Oil crops for the pharmaceutical industry
Lapinskas P. (1993)
In: Shewry P. and Stobart K. (eds.) Seed storage compounds: biosynthesis, interactions and manipulation, pp 332 - 342. Oxford University Press.   ISBN 0 19 857768 0.

Chapter 23

Biochemical background
Therapeutic effects
Choice of source
Development of evening primrose as a crop
Future developments


Most oils (defined as long-chain fatty acids, normally in triglyceride form) derived from plants are used in pharmaceutical products primarily for their physical properties (Reynolds 1989). For instance, olive oil is used as a demulcent (to protect and soothe inflamed mouth tissue), as a mild laxative, as a component in emollient ointments and liniments, and as a lubricant; castor oil is a well-known ingredient of emollient creams, sesame oil is used as a solvent for steroids; and peanut oil is used as a mild enema and for softening ear wax. In addition, sunflower oil and soybean oil are used to provide energy and essential fatty acids for parenteral (e.g. intravenous) nutrition and some oils, such as castor oil, are used for making soaps and detergents which have medical uses. There is also a product (Lipiodol) based on a mixture of iodine with the ethyl esters of poppy seed oil which is used as a contrast medium for the visualization of the lymphatic system and nasal and other sinuses in radiography.

Prior to 1988, this represented the full extent of the use of oils in medicine. However, in that year, a new product (Epogam - gamma-linolenic acid in evening primrose oil) was granted a product licence in the UK for the treatment of atopic eczema, and this has opened up a whole new approach to the treatment of a range of diseases. The remainder of this paper will describe the rationale behind this form of therapy, its development and prospects for the future. The medical and biochemical aspects are based on recent reviews by Horrobin (1990a, b, 1992a, b), with permission.

Up Top of page

Biochemical background

Fat in the diet is primarily used as a source of energy in mammals. However, animals kept on a fat-free diet develop a series of symptoms which progressively deteriorate, leading ultimately to death, and this deterioration can be reversed by feeding certain fatty acids. These fatty acids are clearly required nutrients and are therefore called essential fatty acids. Like vitamins and essential amino acids, a supply of essential fatty acids in the diet is needed to maintain health.

Basic metabolism of n-6 essential fatty acids
Fig. 23.1 Basic metabolism of n-6 essential fatty acids.

The essential fatty acids are carboxylic acids with a long carbon chain containing one or more methylene-interrupted double bonds. They can be classified into two families, the n-3 series, with the first double bond after the third carbon from the methyl end, and the n-6 series, with the first double bond after the sixth carbon atom. The n-3 series essential fatty acids have some interesting metabolic effects, and interact with the metabolism of the n-6 series essential fatty acids, but they do not appear to have the same pharmaceutical potential, and so will not be considered in detail.

The metabolism of the n-6 essential fatty acids in the body is shown in Fig. 23.1. The primary essential fatty acid in the diet is linoleic acid which is desaturated by the D6-desaturase enzyme to form g-linolenic acid. Two carbon atoms are added to form dihomo-g-linolenic acid and then there is a further desaturation by the D5-desaturase enzyme to produce arachidonic acid. The desaturation is the rate-limiting step in the pathway, and the conversion from dihomo-g-linolenic acid to arachidonic acid is also slow, whilst the elongation reaction is rapid.

All these fatty acids are able to reduce the effects of essential fatty acid deficiency, although only linoleic acid, and to a lesser extent arachidonic acid, are present in significant proportions in the normal diet.

The activity of the essential fatty acids is in four main areas. Firstly, they are important components of cell membranes - they affect the physical characteristics of fluidity and permeability, and they can modulate the activity of membrane-bound enzyme systems and receptors. Secondly, the essential fatty acids are involved in the regulation of cholesterol synthesis and transport. Thirdly, they are important in preventing water loss from the skin. Lastly, they are substrates for the production of a wide range of short-lived regulatory molecules known as eicosanoids which include the prostaglandins and leukotrienes and which are involved in the moment-by-moment regulation of cellular function.

The effect of a blockade of
Fig. 23.2 The effect of a blockade of D6-desaturase on n-6 metabolism. PGE1, prostaglandin E1; 15-OH-DGLA,
15-hydroxy-dihomo-g-linolenic acid; PG2, 2-series prostaglandins; TX2, 2-series thromboxanes; LT4, 4-series leukotrienes;
12-OH-AA, 12-hydroxy-arachidonic acid.

Since there are many different eicosanoids, and they often have opposite effects, there must clearly be a sophisticated control mechanism to regulate their production and relative concentrations and this will be dependent on adequate supplies of the relevant essential fatty acids.

In certain circumstances, however, there may be inadequate amounts of g-linolenic acid, dihomo-g-linolenic acid and arachidonic acid, and their metabolites, even though the dietary intake of linoleic acid is adequate. This may be because of factors which interfere with the action of the D6-desaturase enzyme, such as: ageing, diabetes, high alcohol intake, nutritional deficiencies, stress, high cholesterol levels, viral infections, or eczema. Alternatively, it may be because of factors which accelerate the consumption of the products of the D6-desaturase such as excessive oxidation or high rates of cell division which are found in inflammation and cancer cell growth.

An example is illustrated in Fig. 23.2. Arachidonic acid is the substrate for the production of the 2-series prostaglandins, 2-series thromboxanes, 4-series leukotrienes, and 12-hydroxy arachidonic acid, most of which are pro-inflammatory. Their production is inhibited by prostaglandin E1 and 15-hydroxy dihomo-g-linolenic acid, which are produced in turn from dihomo-g-linolenic acid and which also have intrinsic anti-inflammatory effects. If there is an inhibition of the D6-desaturase enzyme, therefore, this will result in a shortfall of production of the metabolites g-linolenic acid, dihomo-g-linolenic acid, and arachidonic acid. However, arachidonic acid is normally found in the diet and can be mobilized from phospholipids. The production of anti-inflammatory and inhibitory compounds from dihomo-g-linolenic acid will therefore be reduced and the production of pro-inflammatory metabolites from arachidonic acid will be enhanced, thereby creating the conditions for an inflammation reaction. However, if, exogenous g-linolenic acid can be supplied, for instance from evening primrose oil, then the inhibition of the D6-desaturase can be by-passed and the normal inflammation control mechanisms can reassert themselves.

Of all the eicosanoids, prostaglandin E1 in particular has a range of effects which are highly desirable in a number of clinical situations. It can dilate blood vessels, lower blood pressure, inhibit platelet aggregation, inhibit cholesterol synthesis, regulate the immune response, have an anti-inflammatory action, or stimulate cyclic AMP formation (thus inhibiting the production of inflammatory compounds from arachidonic acid). It is therefore possible to use exogenous supplies of g-linolenic acid in three situations:

Up Top of page

Therapeutic effects

There are a number of disease states where the metabolism of the n-6 pathway is disturbed and which might respond to supplementation with g-linolenic acid. A great deal of research is currently in progress to determine whether these diseases will respond and to find the optimum means of administration. Before a new therapeutic agent may be used it has to be licensed by the appropriate national authority, and these authorities require data demonstrating the safety and efficacy of the product. The current situation is summarized below.

Atopic eczema

In atopic eczema it appears that, in at least some patients, the action of the desaturase enzyme is blocked. In double-blind placebo controlled trials, Epogam, a product based on a specially selected form of evening primrose oil containing 40 mg g-linolenic acid per 500 mg capsule, was found to give significant improvements in all aspects of the disease, but particularly in controlling itch. It has now been approved for pharmaceutical use in the UK, Ireland, Denmark, Germany, Italy, Greece, South Africa, Australia, and New Zealand. Launched only in 1988, it is now one of the most widely prescribed dermatological pharmaceuticals in the UK.

Mastalgia (breast pain)

Efamast (a product similar to Epogam) was licensed in the UK, again on the basis of clinical trials which demonstrated a significant relief of symptoms. It is believed that the reason for the pain is abnormal sensitivity of breast tissue to normal levels of circulating ovarian hormones brought about by the deficiency of n-6 essential fatty acids. This may result from an abnormally high rate of use of arachidonic acid, possibly exacerbated by a reduction in D6-desaturase activity.

Diabetic neuropathy

Diabetes is characterized by a blockade of the D6-desaturase and D5-desaturase enzymes and by an inhibition of the conversion of dihomo-g-linolenic acid to prostaglandin E1. Clinical trials have indicated that the long-term deterioration of nerve function associated with diabetes can be halted or even reversed by feeding g-linolenic acid in the form of evening primrose oil. It is likely that this will become an important therapeutic use.

Inflammatory and auto-immune disorders

Initial trials in rheumatoid arthritis have indicated that Efamol evening primrose oil gives a substantial relief of symptoms, permitting patients to reduce the dosage of conventional therapy, and hence also reduce the risk of side effects.

Viral infections

The n-6 essential fatty acids appear to be required for the anti viral actions of interferon and are depleted by viral infections. They also have direct anti viral activity, particularly against viruses with lipid envelopes. Initial trials on the effect of g-linolenic acid + eicosapentaenoic acid (an n-3 essential fatty acid) have indicated significant activity in patients suffering from post viral fatigue syndrome (ME), and in people with AIDS.


In vitro, essential fatty acids of both the n-3 and n-6 series kill cancer cells whilst leaving normal cells largely unharmed. In most cell lines, g-linolenic acid and dihomo-g-linolenic acid are the most effective and produce least effect on normal cells. This has been tested successfully in vivo on human cancers transplanted into mice, and human trials are under way.

It is apparent from this work that g-linolenic acid has great potential as a pharmaceutical product in a range of diseases.

Up Top of page

Choice of source

The results discussed above have all been obtained using evening primrose oil as the source of g-linolenic acid. However, there are other oils which contain higher contents of g-linolenic acid, the most important being the seed oils of borage (Borago officinalis) and blackcurrant (Ribes nigrum), and various fungi, notably Mucor javanicus. However, when the oils are tested for their ability to stimulate prostaglandin E1 production, evening primrose oil proved to be twice as effective as fungal oil and an order of magnitude better than borage or blackcurrant at equal doses of g-linolenic acid.

The reason for this is unclear, but it is thought that the other fatty acids present in the oil and the position of the g-linolenic acid on the triglyceride molecule are both significant, and further work is being done to elucidate this. What is clear is that any source of g-linolenic acid must be tested for efficacy before any assumptions can be made regarding its biological effects. The presence of g-linolenic acid does not, in itself, confer activity.

Toxicological studies have shown that the evening primrose oil (as produced by Scotia Pharmaceuticals) is an extremely safe product. It has been subjected to a full range of toxicological tests in four species covering reproductive performance, teratogenicity, carcinogenicity, and long-term effects, and no toxic effects were found. Several thousand patients have been involved in hospital-based clinical studies with this oil, and no adverse events were found to occur significantly more often with the active group as compared to the control group. In addition, some 500 000 one-month prescriptions for Epogam and Efamast have been dispensed in the UK. The level of reported adverse events has been far lower than for most drugs, and there is no pattern to suggest that they are caused by the products.

As for efficacy, it would be unwarranted to extrapolate these results for evening primrose oil to other oils containing g-linolenic acid, and equivalent toxicological studies should be performed before safety can be assumed.

The results of this work on efficacy and toxicology suggest that the preferred natural oil for use in clinical applications is evening primrose oil.

Up Top of page

Development of evening primrose as a crop

The evening primroses used for g-linolenic acid production are a closely related group of species in the subsection Euoenotherae of the genus Oenothera. There is some dispute amongst taxonomists regarding the number of species within this subsection and the boundaries between them (Raven et al. 1979, Rostanski 1985), and this problem is exacerbated by the fact that they are to a great extent interfertile. This permits the easy production of interspecific hybrids, which further blur the species boundaries. For commercial purposes, the species Oe. biennis, Oe. lamarckiana, and Oe. parviflora have proved to be the most useful.

The evening primrose derives its name from the slightly phosphorescent yellow colour of its flowers, which open in the evening. It is completely distinct from the common primroses (Primula spp.) and is more closely related to the willow herbs (Epilobium spp.). The evening primrose is a plant of temperate latitudes, having originated in north America, and is now widely distributed in the wild in these regions in both northern and southern hemispheres.

As a crop, evening primrose is normally grown as a biennial. It is sown in August and forms a rosette of leaves flat to the ground during autumn, and it overwinters in this form. The following spring it throws up a main stem with a variable number of branches which will reach 1.5-2.0 m in height. Flowering normally begins in late June, with fresh flowers opening each evening (each flower lasting only one day), and can continue into September. The seed is very fine, with a 1000-seed weight of 0.5 g, and is produced in pods which ripen over a period from September to November.

Although the first licensed pharmaceutical product based on evening primrose oil was not launched until 1988, it had been apparent for many years that substantial quantities of oil would be required, not only for medical use, but also for the dietary supplement market. A programme was established in 1975 aimed at improving the yield, quality, and reliability of the seed production of a plant which was at that time new to agriculture, and this work is still continuing. All aspects of crop production have been examined, but the main emphasis has been on plant breeding, to overcome the inherent defects within the wild plants, and agronomy, to optimize the techniques of crop and seed production. Breeding in particular has been a challenge because of the peculiar and unique nature of the evening primrose genome.

Most unusually for a wild plant, the genetics of evening primrose species have been the subject of scientific study for more than 100 years and a great deal is known about their structure and organization. The earliest work was performed by de Vries, starting in 1860, and he published his results in 1900 when he became aware of the work of Mendel. He found that Oenothera hybrids do not obey normal Mendelian rules of inheritance, and used these results to put forward a theory of mutation and the origin of new species. This stimulated a surge of experimental work which in various forms has continued to this day, covering the fields of cytology, speciation, evolution, and plastome/genome interactions. The early work was extensively reviewed by Cleland (1972), and more recent developments have been described by Stubbe (1989), Schuster et al. (1987), and Schuster and Brennicke (1990).

During meiosis the chromosomes in certain races of Oenothera do not form into pairs as in normal plants, but join up end-to-end to form a circle. The reason for this is that small segments at the end of non-pairing chromosomes have been exchanged, so that pairing of homologous regions occurs only at the ends of chromosomes. This means that translocation is restricted to these regions and hence the reassortment of genes through translocation has been effectively eliminated. Furthermore, when the chromosomes separate to form the gametes, adjacent chromosomes move to opposite poles. Since these chromosomes correspond to those received either from the maternal or paternal parent, it follows that the gametes produced contain a complete and unchanged set of genes from either the maternal or the paternal parent. Finally, there is a system of lethal genes which are either gametic, where the maternal chromosomes are inactivated in pollen cells and vice versa, or zygotic, where homozygous embryos become inviable through the action of deleterious recessive alleles.

The effect of gametic and zygotic lethal genes on segregation in Oenothera
Fig 23.3 The effect of gametic and zygotic lethal genes on segregation in Oenothera
Crosses are shown (female x male), and the segregation in the F1 is shown on the right.
Male and female signs gametic lethals; Z, zygotic lethals; ABCD, parental genomes (not individual genes).

As a result, these Oenothera races are able to breed true regardless of the degree of heterozygosity. This has interesting consequences for the plant breeder. If hybrids are made between two such races, different results will be obtained depending on the lethal system present, as shown in Fig. 23.3. The important points to note are that splitting occurs in the F1 generation, that only a small number of different progeny classes are produced, and that they will then breed true in future generations. This is in contrast to normal plant species where many thousands of different gene combinations can be produced from a single hybrid through gene reassortment over five or more generations.

However, it is possible for the plant breeder to make progress. Not all races produce a full ring of 14 chromosomes; some produce a ring of 12 and 1 pair, a ring of 10 and 2 pairs, a ring of 10 and a ring of 4, etc. Indeed, all 15 possible configurations have been observed. Furthermore, by making selected hybrids, it is possible to break up the rings to introduce pairing. Those chromosomes which form pairs then conform to normal inheritance patterns and segregation can be obtained, although still subject to the limitations of deleterious recessive alleles. Substantial progress has therefore been possible in the major areas of difficulty in evening primrose production.

As a result of these and other developments, it is now possible to grow evening primrose on a commercial scale at realistic cost and without the risks previously associated with the crop.

Up Top of page

Future developments

One of the problems of using evening primrose oil as the means of administering g-linolenic acid is that it is a relatively dilute source. Each 500 mg capsule of Epogam or Efamast contains 40 mg of g-linolenic acid and so, for a clinically useful dose, 8-12 capsules have to be taken per day. This raises problems of patient compliance, in particular with patients who are either very young or who have difficulty in swallowing. Furthermore, for therapy aimed at boosting the level of desirable metabolites (such as prostaglandin E1), substantially larger doses of g-linolenic acid are required which would raise the number of capsules to be taken each day to impractical levels.

Work is therefore in progress to develop products which contain higher levels of g-linolenic acid. In view of the different efficacies of the natural oils containing g-linolenic acid, a range of different materials is being tested, including free fatty acids, ethyl esters, and various triglycerides. Clearly, proof of biological efficacy and safety is required before any form of g-linolenic acid can be assumed to be effective; there is already an indication that methyl and ethyl esters are less readily absorbed than triglycerides or free fatty acids. The levels of concentration for potential commercial products which have been achieved are as high as 99 per cent g-linolenic acid in the free fatty acid form, and one company has offered to make a range of concentrated products available free of charge to bona fide researchers in order to stimulate research in this area (Scotia Pharmaceuticals).

The use of a concentration process, particularly where g-linolenic acid is separated as a free fatty acid, makes it possible to consider a wider range of g-linolenic acid sources for use as raw materials. The actual source used will depend on the relative ease with which it may be processed, but will ultimately be determined by cost. At this stage, an accurate comparison of costs is difficult, since fungal oil has yet to be produced on a sufficiently large scale for accurate costing, and the status of blackcurrant seed as a by-product of juice manufacture means that supply is inelastic and will be particularly susceptible to demand levels. Raw material costs would indicate a cost per kilo of g-linolenic acid of about £70 for evening primrose and about £35 for borage, and Ratledge (1987) has estimated a cost of £12.50 for fermentation using Mucor javanicus. Since these figures do not include the cost of oil extraction, which is likely to be more expensive for fungal oil, it is likely that the eventual source of choice will be either fermentation or borage oil (or a combination of both). Evening primrose oil is likely to be too expensive unless use can be made of its particular characteristics, or the yields of g-linolenic acid, oil, and seed can be substantially improved. However, evening primrose oil is likely to have a continuing role in the nutritional market where a premium is attached to the use of 'natural' products.

Up Top of page


The use of oils containing g-linolenic acid is an exciting new development in the pharmaceutical industry, which is likely to assume greater importance as more research into the biochemical and therapeutic aspects of essential fatty acids is completed. However, the degree to which this new demand will be supplied by plant seed oils will depend on their cost relative to fungal sources.

Up Top of page


Cleland, R.E. (1972).
Oenothera: cytogenetics and evolution.
Academic Press, London.

Horrobin, D.F. (1990a).
Gamma linolenic acid.
Reviews in Contemporary Pharmacotherapy 1, 1-41.

Horrobin, D.F. (ed.) (1990b).
Omega-6 essential fatty acids: pathophysiology and roles in clinical medicine.
Alan R. Liss, New York.

Horrobin, D.F. (ed.) (1992a).
Treatment of diabetic neuropathy: a new approach.
Churchill Livingstone, Edinburgh.

Horrobin, D.F.(1992b).
Nutritional and medical importance of gamma-linolenic acid.
Progress in Lipid Research 31, 163-94.

Ratledge, C. (1987).
Lipid biotechnology: a wonderland for the microbial physiologist.
Journal of the American Oil Chemists' Society 64, 1647-56.

Raven, PH., Dietrich, W., and Stubbe, W. (1979).
An outline of the systematics of Oenothera subsect, Euoenothera (Onagraceae).
Systematic Botany 4, 242-52.

Reynolds, J.E.F. (ed.) (1989).
Martindale: the extra pharmacopoeia (29th edn.).
The Pharmaceutical Press, London.

Rostanski, K. (1985).
Zur gliederung der subsektion Oenothera (sektion Oenothera, Oenothera L., Onagraceae).
Feddes Repertorium 96, 3-14.

Schuster, W., and Brennicke, A. (1990).
RNA editing of ATPase subunit 9 transcripts in Oenothera mitochondria.
Federation of European Biochemical Societies Letters 268, 252-6.

Schuster, W., Hiesel, R., Wissinger, B., Schobel, W., and Brennicke, A. (1987).
Structure and transcription of the Oenothera mitochondrial genome.
In: Plant molecular biology. Vol. 140 (ed. D.V, Wettstein and N.H. Chua), pp. 115-26. Plenum, New York.

Scotia Pharmaceuticals.
The SCWT programme.
Leaflet obtainable from Scotia Pharmaceuticals Ltd., Woodbridge Meadows, Guildford, Surrey GU1 1BA, UK.
(Now unavailable - April 2000)

Stubbe, W. (1989).
Oenothera - an ideal system for studying the interactions of genome and plastome.
Plant Molecular Biology Reporter 7, 245-57.

Up Top of page

© Peter Lapinskas 1999-2012 Email Peter Lapinskas Last updated: 24 September 2012

Home      Services      Background      Publications      Resources      Contact