HIGH FLYERS THINK TANK

Biotechnology and the future of Australian agriculture

The Shine Dome, Canberra, 26 July 2005

Biotechnology: New plant products (pharming)
by Dr Allan Green, Senior Principal Research Scientist, CSIRO Plant Industries, Canberra

I am going to cover the plant product traits rather than the input traits, which Jeff Ellis has already covered and which we have heard mentioned earlier.

(Click on images for larger versions)

The first wave of GM crops to be released into agriculture really did cover what have been called the input traits – things like herbicide tolerance, insect resistance and hybrid varieties. Now we are starting to see, as Jeff mentioned, what might be the second group of those input traits, perhaps disease resistance and stress tolerance.

But it is fairly clear that, at least in some markets – not all – the difficulties associated with market acceptance have been slated home to the lack of direct benefits to consumers of some of these products.

(Click on image for a larger version)

There is quite a lot of hope, I guess, expressed by many people that GM crop acceptance will be greater when we can demonstrate consumer benefits, particularly in the area of improved nutritional quality and functionality, perhaps through things like healthier oils, enhanced nutrients or allergen-free products.

(Click on image for a larger version)

So it is in those areas that I want to go through a quick set of examples of the types of products that are being developed in our research labs and entering commercial development. They fall into two categories. What is being called the second wave of the output traits, or the quality traits, are basically improvements of our current food crops.

The types of things we might think about there are healthier oils – for instance, low saturates, zero-trans cooking oils, allergen-free grains, vitamin-enriched grains (you would all be familiar with the Golden Rice example), and omega-3 plant oils. I will talk about each of those briefly.

(Click on image for a larger version)

Then there is what has been called the third wave, which is really the move to industrial and pharmaceutical crops, making our plants produce products they don't currently make. Some examples would be reticuline-containing poppy rather than morphine-containing, new oleochemicals for industrial use, biodegradable plastics in plants and, as an example of a pharmaceutical, human insulin produced in plants. I will cover each of those very briefly.

(Click on image for a larger version)

I think it is important to realise that there are two main technologies that we use in modifying the metabolic pathways in plants. Firstly, we can silence endogenous genes, particularly by using RNAi silencing, the newly developed technique there. In that way we can shut down pathways and change the product composition of plants. The other way is to introduce new biosynthetic pathways. We can do that in either a very simple way – maybe introduce a gene for an enzyme that has a very simple single-step function – or in some cases we can contemplate introducing completely very complex multi-step biosynthetic pathways. And there are examples of each of these.

Firstly I have here some examples of how we can change product quality by silencing endogenous genes.

(Click on image for a larger version)

One of the first cases of successful use of RNAi silencing has been work actually done in CSIRO to modify the oil composition in cotton and canola. In both of those crops it has been possible both to change the level of saturates, to produce very low saturated fatty acids, and also to switch from polyunsaturates to monounsaturates. The commercial importance of this is that polyunsaturates are converted industrially into monounsaturates by hydrogenation, a process that delivers trans fatty acids into the product, and these are cholesterol-raising. So by using a biological change here we can produce a stable, healthy cooking oil that will have very low saturates and no trans fatty acids.

(Click on images for larger versions)

Another example is morphine-free poppy. CSIRO, working with Johnson & Johnson Research through Tasmanian Alkaloids, have been interested to see what new pharmaceutical compounds we could make in poppy, other than the narcotic morphine. As you would know, poppy is a source of morphine, and from morphine we derive codeine, which is one of the main pharmaceuticals of importance from poppy.

What if we could change poppy so that instead of producing morphine it produced codeine and we didn't require that industrial processing step? Or what other opportunities are there in poppy to produce new pharmaceuticals?

One of the early successes of Phil Larkin and his team has been to silence one of the enzymes late in the pathway of morphine synthesis, codeinone reductase. By doing that, they were able to remove morphine from the poppy and build up the level of an early alkaloid, reticuline, which is a potential scaffold for developing other pharmaceuticals of importance. That was recently published in Nature Biotechnology.

(Click on image for a larger version)

As Ian Edwards, I think, mentioned earlier, allergens are a potential area where gene silencing could play a really important role in improving nutritional quality of foods. There are many allergens in the food supply which have disastrous consequences, particularly with peanut – you would have heard recently of a child dying in Australia. The proteins in these foods which are responsible for the allergenicity are being identified, and the genes that control them are being identified as well. We can potentially use gene silencing to inactivate the synthesis of those proteins and move down the path of reducing allergens in grains and other foods.

I have also given there a ricin-free castor as an example of an industrial crop which has a major toxin, which is a severe limitation in using that crop in agriculture. There are efforts to use gene silencing to remove ricin protein from castor bean.

I will switch now to the introduction of new biosynthetic pathways.

The first example I want to give is a simple single-gene insertion. This is work that we have done at CSIRO, where we have taken a gene from a wild plant, Crepis palaestina, which produces an unusual fatty acid that has quite a valuable industrial use, and we have been able to transfer that into an oilseed plant and have that oilseed produce the modified fatty acid. So a single biosynthetic function can change the composition of the oil of, in this case, linseed. Now we need to do more work to increase the yields and that may involve additional steps, but to have the product produced only requires one gene.

(Click on image for a larger version)

Another fairly simple genetic change, one with quite an interesting application, I think, relates to the potential to develop human insulin in plants. A company in Canada, Symbiosis Genetics, is fairly well advanced down this path, which relies on the fact that, in the cotyledon of oilseeds, oils accumulate in spherical bodies called oleosomes, and these bodies are surrounded by proteins. Symbiosis found that you could engineer those proteins without affecting the structure and functioning of the oilbodies, and you could actually insert into that protein a small peptide – perhaps a pharmaceutical peptide, such as insulin – and have that accumulate in the proteins around the oilbody. And then, because of the particular physical properties of oilbodies, you can extract those proteins with very high efficiency. Then you can cleave the peptide out of that protein, using an inserted proteolytic scythe.

(Click on image for a larger version)

This is a very interesting production system in plants for producing high-purity peptides of pharmaceutical interest.

(Click on image for a larger version)

That is an opportunity, probably, for a couple of farmers. It is a big-value opportunity, but it is a relatively small-value production system. So it is one that will probably be realised to the advantage of the biotech industry rather than to the advantage of agriculture.

(Click on image for a larger version)

Moving into more complex pathways, I think the one that we are all aware of, that gets so much publicity, is Golden Rice. This results from the insertion of a pathway into rice to produce β-carotene, which is the precursor of vitamin A in the human body. The initial product that was developed involved the insertion of the three genes shown here in red, basically three genes from daffodil and a soil bacterium, which encode the pathway for production of β-carotene in plants.

The initial levels were fairly low – in Golden Rice initially there was about 1.6 μg/gm β-carotene – and some of the opposition to that development was centred around the idea that this was too low a concentration and you would have to eat a kilogram of rice a day, or something, to get adequate levels. But recently the second generation of Golden Rice has been developed, where the daffodil gene for the first step in the pathway, phytoene synthase, was replaced with a maize gene, and that led to about a 20-fold increase in β-carotene. This is now a feasible level for introduction into diets, particularly in developing countries. Something like 100 gm a day of rice containing this would give you adequate vitamin A levels.

(Click on image for a larger version)

One that I am a little closer to is the omega-3 plant oils. The omega-3 fatty acids DHA and EPA, which we currently get from fish, are pretty well recognised as the highest functional food ingredient on the radar of the community. People are now starting to realise that they are not getting enough of these nutrients, and we are becoming more and more aware of the health effects and benefits from taking higher levels of DHA and EPA. But our supplies from fish are obviously compromised, and so there is the interest to develop alternative commercial sources of DHA and EPA.

Plants are an obvious target. They only produce the first fatty acid, alpha-linoleic acid (ALA); they don't go on to produce the longer-chain, more unsaturated polyunsaturates EPA and DHA. But we have been able to take the genes from microalgae – in this case by five biosynthetic steps, elongation of the chain and desaturation – and CSIRO recently announced that we had achieved DHA production in Arabidopsis. Since that announcement, there have been two other reports in the literature of similar achievement.

I think this is one of the really promising applications for human nutrition. There is a fair way to go yet, though, because we have to get levels up to commercially viable or appropriate levels.

(Click on image for a larger version)

The last example I want to give is biodegradable plastics from plants – just to switch to an industrial product. Polyhydroxy-alkanoates are a series of biodegradable plastics which are based on a monomer of three carbons, and depending on the number of carbons in the side chain you can produce a range of different plastics.

(Click on image for a larger version)

Polyhydroxybutyrate is the four-carbon monomer, and that is actually produced in soil bacteria by just three enzymes from basic building blocks which are present in plants.

(Click on image for a larger version)

So it was possible to take those three genes, insert them into Arabidopsis – in this case – localise them in the chloroplast and get accumulation of polyhydroxybutyrate in the cell, without affecting the viability of the plant. This is a technology which is undergoing commercial development by a company called Metabolics.

(Click on image for a larger version)

To conclude, I thought it would be worth highlighting what some of the issues around novel plant products might be for Australian agriculture.

The first issue would be consumer acceptance of products. Will consumers actually embrace GM foods with modified composition? We think they will, but now we are actually changing the composition of what they are eating, not producing the same product from an altered plant. Will the opponents and the activists acknowledge the value of community health and environmental benefits? If we are shifting to renewable sources instead of petrochemicals, will they embrace the use of gene technology for that, or will we encounter more opposition? So they are issues.

Another issue is the commercial viability of products. As in Alan Finkel's opening talk, which products will it be commercially viable to develop and introduce? This is really about market size and value in the market, as compared with the cost of development and regulatory approval. And will premiums in the supply chain offset the increased production and segregation and processing costs?

Finally, can we integrate these new crop products into our existing food-based agricultural systems? Can we have coexistence of pharmaceuticals or industrial products with food crops? Where will the liability fall if cross-contamination occurs? And, although we might settle for acceptably low tolerance levels in food crops, where the trait is not a modified product, when we come down to producing industrial products in agriculture we will actually have a zero contamination tolerance in some of those cases, for products that won't be approved for food use. So how will we grapple with those issues?