The commercial development of food plants used as medicines
Horrobin, D.F. and Lapinskas, P. (1998)
In: H.D.V. Prendergast, N.L. Etkin, D.R. Harris and P.J. Houghton (Editors), Plants for Food and Medicine, pp. 75 - 81. Royal Botanic Gardens, Kew.    Buy the book!


The cost of development and the issue of patent protection
Some problems of pharmaceutical crop development
Gamma-linolenic acid as a drug
The evening primrose as a GLA source


For most of the history of humanity, the majority of medicines have been derived from natural sources. Plants have been the main sources, but micro-organisms have also been important. In the 1950s, however, the situation changed with respect to plants. The success of some wholly synthetic drugs led scientists and managers in the pharmaceutical industry to the idea that to use natural products as drug sources was passé. The only exception was the antibiotic field where it continued to be recognised that micro-organisms might be better than human chemists at developing antibacterial strategies. Most antibiotics are therefore still derived from microorganisms, although the molecules may be subjected to chemical modification after extraction from the fermentation broth.

Plants, however, effectively disappeared as sources of new drug substances and for 40 years there was almost no interest in using plants for drug development. Very slowly that situation is changing, and some companies are beginning seriously to look again at the plant world. There are several reasons for this.

  1. The failure of drug discovery programmes
    Drug discovery has become less and less efficient. No one is quite sure about the prime causes of this trend, but there is no doubt about the fact. If one takes the top 20 drug companies in the world, adds up their research and development (R&D) expenditure over the past 15 years, and divides that total by the number of genuinely new compounds they have discovered and brought to market, the results are astonishing. Depending on the company, each new compound is costing between $750 million and $2000 million to develop. Of course, that way of assessing costs includes the failures and emphasises the scale of the overall inadequacy of pharmaceutical R&D.

  2. The difficulty of discovering new drugs without botanical leads
    One consequence of the failure is the recognition that, with respect to the development of completely novel chemical structures, plants may still have the edge over humans. It may be easier to find entirely novel structures by looking at plants than to discover them de novo.

  3. Lower risk of toxicity in plants
    One reason for the failure of pharmaceutical industry R&D is that many compounds demonstrate unexpected toxicity. It would be absurd to say that all plant chemicals are safe, since some are clearly highly toxic. But it is true that, if a plant has been widely used by humans, then substances within that plant are less likely to be seriously toxic than is a novel synthetic entity.

  4. Plant medicines represent generations of trial and error
    Another reason for failure is that with many diseases we simply do not have good chemical ideas about where we should start. A new sense of humility has led to the view that traditional remedies from ancient systems of medical care may have more to offer than has been previously thought.

  5. Interest and initiatives in the public sector
    On this positive side, the dramatically increased concern about "green" issues has led the public to be more interested in using medicines from natural sources and in the idea of using the rain forest, for example, as a source of new drugs.

For all these reasons, for the first time in 40 years, a more positive climate exists with respect to the development of pharmaceuticals from plants. In addition to prescription drugs there is also now great interest in "nutraceuticals", natural products and foods which, in addition to providing nutrition, foster specific therapeutic effects that can be claimed on the label. This discussion outlines some of the opportunities and problems in the field.

Up Top of page

The cost of development and the issue of patent protection

In all countries, the real costs of drug development have escalated rapidly, even if one ignores the failures and inefficiencies of the discovery and development process. These costs have resulted from the imposition of "Good Laboratory Practice", "Good Manufacturing Practice", and "Good Clinical Practice", and from the steadily increasing requirements associated with the demonstration of the safety and efficacy of a compound. These standard requirements in the EU and the USA are now so onerous that, even if everything goes perfectly from day one, a new drug could not be developed with expenditure of less than about $100 million.

What this means is that no sensible company could seriously contemplate entering a drug development programme unless it could see that the returns, after all expenses, would reach at least $300 million. Because of the cost, the long delays in the development and regulatory processes, and the inevitably long pay back period, most companies will require much larger potential profits than that. They also will want to be reasonably sure that they will have some protection of intellectual property; without that no investment could be made.

  1. Intellectual Property The intellectual property issue used to be a serious barrier to the development of drugs from plants. Except in the USA, it used to be impossible to obtain patent protection for new uses for known chemicals. If a chemical found in a plant was known for any reason at all, its use as a pharmaceutical could not be protected. This situation has changed, and most developed countries have now come into line with US practice and allow the patenting of new pharmaceutical uses of known compounds. This means that a company can make an investment in a plant-source drug even if the compound is already known.

    Further forms of intellectual property protection are being developed. Governments have recognised that often there may be new drugs that cannot be patent-protected. These may, for example, arise from traditional sources such as Ayurvedic medicine, and thus be well-known, or alternatively be developed by people in academic institutions who have no knowledge of patent law and who consequently publish without realising that the act may completely destroy the possibility of patenting. Governments have therefore developed various forms of marketing protection that give companies who make the necessary investments in unpatented compounds to take them through the approval process, and to have the opportunity for a market monopoly for anything between three and ten years. These marketing protection provisions, it has to be said, are not always popular even in Government departments other than those which are concerned with intellectual property. Scotia Pharmaceuticals, for example, believed that it had marketing protection in the EU for the drug use of evening primrose oil (from Oenothera spp.; Onagraceae) for atopic eczema. But the Departments of Health in several European Countries did not like the EU marketing laws and granted licenses to generic competitors who had spent nothing on drug development. Scotia Pharmaceuticals had to take its case to the European Court to win recognition that European marketing laws really do have force.

  2. Nutraceuticals. The problems facing nutraceuticals are very far from any equivalent resolution. At present, in most Western countries, most health claims for a product are illegal unless it has gone through a full regulatory approval process with all its costs. If these rules are reduced to allow food substances to be associated with health claims when there is reasonable evidence, some form of intellectual property provision also must be put in place if a responsible industry is to develop. Even with greatly relaxed rules, it is doubtful whether reasonable evidence for a nutritional product's safety and efficacy for a particular disease could be bought for less than $20 million or so. No one is going to invest such a sum unless they can be confident that they will not immediately be copied by competitors who do no research and incur no similar costs. Some form of claims approval system with marketing protection for a reasonable number of years will therefore have to be put in place if the nutraceutical industry is going to develop in a way that makes a real contribution to society.

Up Top of page

Some problems of pharmaceutical crop development

There are several issues that must be addressed if a plant source is to be developed successfully as a pharmaceutical.

  1. Prediction of future safety and efficacy

    A disease must be identified that responds to a chemical substance found in a plant. This in itself will require considerable expenditure since the compound must be present in reasonable quantities, it must be found to have acceptable safety in animals and humans, and it must be shown to be clinically effective in pilot studies of animal models or in humans or both. In essence, the substance must be shown to be safe and to have a reasonable prospect of being effective. Judgement on these matters must be made by industry scientists at times when outside observers and government regulators would certainly regard the evidence as inadequate. The skill of the industrial scientist involves the accurate prediction of future safety and efficacy on the basis of inadequate evidence.

  2. Financial assessment If a positive judgement is made about likely future safety and efficacy, in most companies a financial assessment also must be made. Are the returns from sale of the product for the disease concerned likely to generate a certain minimum sum after all expenses have been met? Criteria will vary from company to company, but this sum is unlikely to be less than about $300 million.

  3. Suitability of synthetic preparation

    Can the chemical concerned be prepared at reasonable cost via a synthetic route? There is no doubt that most pharmaceutical companies are more comfortable with chemical syntheses than with natural sourcing, and so if they can make the compound synthetically they will. Most organisations will prefer to use the plant source as the idea for the structure of the drug and then move as far away as possible from that source. The only exceptions are situations where the cost of synthesis is considerably higher than the cost of producing the compound in plants. Compounds that lead to the favouring of a plant source are exceptionally complex and their synthesis involves a large number of separate steps and/or ones in which there are multiple isomers, only one of which is biologically active, and in which the plant produces the desired isomer in pure form.

  4. Abundance of the compound in the plant

    Does the plant contain a reasonable amount of the compound, preferably in easily harvested parts such as leaves or seeds? A plant is likely to be a practical source only if it produces a reasonable quantity of material in relation to likely demand, and only if that material can be extracted and purified with relative ease. Some potentially useful compounds, for example, might be found in the bark or roots of slow-growing trees. From an agronomic point of view such sources are unlikely to be viable.

  5. Potential for genetic manipulation

    If the drug is a success, is there a reasonable possibility of modifying the plant either by conventional breeding or by newer forms of genetic manipulation, so that production can be rapidly expanded in a reliable way? Only if the prospect for this looks reasonable is there a likelihood that a company will make a decision to invest.

These are only some of the most important problems that must be addressed if a plant source is going to be useful.

Up Top of page

Gamma-linolenic acid as a drug

In the late 1970s and early 1980s gamma-linolenic acid (GLA) was identified by researchers at Scotia Pharmaceuticals as a compound likely to be of benefit in the management of three important disorders: atopic eczema, diabetic nerve damage, and some forms of cancer. Since in the developed world there are perhaps 10 million people with eczema, 30 million with diabetes, and 10 million with cancer, and since all three conditions are managed with only partial success, it seemed as though the market size was likely to be adequate if GLA was clinically effective in even one of these conditions.

It also seemed likely that GLA would be safe. GLA is a normal intermediate in human metabolism. There is little if any GLA in the diet, except in human milk, where it occurs in substantial amounts. GLA in humans is made from dietary linoleic acid by delta-6-desaturation. GLA is then metabolised to a range of substances that are absolutely required for normal membrane structure and normal cell signalling mechanisms. GLA therefore plays a central role in the economy of most organs.

Researchers from Scotia Pharmaceuticals and from academia demonstrated that in atopic eczema there is reduced conversion of linoleic acid to GLA and that this plays an important role in the development of the skin lesions. In diabetes, the reduced formation of GLA, combined with abnormal glucose metabolism, leads to serious nerve damage. Finally, some cancer cells completely fail to produce GLA; as a consequence the cells multiply rapidly, a process that can be completely stopped by the direct administration of GLA.

Although at the time the evidence was only limited, nevertheless the Scotia Pharmaceuticals scientists judged it sufficient to develop a major programme for the production of GLA on a large scale. That decision has been justified because GLA has been shown to be clinically effective in atopic eczema (Morse et al., 1989), diabetic nerve damage (GAMTG, 1993), and pancreatic cancer (Fearon et al., 1996). GLA has been approved for the treatment of eczema in 14 countries including Australia, Denmark, Germany, Kenya and the UK. Applications for approval for its use in diabetic neuropathy and in pancreatic cancer have also been filed in a number of countries. There can therefore be no doubt that a market of adequate size is there.

GLA also meets many of the criteria required for successful development of a plant source as a manufacturing route. It is found in the seeds of a number of plants, notably evening primroses Oenothera spp. subsect. Euoenothera, borage Borago officinalis (Boraginaceae), and members of the Grossulariaceae. It has three carbon-carbon double bonds, each of which can be in either the cis or trans configuration, giving eight possible isomers. Only the all- cis isomer is biologically active, the other isomers acting as competitive inhibitors. The plant produces 100% of the all- cis compound, whereas all synthetic routes produce some mixture with inactive and competitive inhibitory trans isomers. The synthetic routes and the subsequent separations can be performed but are extremely expensive, so making plant sources reasonably competitive. The plant sources were judged to be capable of improvement and so a decision was made to go for them as the main manufacturing route.

Up Top of page

The evening primrose as a GLA source

The Oenothera spp. used for GLA production originated in North America, but have spread around the world (see Lapinskas, 1993, for more details). They are primary colonisers, growing well on waste ground but failing to compete effectively in richer environments where many other species are present. Most of them are biennial, with the seeds normally germinating in the spring and the plants flowering in the following year. Such plants grow initially as a flat rosette and overwinter in this form. In the next spring the plants bolt, producing a flowering head with yellow flowers that open in the evening and die the next day. Flowering starts at the bottom of the spike and spreads steadily upwards so that flowering and the setting of seed may be spread over a number of weeks, usually completing in September or October. The seeds are found in capsules that split open as they ripen, scattering the seed over the ground. At any one time a spike may contain flowers at the top, green and ripening closed seed pods in the middle, and ripe open pods that are losing their seed at the bottom. The oil comprises 20-25% by weight of the mature seed and in the wild contains about 6-9% of GLA. Unusually, and offering a major advantage over other seed sources, the composition of the oil is remarkably simple with 70-75% consisting of linoleic acid and the remainder almost entirely of oleic and palmitic acids. It is therefore relatively easy to prepare pure GLA from evening primrose oil.

However, it is immediately apparent from the description that the wild plant has characteristics that make it less than desirable as a crop. The main ones are:

Since the late 1970s Scotia Pharmaceuticals has had a group of plant-breeding and agronomy specialists working to improve the crop. Well over 3000 wild varieties have been collected from around the world, examined for all relevant characters and, when appropriate, entered into breeding and agronomy programmes. As a result the reliable yields of GLA/hectare have increased 20- to 30-fold, mainly because of improved seed oil content, improved oil GLA content and the introduction of non-seed-shedding varieties that retain all or most of the seeds until all the seeds on the spike are ripe.

In addition the crop has been made much more farmer friendly by the:

As a result the evening primrose has become established as a reliable source of GLA for both nutraceutical and pharmaceutical uses. It faces competition from borage as a source of pure GLA for pharmaceutical purposes. However, borage oil, either natural or partially purified, is less attractive than primrose oil for nutraceutical purposes since it contains some unidentified potentially toxic material that stimulates platelet aggregation and may therefore increase the risk of thrombosis. Primrose oil carries no such risks. In addition, the image of the evening primrose has become enormously attractive to consumers; and this together with the improved varieties is likely to ensure a long-term commercial future for the evening primrose.

For pharmaceutical purposes, where origins are unimportant and where economical production of the pure compound is what matters, algal and fungal fermentation and genetic manipulation of cheap oilseed crops so that they will make GLA are all threats to the evening primrose. Scotia Pharmaceuticals is involved in all these areas because we cannot afford to miss out on a development that would deliver large volumes of inexpensive, high quality GLA. However, somewhat to our surprise, the evening primrose developments have more than kept pace with the others, although how long that will continue must be a matter for speculation.

Up Top of page


The climate for plant-derived pharmaceuticals and nutraceuticals has recently improved considerably. Particularly with nutraceuticals, the attractiveness of the plant source to many people may offer important commercial advantages. However, very substantial investments will be required; and the field urgently requires an appropriate framework for intellectual property protection.

Up Top of page


Fearon, K.C.H., Falconer, J.A., Ross, J.A., Carter, D.G., Hunter, JO., Reynolds, P.D. and Tuffnell, Q. (1996).
An open-label dose escalation study of the treatment of pancreatic cancer using lithium gammalinolenate.
AntiCancer Research 16:867-874.

Gamma-linolenic Acid Multicenter Trial Group (GAMTG) (1993).
Treatment of diabetic neuropathy with gamma-linolenic acid.
Diabetes Care 16:8-15.

Lapinskas, P. (1993).
Oil crops for the pharmaceutical industry.
In: P.R. Shewry and K. Stobart (Editors), Seed storage compounds, biosynthesis, interactions and manipulation, pp. 332-342. Proceedings of the Phytochemical Society of Europe. Clarendon Press, Oxford.
(Full text)

Morse, P.F., Horrobin, D.F., Manku, M.S., Stewart, J.C.M., Allen, R., Littlewood, S., Wright, S., Burton, J., Gould, D.J., Holt, P.J., Jansen, CT., Mattila, L., Meigel, W., Dettke, T., Wexler, D., Guenther, L., Bordoni, A., and Patrizi, A. (1989).
Meta-analysis of placebo-controlled studies of the efficacy of Epogam in the treatment of atopic eczema. Relationship between plasma essential fatty acid charges and clinical response.
British Journal of Dermatology 121:75-90.

Up Top of page

© Peter Lapinskas 1999-2012 Email Peter Lapinskas Last updated: 3 July 2012

Home      Services      Background      Publications      Resources      Contact