attack cell membranes enzymes and proteins, disrupting normal cell activities. Unsaturated fatty acid components are very susceptible to these attacks and cause damage in cells that line our arteries leading to their hardening and thickening and eventually to heart attacks and strokes. Free radical attacks on the protein in collagen can cause cross-linking resulting in stiffness in the body, reduce suppleness and increase wrinkling in the skin. It has been estimated that at least 85% of chronic and degenerative diseases are the result of oxidative damage³.

Given the hostile environment that living organisms are exposed to, it is miraculous that they survive and thrive as they do. Living cells maintain an elaborate anti-free radical defense systems known as antioxidant systems. Among the enzymes one of the best known for this is superoxide dismutase which disarms the superoxides by altering them to a much less reactive form. The body also makes use of vitamins like C and E, trace minerals like selenium and other nutritional components like β-carotenei and phenolic constituents to cope with free radicals. Many researchers believe that to combat free radical damage effectively the general levels of the free radical fighting nutrients need to be higher than nutritional experts previous thought. Support for this view is available from a Finnish study of 12,000 people over a four year period found that those people with low levels of antioxidants in the bloodstream were eleven times more likely to get cancer⁴. In fact the free radical research has produced a new approach to good nutrition. While vitamins and minerals are identified with the recommended Daily Allowances published by health authorities to enable the body to function correctly, antioxidants are required to ward off the constant waves of free radical. For this purpose, research makes it seem likely we must not only take regular but larger doses of antioxidants than we have previously thought necessary.

One particularly effective and desirable antioxidant is 1, 2, 3, 4, 6 - pentagalloyl glucose (I).

It is an object of the invention to provide new antioxidant preparations with useful properties and/ or methods for the purification of 1, 2, 3, 4, 6 - pentagalloyl glucose and/ or provide the public with a useful choice.

In accordance with one aspect of the invention there is provided an extract of an Oenothera species comprising antioxidant. Preferably the extract comprises 1,

2, 3, 4, 6 - pentagalloyl glucose. More preferably the extract further comprises at least one of catechin (II) and procyanidin gallate (III). Most preferably the extract comprises 1, 2, 3, 4, 6 - pentagalloyl glucose, catechin and procyanidin gallate.

Preferably the species is Oenothera biennis (evening primrose). Preferably the extract is a seed extract.

where R = H or

and n = 1,2 . m - 1,2 Preferably the extract is a polar solvent extract.

Preferably the polar solvent comprises at least one of acetone, ethanol, methanol and water.

Preferably the polar solvent is aqueous ethanol, aqueous acetone or aqueous methanol.

In a preferred embodiment of this aspect of the invention evening primrose seeds after oil removal are treated 10- 100% (preferably 30-70%) acetone, 10- 100% (preferably 40-90%) ethanol or 10- 100% (preferably 40-90%) methanol.

In another aspect of the invention there is provided a method for preparing an antioxidant preparation comprising extracting seeds of an Oenothera species with a polar solvent.

Preferably the polar solvent comprises at least one of acetone, ethanol, methanol or water.

Preferably the polar solvent is aqueous ethanol, aqueous methanol or aqueous acetone.

Most preferably the extraction is carried out with 10- 100% (preferably 30- 70%) aqueous acetone, 10- 100% (preferably 40-90%) methanol or 10- 100% (preferably 40-90%) aqueous ethanol and the starting material is evening primrose.

The preferred Oenothera species is Oenothera biennis.

In a preferred embodiment there is provided a method for preparing a 1, 2, 3, 4, 6 - pentagalloyl glucose extract comprising the step of extracting seeds of an Oenothera species with a polar solvent.

Preferably the polar solvent comprises at least one of acetone, ethanol, methanol or water. Preferably the polar solvent is aqueous ethanol, aqueous methanol or aqueous acetone.

The 1, 2, 3, 4, 6 pentagalloyl glucose extract may be further purified by chromatography, preferably by adsorption chromatography, more preferably by chromatography on Sephadex LH20.

The product of the above process is a preferred antioxidant for use in ingestible and topical compositions. It may be used in such compositions either alone or in combination with other antioxidants including other antioxidants from an Oenothera species eg catechin and/or procyanidin gallate.

In a further aspect of the invention there is provided an ingestible composition comprising an antioxidant-containing extract from an Oenothera species, preferably Oenothera biennis.

Preferably the ingestible composition comprises a preferred extract of a previously described aspect of the invention.

Preferably the ingestible composition is prepared in a unit dosage form containing 0.1 mg- 1000 mg (preferably lmg-250 mg) of extract. Most preferably the unit dosage form is a capsule or tablet.

In a further aspect of the invention there is provided a topical composition comprising an antioxidant-contai-oing extract of an Oenothera species. Preferably the extract is a preferred extract of a previously described aspect of the invention.

Preferably the topical composition also comprises a cosmetic base.

In a preferred embodiment of the invention a 60% aqueous acetone extract was prepared from spent Oenothera biennis seeds from which the oil had previously been removed. The extracts are concentrated by a rotary evaporation and the residue diluted with water and dried to produce a powder. This powder may be incorporated into tablets or capsules eg gelatin capsules.

Alternatively for topical compositions of the invention they may be incorporated into a cosmetic cream. Such a cream may be prepared using either a water in oil emulsion or an oil in water emulsion. Tablets, capsules and topical compositions may be prepared in a wide variety of ways known to those skilled in the preparation of pharmaceuticals, cosmetics and the like. The critical feature is that the tablet, capsule or topical composition contains an antioxidant composition of the invention.

EXAMPLES

The following examples further illustrate the practice of the invention. Ratios of solvents are by volume unlessj otherwise indicated.

EXAMPLE 1

Materials: Linoleic acid 99%; and b-carotene 95% were purchased from Aldrich Chemical Company. Tween 20 came from Serva Feinbiochemica GmbH & Co. Spent evening primrose seeds were obtained from Industrial Research Limited's Chemical Processing Unit.

Extraction: The spent seeds ( 100 g) were treated with 60% aqueous acetone (500 mL) with occasional shaking in a conical flask at ambient temperatures for several hours. The mixture was filtered and the residue treated twice more with 50% acetone. The combined extracts were concentrated under a rotatory evaporator and the residue diluted with water (50 mL) and freeze dried to produce a brown amorphous powder (13.0 g).

In a separate experiment, a new batch of spent evening primrose seeds

(100 g) was sequentially extracted with ethyl acetate, ethanol, and finally water. The extracts were concentrated and freeze-dried separately to produce ethyl acetate extract (5.3 g), ethanol extract (6.3 g), and water extract (9.0 g).

Chromatography: The 60% aqueous acetone extract was fractionated over a column of Sephadex LH20 prepared in 15% aqueous methanol. Extract was dissolved in minimum of this solvent and column eluted with aqueous methanol containing increasing proportion of methanol (15% → 100%), and finally the column was washed with 60% aqueous acetone. The fractions were monitored by spotting on cellulose TLC plates, developed with 6% aqueous acetic acid, and visualised by spraying with ferric chloride-potassium ferricyanide spray reagent. Fractions containing similar types of materials, judged by their RF values on the TLC plates, were combined and concentrated before freeze-drying.

Antioxidant Activity: The detection of antioxidant activity of the fractions was done by the method of Nakamura⁵. Fractions were separately spotted on a silica gel plate and the plate sprayed with a dilute solution of β-carotene, followed by irradiation with ultra-violet light. After several minutes of exposure, the plate was decolorized and fractions with antioxidant activity appeared as yellow spots.

The antioxidant activity of extracts was measured using the coupled β- carotene/linoleic acid reaction of Miller⁶. In this process, 2.0 mL of β-carotene solution in chloroform (0. 167 mg/mL) were mixed with 50 mg/g linoleic acid and 400 mg of tween 20. Chloroform was evaporated off in a rotovapor and 100 mL of oxygenated distilled water was added to the residue in small volume with vigorous agitation so as to form an emulsion. 5.0 mL of this were added to tubes containing samples (10- 100 mg/mL) for evaluation. The absorbance of the mixtures at 470 nm were measured immediately after the samples were mixed at 35°C, and then at regular intervals. In most cases, each extract or fraction was evaluated using 3 or 5 different concentrations in duplicates.

Calculation: Calculation of antioxidant activity coefficient⁶ was made using the expression:

AAC=[A_E(t)-A_C(t)]/[A(c(θ)-A(c(t))

AAC is the antioxidant activity coefficient ranging from 0- 1000; AE is absorbance at t min for extract or sample; and Ac(o) and Ac(t) are the absorbances of the control at 0 and t min, respectively. Antioxidant Activity of Extracts: The rate of decline of the β-carotene 470 nm peak is related to the effectiveness of the antioxidant activity of the extracts. When compared with the control, all extracts showed varying degrees of antioxidant activity. As expected, the hexane extract produced the least amount of materials with the lowest antioxidant activity. The. evening primrose seeds, after all, had been previously stripped with hexane so that most hexane soluble materials would have been removed. Both the 60% aqueous acetone and EtOH extracts showed good antioxidant activity (Table 1), as measured by the coupled β- carotene/linoleic acid reaction with the EtOH being a little superior to aqueous acetone. Extracts produced with ethyl acetate and water also possessed activity, but less so than the EtOH or aqueous acetone. In an attempt to optimize further the activity of the crude extracts, 20%, 50% and 80% aqueous EtOH were used as solvents, however, their antioxidant activity were all comparable to one another (Table 1).

The antioxidant activity of these extracts were all concentration dependant with the most marked increase from 10-50 mg/L of extracts, but the rate of the increase was reduced at higher concentrations. H₂0 extract also exhibited good antioxidant activity, significantly superior to EtOAc extract, showing that the most active components were likely to be polar compounds.

When linoleic acid was excluded from the experiments, the β-carotene was relatively stable well beyond the 90 min period of the measurements. This behaviour would suggest that the β-carotene was degraded by the linoleic acid oxidation products. The mode of action of the extracts would therefore appear to be the inhibition of linoleic acid oxidation, probably by serving as free radical scavengers and by blocking the oxidized linoleic acid from reacting with the β- carotene molecule leading to the latter breakdown.

Isolation of Compounds with Antioxidant Activity: Treatment of the

60% aqueous acetone extract on a Sephadex LH 20 column gave a major active fraction that was further purified by an MCI CHP 20 to produce pure compound. Assessment of the antioxidant activity of this compound by the coupled β-carotene linoleic acid reaction showed it had good activity in a concentration dependent manner. The chemical constitution of the pure sample was analysed by NMR. The ¹ C NMR spectrum of the sample showed 15 carbons altogether, with 3 being aliphatic carbons and the remaining being aromatic. Its phenolic character was apparent by the presence of quaternary low field peaks: 3 at 156.5, 157.0 and 157.4 ppm (characteristic of a phloroglucinol ring); and 2 at 145.5 and 145.6 ppm (characteristic of a catechol ring). The compound produced a reddish coloration when sprayed with vanillin-HCl, and a blue coloration with ferric chloride-potassium ferricyanide reagent, suggesting it to be a flavan. This deduction was also consistent with the remaining carbon chemical shifts in which the three aliphatic carbons (28.2, 68.0, and 82.3 ppm) were attributed to the heterocyclic C-ring. The downfield shift (82.3 ppm) assigned to the C-2 of the C-ring indicated a 2,3-trαns relationship, and this was confirmed by the large proton-proton couplings observed for their corresponding proton chemical shifts. The compound was therefore identified as catechin (II)⁷ and confirmed by a chromatographic comparison on 2D TLC with authentic sample. Antioxidant activity results for this compound are shown in Table 1.

A proanthocyanidin fraction, obtained by eluting the Sephadex LH20 column with 60% aqueous acetone after it had been through with the EtOH described earlier, also showed good antioxidant activity. The freeze-dried material was analysed by ¹³C NMR, which showed many of the features exhibited by catechin, thus indicating the proanthocyanidin structure based at least in significant proportion on the catechin structure. The relatively strong peak (-76 ppm) in the heterocyclic region indicated that an epicatechin⁷ structural unit also was involved. In addition, a relatively strong signal (166 ppm) was present, and the high field position of the signal indicated the presence of ester carbonyl units. Also present were ¹³C NMR peaks at 1 10, 120, 139 and 145 ppm, which were consistent with gallic acid functionality, and the material was hence a procyanidin gallate polymer. The observation of multiple ¹³C NMR signals in the general upfield region (34-38 ppm) attributed to substituted benzylic carbons) provided the evidence that the gallic acid units formed ester linkages on some of the hydroxyls of the heterocyclic C-ring of the catechin or epicatechin structural units (III)⁸ _' ⁹.

Another major polyphenol with low Rf value (0.06) in 6% aqueous acetic acid was isolated and purified by repeated chromatography on Sephadex LH20 with

50% aqueous ethanol as eluant. The ¹³CNMR spectrum of the compound showed six peaks at 63.0, 69.5, 71.9, 73.4, 73.8 and 93.4 ppm which were characteristic of a sugar moiety. In addition there were four clusters of aromatic peaks centering at about 110.2, 120.5, 139.5 and 146.0 ppm as well as a group of carbonyl chemical shifts in the 166.0 ppm region all of which were consistent with galloyl moieties. The carbonyl signals were sufficiently resolved to reveal five peaks (165.2, 165.9, 166.0, 166.2 and 166.6 ppm) indicating the presence of five galloyl moieties. This chemical constitution was also corroborated by ^!HNMR which showed the presence of five sets of two proton singlets at d6.86, 6.89, 6.94, 6.98 and 7.05 corresponding to the five sets of degenerate H-2, H-6 of the galloyl ring protons. In addition, the sugar protons were all observed in the downfield region of the spectrum indicating their respective sugar hydroxyls were all acylated thus establishing the compound to be 1, 2, 3, 4, 6 - pentagalloyl glucose. The structure was also confirmed by spectra comparison with published data¹⁰ on the compound.

Antioxidant Activity Coefficient: The antioxidant activity of the various extracts determined by the coupled b-carotene/linoleic acid reaction was expressed as the antioxidant activity coefficient used by Mallet¹¹. The maximum value of 1000 indicated a perfect antioxidant, an ideal system that existed only in theory. The higher the value the more efficient the sample, and a value of zero indicated that the sample had the same activity as the control. A sample that showed pro-oxidant activity produced negative values.

Table 1 Antioxidant Activity Coefficient (AAC) of Extracts from Spent Evening Primrose Seeds

ACC

Sample 10 mg/L 50 mg/L 100 mg/L 200 mg/L

60% Acetone ext 106.58 569. 16 794.78 857.14

H₂0 ext 43.08 193.88 399.09 563.49

EtOAc ext 44.12 1 18.70 223.74 64.08

EtOH ext 285.71 728.99 835.08 890.76

Hexane ext 1 10.92 58.77 193. 12 405.35

Hexane ext 2 ,20.99 72.63 203. 19 470. 19

20% EtOH ext 58.76 307.69 722.22 686.97

50% EtOH ext 47.01 387.82 488.25 662.39

80% EtOH ext 130.82 534.37 652.99 773.84

Catechin 175.63 426.52 598.57 779.57

Procyanidin gallate 60. 14 436.43 664.56 809.07

The results summarised in Table 1 show that all examined extracts exhibit antioxidant activity, with polar solvents offering the better antioxidant extracts. Catechin and procyanidin gallate are some of the principal antioxidant components present in evening primrose seeds.

EXAMPLE 2

Test sample: The test sample was prepared by extraction of spent evening primrose seeds (after oil removal) with aqueous acetone. The extract was concentrated and the resulting aqueous suspension diluted with H₂0 and filtered over glass wool and the filtrate purified over XAD-4 resin to give the "activated" extract as freeze-dried powder. The sample was dissolved in DMSO prior to testing.

Chemicals: 2-Amino- l-methyl-6-phenylimidazole [4,5-b] pyridine (PhIP) was obtained as a powder from Toronto Research Chemicals Inc and was dissolved in DMSO to form a 1 mg/mL solution. Cytochalasin B was obtained as a powder from Sigma Chemical Co USA and was dissolved in DMSO to form a 2 mg/mL solution. Hoechst 33258 DNA specific fluorescent stain was obtained from Hoechst and dissolved in 50% ethanol to form a 80 mM solution.

Chinese Hamster Cells: The Chinese hamster cell line V79-5 was originally obtained from W R Inch (London, Ontario, Canada) and has been maintained in the Cancer Research Laboratory since 1979. It is stored as frozen aliquots in liquid nitrogen and, once thawed, it is maintained in a CO₂ incubator at 37°C by weekly trypsinisation and subculture in T25 flasks. The culture medium used for maintaining the cell line is α-MEM, without ribo- and deoxyribo- nucleosides containing 10% heat inactivated foetal calf serum from Gibco, New Zealand, penicillin (100 units/mL) and streptomycin (100 μg/mL).

Rat Liver Microsomal Enzymes and Cofactors: S9 was obtained as Arclor 1254-induced Sprague Dawley male liver lyophilised S-9 (MolTox LS-9) lot number 0406, from Molecular Toxicology Inc Maryland, USA. MolTox LS-9 vials were stored -70° and were reconstituted to 2.1 mL immediately prior to each test.

S9 mix was prepared freshly for each test and consisted of various components in the following proportions: Component Volume

S9 Fraction 0.5 mL (5%)

MgC-₂ (0.4 M) - KC 1 (1.65 M) salts 0.4 mL

1 M glucose-6-phosphate 0.1 mL 0. 1 M nicotinamide adenine dinucleotide phosphate 0.8 mL

0.2 M, pH 7.4 phosphate buffer 5.0 mL

Sterile distilled water 3.2 mL

The mixture was stored on ice for the duration of the assay, and any remaining at the end of the day was discarded.

Micronucleus Assay: For each experiment V79 cells were dispersed at 1 x 10⁵ cells/mL in 100 mm tissue culture dishes containing 15 mLs of α-MEM and 10% foetal calf serum and incubated for 24 hours in a C0₂ incubator at 37°C. The medium was removed and cells were washed with phosphate buffered saline then 10 mLs of α-MEM, containing an appropriate dose of test substances, 1 mL of "S9" mix and no foetal calf serum, was added to each dish. The cells were then incubated or 4 hours in the presence of the test substances. After 4 hours the medium containing the test compounds was removed, the cells were washed with phosphate buffered saline, trypsinised and resuspended in fresh growth medium. Cell densities were determined using an electronic Particle counter (Coulter Electronics).

To assess micronucleus induction, cells were plated in a 60 mm tissue culture dish containing a sterile coverslip at a concentration of 1 x 10⁵ cells/mL in 5 mLs of α-MEM containing 10% foetal calf serum. Cytochalasin B was added to each dish to give a final concentration of 3 μg/mL and the cells were incubated for 20 hours then stained as described by T R Chen¹². Then all but 1 L of medium was removed, and 1 mL of Carnoy's fixative (redistilled methanol: analytical grade acetic acid, 3: 1) was added at room temperature. After 2 minutes the fixative was aspirated and the coverslip allowed to air dry in the petri dish. The coverslip was then stained by flooding with a sufficient volume of 80 μM Hoechst-33258 stain. After 8 to 10 minutes, protected from light, the coverslip was washed in deionised water, mounted on a slide and examined under a Nikon episcopic fluorescence microscope. Binucleate cells were scored for micronuclei identified using the criteria of Heddle et al¹³. At least 1000 cells were scored from each experiment.

Initially a dose range of the sample was tested in the cell line with the addition of exogenous S9 mix, in the absence of any known mutagen. This was in order to establish that the test sample itself did not induce micronuclei in the cell line. A dose range of PhlP was also tested alone to establish a suitable dose to use and investigate any protective properties the evening primrose extract might have. A dose range of the evening primrose extract was then tested with PhlP at 0, 2, 3 and 5 μg/mL. A solvent control was included in each experiment. The results are shown in Table 2.

The number of micronuclei and multiple micronulei counted in binucleate cells are summarised in Table 2 which shows that the evening primrose extract is associated with a significant and dose related decrease in the number of cells containing micronuclei induced by 3 μg/ l and 5 μg/ml PhlP (p>0.01).

Table 2

EXAMPLE 3 Gelatin capsules

Gelatin capsules are prepared by incorporation of lOOmg of the freeze dried powder from the 60% aqueous acetone extract of Example 1 into each capsule. References

1. Hickman, KCD Proceedings of Josiah Macy Conference on Physiological Antioxidants, New York, 1949.

2. Evans, R J and Reynhout G INFORM, ( 1994) 5, 467.

3. Niwa Y, Hanssen M "Protection for Life's How to boost your Body's

Defences Against Free Radicals and the Aging Effects of Pollution and Modern Lifestyles" Thorsons Publishers, Wellingsborough ( 1989) p9.

4. Salonen J.T et al. British Medical Journal ( 1985) 290, 417.

5. Nakamura, T; Mukaiyama, T; and Nagayama K ( 1991) J Fac Agric

Kyushu Uni 36, 93.

Miller, H E ( 1971) J Amer Oil Chem Soc 48, 91.

7. Czochanska, Z; Foo, L Y; Newman, R H; and Porter, L J ( 1980) J

Chem Soc, Perhάn Trans, 2278.

8. Sun, D; Zaho, Z; Wong, H; and Foo, L Y ( 1988) Phytochemistry, 27, 579.

9. Foo, L Y; and Porter L J ( 1980) Phytochemistry, 19, 1747.

10. Nishizawa, M., Hamagishi, T., Nonaka, G.I. and Nishioka, I., (1982) J. Chem. Soc. Perkin Trans I, 2963.

11. Mallet, J F; Cerrati, C; Ucciani, E; Gamisans, J; and Gruber, M ( 1993) Food Chemistry, 49, 61.

12. T R Chen (1977) Experimental Cell Research, 104, 255-262. 13. J A Heddle; A S Raj; and A B Krepinsky (1981). H F Stich and R H

C San (eds). Short-term tests for chemical carcinogens, Springer Verlaf, New York, pp 250-254.

Aspects of the invention have been described by way of example only and it should be appreciated that modifications and additions thereto may be made without departing from the scope of the invention.

Claims

CLAIMS:

1. A food-, cosmetic- or pharmaceutically-acceptable extract of an Oenothera species comprising antioxidant.

2. An extract as claimed in claim 1 comprising 1, 2, 3, 4, 6 - pentagalloyl glucose.

3. An extract as claimed in claim 1 comprising at least one of catechin and procyanidin gallate.

4. An extract as claimed in claim 1 comprising 1, 2, 3, 4, 6 - pentagalloyl glucose, catechin and procyanidin gallate.

5. An extract as claimed in claim 1 wherein the species is Oenothera biennis.

6. An extract as claimed in any one of claims 1-5 wherein the extract is a seed extract.

7. An extract as claimed in any one of claims 1-6 wherein the extract is a polar solvent extract.

8. An extract as claimed in claim 7 wherein the polar solvent comprises at least one of acetone, ethanol, methanol and water.

9. An extract as claimed in claim 8 wherein the polar solvent is ethanol, aqueous acetone or aqueous methanol.

10. An extract as claimed in claim 6 wherein the seeds after oil removal are treated with 10-100% acetone, 10- 100% ethanol or 10- 100% methanol.

11. An extract as claimed in claim 10 wherein the seeds after oil removal are treated with 30-70% acetone, 40-90% ethanol or 40-90% methanol.

12. A method of preparing a food- or pharmaceutically-acceptable extract of an

Oenothera species comprising antioxidant having the step of extracting seeds of an Oenothera species with a polar solvent.

13. A method of preparing 1, 2, 3, 4, 6 -pentagalloyl glucose comprising the step extracting seeds of an Oenothera species with a polar solvent.

14. A method as claimed in claim 13 wherein the polar solvent extraction is preceded by an oil removal step.

15. A method of claim 13 or 14 comprising the further step of purification using chromatography.

16. A method of claim 15 wherein said chromatography is adsorption chromatography.

17. An extract obtainable by the method of any one claims 12- 16.

18. An ingestible composition comprising an antioxidant- containing extract of any one of claims 1- 1 1 and 17.

19. An ingestible composition as claimed in claim 18 in the form of a unit dosage form containing 0. lmg- lOOOmg of extract.

20. An ingestible composition as claimed in claim 19 wherein the dosage form contains lmg-250mg of extract.

21. An ingestible composition of claim 19 or 20 in the form of a capsule or tablet.

22. A topical composition comprising an antioxidant-containing extract of any one of claims 1- 11 and 17.

23. An extract of any one of claims 6-9 and 17 wherein the extract has been prepared from seeds from which oil has previously been removed.