Safeguarding Australia
Plant science at the frontiers: Safeguarding Australia’s primary industries
Tuesday 1 April 2008
Dr John Manners
Assistant Chief, CSIRO Plant Industry
Research program leader, CRC for Sugar Industry Innovation through Biotechnology
Dr John Manners is a molecular plant pathologist with research interests in fungal pathogens and the molecular dialogue between plants and pathogens that determines disease outcomes. He also has interests in applications of plant biotechnology for crop improvement and disease control.
Plant science at the frontiers: Safeguarding Australia’s primary industries
(Click on image for a larger version)
I would like to start by thanking Jim Peacock, and the other members of the Academy who organised this lecture series, for the invitation to present to you. The title of my talk is actually a bit of a double entendre, in that I am going to be talking partly about some episodes of battle at the geographical frontiers between Australia and invading pathogens, and how we have responded and how we assess such threats. On the other hand, I am also going to be talking about plant science and the study of plant-pathogen interactions, and how that ‘frontier' of science is helping provide more durable resistance for the future.
I know that many of you here are not plant pathologists, so I thought I would start by giving you some historical cases so you can appreciate that classically diseases have been very important both to plants and to humanity.
(Click on image for a larger version)
The iconic plant disease epidemic is that of the Irish potato famine. Recently, after a lot of study of records, some data has appeared of the impact of that famine on the population. The table shown here illustrates that this disease arrived in Ireland in the 1840s. It is caused by a pathogen called Phytophthora infestans – ‘plant destroyer', in a literal translation. This is a pathogen which is fungal-like in form, but evolutionarily is actually more related to algae.
Looking downwards from the top of this table, you can see at the start, before the epidemic in 1840, large numbers of acres, a large amount of yield and a population of around 8.2 million people. By the time this epidemic was over, down at 1848–1851, yields had been reduced to about 20 per cent of what they were originally, very little potato was being planted and the population had been reduced by two million – of which one million died of starvation and one million emigrated. They emigrated to many places, including Australia, so this has global importance.
(Click on image for a larger version)
The second point I would like to make is that even though this is an iconic disease and very well understood, we are not immune through that understanding to further invasions of the same pathogen. So we had further epidemics from this disease in the 1990s.
In the US, new strains of this fungus which emerged from Mexico overran the potato industry, and other strains migrated to Europe, where a second epidemic occurred. This invasion is really what I call a genotypic invasion, where a new genotype arrived and overtook the old genotype, as displayed on the graph here, through the US. The lesson is that it is not only the disease and the pathogen that we have to consider as a threat, but also the genes and the genotypic complexity of the pathogen.
(Click on image for a larger version)
I know that potatoes may be a staple in Ireland, but probably other staples exist in Canberra, so I thought I would tell you a little bit about grapevine and wine. Again this is a historical account of an epidemic, and it leads to an indication of the the legacy of incursions.
Powdery mildew disease is a fungal disease which infects the leaves and the fruit of grapevine. Until the mid-1800s, grapevine cultivation in Europe was fairly disease-free. Vines were brought in from a new centre of origin, a Vitis species in North America, and with that came powdery mildew disease. It was first discovered in Kent and subsequently in Paris, and then an epidemic moved through southern Europe. There are some fairly eloquent terms on this slide to describe the disease, including that it might be caused by ‘evil emanations from the telegraph wires'.
It was noted that some American vines growing in Tuscany were immune, or as people would say at the time, they remained ‘virgin under the malady's embrace'. Subsequently, further American vines were brought into Europe for breeding, but because there was no biosecurity involved, this actually brought the other two major pathogens of grapevine into Europe. (Those American vines, although they brought resistance, were never successfully used in breeding, because they had other, undesirable, attributes.)
(Click on image for a larger version)
The legacy of this is that grapevine cultivation is now the major user of fungicides in Europe and many other grapevine-growing areas. This slide shows fungicide usage in France through the 1990s. The blue line indicates what is applied to grapevines, the other lines are for what is applied to cereals – which are 70 per cent of the crop area, whereas 70 per cent of the fungicide is applied to grapevine. So once diseases arrive, it is irreversible and their impact is substantial and there for you to manage forever.
(Click on image for a larger version)
When we are looking at plant biosecurity, we have to look at three major aspects. With regard to pathogens and pests: what is the threat, what is the likelihood of incursion, what is the likelihood that the pathogen is going to evolve, either exotically or endemically? Second, how do we use the host diversity, host resistance, how do we deploy those? And also, with regard to the environment, how do we actually conduct our agriculture and what is the climate impact on diseases?
Most modern biosecurity workers would say there is a fourth dimension to that, which is economics and politics, which are involved in the discussions around quarantine and trade blockouts. But I am not going to discuss those today.
A lot of current biosecurity research and industry action revolves around the issues of assessing threats and assessing what we are going to do if we have an incursion.
(Click on image for a larger version)
Australia has one of the most, if not the most, vigilant plant biosecurity networks in the world. If you go to the web sites of these agencies, you will find an enormous amount detail on the threats to particular industries and action plans for those industries. I am not really going to talk about that type of science, very important though it is.
(Click on image for a larger version)
In biosecurity research there is often the type of program set out on this slide. This is straight out of the Plant Biosecurity CRC [Cooperative Research Centre]: preparedness and prevention, diagnostics, surveillance and managing impact. The emphasis of my talk is going to be genetic preparedness, which is a part – only a part – of the impact management area of biosecurity research.
Genetic preparedness really means understanding the genetic structure and the genetic threat of the pest or the pathogen, its population genetics, its evolution, and developing tools that allow you to understand and detect and dissect that in diagnostics.Also, at the host side, it allows one to undertake either pre-emptive breeding or rapid-response breeding to identify resistance sources, resistance genes, novel resistance and plan how they are used in crop improvement.
(Click on image for a larger version)
In tonight's talk I am going to start with some discussion of the challenges for plant biosecurity, in particular.
(Click on image for a larger version)
One of the issues with plant biosecurity is really how it is perceived, firstly in the community. In respect of animal diseases as shown on this slide, I am sure if you went down the mall here and asked people whether they had heard of avian influenza or foot-and-mouth disease or rabies, you would probably get a yes from most people. If you went to them and said, ‘What about karnal bunt and plum pox and pine pitch canker?' probably people wouldn't know. But these are the major threats through plant diseases, and we have to maintain vigilance of our basal plant production systems.
I will use this slide also to illustrate diversity. If you are dealing with domesticated animals, you are really dealing with a handful of species. If you are dealing with plant threats, you are dealing with a very, very large number of plant species and each particular plant species has an extremely large number of pathogens. So there are issues around prioritising what you are going to work on.
(Click on image for a larger version)
This slide illustrates what I have just said, in this case for bananas. There are many fungal, nematode, viral and bacterial diseases of bananas, several of which are not in the country. And, essentially, you have to manage plans for all of those, be they plans for genetic preparedness or be they for incursion management.
(Click on image for a larger version)
Finally, most microbes that infect plants, as Jim indicated in his introduction, undergo rapid evolution. This slide looks a bit complicated, but I will walk you through it.
If you are actually managing a pathogen, you have to try to manage to minimise the diversity of that pathogen that confronts the crop. You see here a risk assessment framework that was developed by Bruce McDonald.
On the vertical axis we assess the potential for genetic recombination of a pathogen. If its reproduction is just asexual, it is likely to be clonal, therefore it is going to have low genetic recombination. If it is only sexual, it will have a much higher level of genetic recombination. If it is inbreeding it will be at the low end of the ‘Sexual' box, and if it is outbreeding, outcrossing, it will be at the high end. And if it is both, it will be much higher again. The evolutionary potential will be higher.
Along the horizontal axis is a measure of the pathogen's ability to undertake dispersal or genotype and gene flow. If it is in a soil-bound organism it will be low; if it is splash-dispersed it will be high; if it is wind-dispersed it will be even higher.
Just to illustrate another classical biosecurity event: prior to 1930, particularly in the US, Puccinia graminis (the stem rust fungus, the rust fungus which infects wheat) was up in the top right-hand corner of this framework, because it was basically a sexually and asexually reproducing population, and it has wind-dispersed spores. The sexual reproduction in this fungus actually occurs on an alternative host called barberry, or Berberis vulgaris. In the US the countermeasure was essentially undertaken as a process of elimination, which moved the risk assessment of the stem rust fungus down. It is still a considerable foe, but we know that the genetic complexity now has been suppressed through this action.
(Click on image for a larger version)
Finally in this regard, shown here is an analogy that one uses in crop protection in general. It is taken from Lewis Carroll's Through the Looking Glass, where Alice is running as fast as she can but going nowhere. The Red Queen explains to her that, basically, ‘That is just what you do here. If you don't keep running fast you go backwards, and if you actually want to get anywhere you have to run twice as fast.' So if you are countering migrating and evolving pathogens and pests, you have to keep active just to keep things as they are. And if you don't keep vigilant you will go backwards.
(Click on image for a larger version)
Now I will describe some recent events around plant biosecurity, and some incursions and some threats. I am going to start with sugarcane smut.
(Click on image for a larger version)
The sugarcane industry in Australia is an important industry – it has a value of about $1 billion to $2 billion, depending on the world sugar price. It extends up the coastal area from northern New South Wales to Far North Queensland, with a small growing area in north-western Australia, on the Ord River.
This was a disease that previously was not present in Australia. It is a smut caused by a basidiomyces fungus. It forms black, whip-like structures such as the one shown in the panel at the top right-hand side of this slide, which are covered in spores. The fungus invades the bud in the stalk at planting, grows through the host and emerges at the growth point with this massive whip structure, which is essentially loaded with spores. It can also cause severe stunting.
The spores are primarily wind-blown, and it has a very high contagious potential. You see here some recent data here from BSES Ltd, where they have measured 1014 spores per day being produced on infected paddocks.
(Click on image for a larger version)
I will just run quickly through the spread of this disease. (There is a sense of inevitability that builds as we go through this slide.) It was first identified in South Africa in the 1800s, then India and later Brazil, then Hawaii in the 1970s, Central America, the USA, and finally, in 1979, Indonesia. When you look at that, it is actually quite surprising that it took almost 20 years to move from Indonesia to north-western Australia. I think that is a credit to the quarantine processes that are in place in Australian industry. Subsequently, in 2006, it has moved to the major sugarcane production areas in Eastern Australia.
(Click on image for a larger version)
For a wind-blown spore, it is not difficult to see how something arrives in Australia from Indonesia. This slide includes a shot of a storm system off the north of Australia, covering Cape York, New Guinea, Timor and the Indonesian islands. We believe that the disease arrived essentially through similar climatic events. We know from DNA analysis at BSES Ltd. that the fungus that arrived in Australia is exactly the same as the fungus that is present in Indonesia, and different from that in other parts of Asia.
Despite attempts to limit its spread from Western Australia to the eastern coast, it arrived in 2006. Realistically, for a pathogen like this, once incursion has occurred and it has completely spread, resistance is the only control option. It is not present in all regions in Queensland, but it is in at least three.
The genetic preparedness here was basically around developing cultivar exchange programs with other sugarcane industries that had smut, but most important was developing a cooperative arrangement with Indonesia, where the Australian industry could actively evaluate its clones and its breeding lines against smut pathogen. This took place before the incursion occurred here.
(Click on image for a larger version)
Even so, if you look at the ranking shown here of susceptibility versus resistance of Australian lines, you see that most of the varieties that were in current use when the incursion occurred in Queensland are highly susceptible. This is primarily because smut resistance comes with a cost. Most smut-resistance lines were not extremely high sugar-yielding lines, and therefore there was a tendency to keep smut-sensitive lines in the field.
(Click on image for a larger version)
Nonetheless, the breeding program has undergone a change already. If we go back to 2000 we see that most lines in sugarcane breeding were sensitive; if we come now to 2007 we see that virtually none are smut-susceptible. Almost all lines are either resistant or intermediate resistant.
We see in this graph the take-off of resistant lines from 2006. This is the product that came from the Indonesian program. So the industry is in a state of massive change in terms of replacing all its current varieties and its entire breeding program.
(Click on image for a larger version)
Because smut comes at a cost and because we would like to combine multiple types of resistance, what we would really like to do is to have DNA markers that we can tag for disease resistance. In sugarcane this is not straightforward, because sugarcane has the most complex genome of all crop plants. Shown at the top left of this slide is a stylised version of the sugarcane genome. It is an interspecific hybrid, so there are two colours here, and it is essentially a polyaneuploid, so there are variable chromosome numbers between homeologous groups, and each individual has a variable chromosome number.
One of our best genetic maps is shown here stylised, at the top right, about 65 per cent complete. Using tools like the new sugarcane DArT chip, which is a marker system with 7000 markers on it, we have been able to identify candidate rust resistance markers – and we hope that our sugarcane breeders will not be navigating the world with maps like the one depicted here as used by 17th century explorers, but will very soon be using maps more akin to those used by 747 pilots.
(Click on image for a larger version)
Staying in the north: black sigatoka is a disease of bananas. It is caused by the fungus, Mycosphaerella fijiensis. This is widely distributed around the world in most growing areas and is controlled through fungicide applications – and quite massive fungicide applications. Essentially, if you are in Central America this disease is controlled by weekly applications of fungicide. It is absent from Australia, although we do have a similar but less virulent disease called yellow sigatoka.
Black sigatoka is controlled by quarantine. It is present in Papua New Guinea and in the Torres Strait, and we have had incursions of this disease in North Queensland which have been managed through eradication.
DNA tools were quite critical in the eradication program, because one could go to fields and immediately distinguish between black sigatoka infected fields and yellow sigatoka infected fields, which would guide the eradication. At the time, I was in the CRC for Tropical Plant Protection (not involved with this program directly) and it was a wonderful thing to see our young scientists taking their PCR [polymerase chain reaction] machines up to Tully and places like that, and working with the pathologist, doing real-time analysis of thousands of samples to guide the eradication, which really gave me a lot of faith in the power that DNA diagnostics have in guiding this exercise.
Genetic preparedness with bananas is probably not very high, because there is no banana breeding program and also the supply chain is very attuned to a single cultivar. This is probably a crop where a GM [genetic modification] solution would be very useful, but it is in bananas, which is a fresh fruit, and that may be some way off in the future in terms of GM acceptance.
(Click on image for a larger version)
The last example of threats I would like give is karnal bunt. This threatens our wheat industry, our major crop across Australia, worth $5 billion plus per year. This is an interesting disease. It is reasonably well distributed now in many wheat-growing areas. It is caused by another basidiomyces fungus and infects the florets but only partly infects the grain.
Shown here is an infected grain lot and, below it, an infected kernel where you can see only a relatively small dark infected area, whereas we have another endemic disease at the moment which is very similar, common bunt, which affects the whole grain and basically disintegrates it. Karnal bunt doesn't affect yield but it does produce a ‘fishy' odour, which is a major problem for quality.
At the moment there is an awareness campaign and a lot of preparedness for the incursion of this, should it occur. Because it occurs elsewhere, resistance sources have been identified and DNA markers are now emerging, and these could be used in Australia if this disease were to arrive and take hold. It is unlikely that they will be used, I think, in a pre-emptive form. We would probably wait until the actual disease is here, but as long as the tools are ready, you have the preparedness.
(Click on image for a larger version)
Now I am going to move on to rust diseases as an example of evolving pathogens – in particular, rust diseases of wheat – and I will move also from describing a disease threat into how molecular tools may be able to assist, both now and in the future.
(Click on image for a larger version)
In wheat there are three major diseases – stem rust, leaf rust and stripe rust – caused by different types of pathogens. These are present in multiple races, their spores are airborne, and they have a biotrophic lifestyle, which essentially means that they live on living tissue.
(Click on image for a larger version)
Australia has the benchmark system for managing rusts globally, through the Australian Cereal Rust Control Program, with the participants listed on this slide. This undertakes surveys of races, resistance testing et cetera to guide the breeding programs.
(Click on image for a larger version)
We know that rusts are going to keep coming. You only have to look at history to see that – this slide shows a list of rust arrivals over the last century. We also know that rust is going to evolve, even endemically. Just recently in Western Australia a resistance gene that was deployed against stripe rust, Yr13, was overcome by a new strain which appears to be just a simple mutant of the pre-existing strain, and most of the significant Western Australian wheat varieties listed here have gone from being resistant to being either highly susceptible or moderately susceptible.
(Click on image for a larger version)
This is not just an Australian problem; it is now a global problem. At the moment there is major concern over a new race of the wheat stem rust fungus called Ug99, which is on the move. This is a strain that defeated a resistance gene called Sr31. It emerged in Uganda in 1999 and has now moved through Kenya and Ethiopia into the Arabian Peninsula, and it is thought that it will most probably move through the Middle East and into the production systems of India.
Sr31 is not a highly important gene in Australia, because it is closely linked to the sticky dough locus. But there have been reports of mutation of this strain to overcome Sr24, and therefore it is something that needs to be taken very seriously. It is a highly virulent strain. But many Australian scientists are involved in this as a global problem.
(Click on image for a larger version)
So how do we counter these evolving and migrating rust pathogens? It is really about combining various forms of resistance: resistance to the different diseases and also to different races/pathotypes of those diseases. There are two types of resistance from the plant perspective – seedling resistance and adult resistance – and they also need to be combined.
(Click on image for a larger version)
You can only do this type of combination using DNA markers. You identify pieces of DNA closely linked to the gene, and you follow the pieces of DNA, not the actual resistance response because once you have more than one or two resistances together, it becomes very difficult to know whether you actually have the full gene set or not.
(Click on image for a larger version)
This is called pyramiding genes. Shown here is a slide from Jeff Ellis, who is in the audience. This is not totally straightforward, because when you are actually making crosses to bring together all these genes, the chances of identifying a wheat genotype that has them all together are very low, so you have to look at extremely large numbers. And if you are in a breeding program, if you do identify a genotype that has got a pyramided set of genes, then as soon as you use that as a parent in a cross the pyramid all falls apart and you have to start again.
(Click on image for a larger version)
GM technology really does offer, ultimately, a solution to this, because if you clone the genes and you put them all together on a single cassette, then in effect you create a multigene resistance locus which you can use in breeding, just as you can any other major gene. Cloning these genes out of wheat is not exactly straightforward but it is ongoing work, and there are lots of promising results coming out in that area.
(Click on image for a larger version)
Perhaps the most exciting things that have happened in plant pathology, and some of the most exciting research I have seen in CSIRO, are in the work on understanding resistance gene proteins for rust pathogens. This has been undertaken in Jeff Ellis's team, with participants such as Peter Dodds, who is also here tonight. This has led to a deep understanding of the molecular interactions. We know that resistance genes code for proteins that are like receptors, and we also know that those receptors bind to protein molecules that are released by the pathogen. It is this specificity that gives us a resistance response.
Genetically, we have known that there have been genes in the pathogen and genes in the host that act complementarily, but now they are being identified. Not only are they being identified, but structural models for the interaction are beginning to emerge, so that one can start to actually develop design principles for resistance genes which may lead to more durable resistance whereby this interaction cannot be mutated away in the fungus.
(Click on image for a larger version)
In addition to that, the actual cellular processes in the interaction are being elucidated. I am using here a slide prepared for rusts from Jeff Ellis and Peter Dodds, but this is actually now emerging as a common theme both for bacterial pathogens and for several other pathogens, such as Phytophthoras.
Basically, here in this rust example, the fungus invades a plant cell and forms a fungal structure inside the cell which is invaginated by the plasma membrane. So the fungus has, in a sense, stuck its fist in a balloon inside the cell.
The fungal structure actually secretes proteins which enter the host cytoplasm. These are called effectors, and the role of these effector molecules is to manipulate the host's defence – to suppress host defences so that we have a susceptible interaction. A few of these have been identified in rust fungi; it is now suspected from fungal genomics that there are hundreds of them that are injected into cells.
Resistance genes appear to have evolved and appear to actually recognise specific members of these effector genes and thereby alert the host to the invasion and trigger an immune reaction. In a pathogen like rust, which is a biotroph and needs living cells, a common immune reaction is for the cells to die, a hypersensitive response.
The possibility is, therefore, to try and design resistance genes which recognise effectors that the pathogen just cannot do without in order to be pathogenic. This is a long-term strategy for durable engineered rust resistance.
(Click on image for a larger version)
Perhaps a slightly less molecularly based process at the moment is understanding how resistance works in other cereals. Most cereals except rice have major rust fungal pathogens. Rice seems to have very strong non-host resistance to wheat rust, and has a resistance gene-like response to rusts.
Shown here is some work from Mick Ayliffe, in CSIRO. When the wheat rust fungus infects rice and attacks rice cells, as shown at the bottom right of this slide, you end up with a host response underneath that which eventually leads to cell death. (A macroscopic-type symptom is shown in the image to the left of that one.) This hypersensitive response is typical of resistance gene-type response, and it will be amenable to gene isolation. We know this is very durable, because rice does not have these pathogens, and it is an exciting route now to take, to try and isolate these genes.
(Click on image for a larger version)
There are a number of lessons from the rust program. First of all, it is a major disease issue for major crops and has led to world-class monitoring systems. But we need multiple resistance sources, we need to stack them, and there are now some exciting advances going on in understanding the molecular interactions, which lead to the possibility of very robust GM solutions to rust pathogens.
(Click on image for a larger version)
If you are waging a war, the best thing to do is to open up another front on your enemy, and wage war on two fronts. I have described the biotrophs, but our fungal invaders have also now developed other types of pathogenic forms which are becoming increasingly important: necrotrophic fungi.
If we look at the major wheat diseases listed in 1969 we find mostly biotrophs: smut, powdery mildews and rusts. But when we look in 2005 we find a whole cohort of new diseases: crown rot, head blight, take all, tan spot, leaf and glume blotch – these are all caused by necrotrophic fungi. And the reason these have emerged is new practices. Basically, in conservation farming these days, growers are using stubble retention to retain water. Necrotrophs don't need living tissue; they can grow quite happily on dead tissue. So they colonise the stubble, they can overseason, and the inoculant can build up. This has led to several global epidemics. One is head blight, caused by Fusarium, in the US and in Europe and China. And most of these fungi are toxin producers.
(Click on image for a larger version)
I am just going to use an example which I have borrowed from a colleague, Richard Oliver, who works in Western Australia and is Director of the Australian Centre for Necrotrophic Fungal Pathogens. It is an example of how these pathogens infect, and also of an evolutionary basis for the emergence of these pathogens.
(Click on image for a larger version)
Pyrenophora tritici-repentis, which is a fungal pathogen, is an important pathogen of wheat these days, but it wasn't really considered to be a major pathogen of wheat until the 1940s. The pathogenicity mechanism for this pathogen has been elucidated by workers in the US. The fungus produces a protein toxin called ToxA which then is recognised by wheat cells containing the gene Tsn1, which is in most wheat varieties. Once this happens, it triggers cell death in the host. And because this is a necrotroph, it can then feed off the dead cells. So it is the opposite of the rust pathogen.
(Click on image for a larger version)
Another pathogen of wheat, which has been around for much longer, is Stagonospora nodorum. The illustration on this slide shows the symptom. It too is thought to produce toxins.
(Click on image for a larger version)
One of the exciting things happening in fungal pathogens at the moment is the availability of genome sequences. Multiple fungi or fungal pathogens now have had their genomes sequenced, and we are getting to the point where we can start to do comparative genomics across the suite of pathogen threats that we are confronted with.
The genome sequence for Stagonospora and the genome sequence for Pyrenophora are available, and when these were compared it was found that the ToxA gene in Pyrenophora and in Stagonospora was virtually identical in sequence. If we look at the sequence similarity as shown on this slide, we see that after an initial region where it is around 80 per cent, as soon as it hits the next region it is up to 100 per cent and then it drops off again to around 80 per cent. So they have, effectively, identical toxin genes – and also, intriguingly, transposase, which is a mechanism for moving genes around, next to the ToxA.
(Click on image for a larger version)
When you look at the populations globally of these pathogens, in Stagonospora you can find 13 ToxA types of gene; in Pyrenophora you can only find one. So this has probably emerged very recently in Pyrenophora, and the simplest explanation is that this has actually acquired the gene from Stagonospora, probably around the 1940s when it became a major pathogen of wheat.
I think that intriguingly this opens up the notion, proposed by Richard Oliver, that we should be thinking about breeding generically against toxins, and understanding the toxin threats more, perhaps, than the actual organism threats per se.
(Click on image for a larger version)
I am now going to move through what is the status of GM plant protection in plant biosecurity – I have alluded to it in previous talks. At the moment, most GM crops that have been deployed are either insect-resistant or herbicide-resistant involving weed control, or a combination of the two. There are very few instances where GM technology has been used for disease. I am going to talk about one which is actually a GM approach to a biosecurity question, the papaya ringspot virus, as it was used in Hawaii to counter this viral invasion.
(Click on image for a larger version)
Papaya ringspot virus is a major problem of papaya, or pawpaw, production worldwide. It is transmitted through insects, and there is no known resistance.
(Click on image for a larger version)
The virus was first detected in Hawaii island in about 1978 in the Hilo area, as shown on this slide. This was a small production area. In Puna, which is the main production area, it was not detected until 1992.
(Click on image for a larger version)
But when it was detected it had a substantial impact on papaya production. Over the next five to six years, papaya production was, effectively, halved.
Interestingly, at the same time Dennis Gonsalves, at the USDA [United States Department of Agriculture] had already developed papaya plants which showed strong resistance to papaya ringspot virus, using GM technology. Dennis used the technology which was initially demonstrated in Roger Beachy's lab, simply using the pathogen gene expressed in the plant itself. This, for what was then a reason unknown, gave rise to resistance.
(Click on image for a larger version)
That technology was then rapidly commercialised in Hawaii and released in, I think, 1998. The view on this slide shows the impact of this. The papayas labelled ‘Solution' are GM papayas in 1999, and some old, non-GM papaya plantations, labelled ‘Problem', can be seen next to that.
This gene is, effectively, a coat protein gene which works in some respects like an antisense gene. But nowadays we understand the mechanism for antiviral technology, through the work of scientists such as Peter Waterhouse, in our Division. We now know that the reason this works is that when the pathogen identifies foreign double-stranded RNA it will, in fact, chop that RNA up. This is a natural antiviral defence system; it is also a system that the plant uses to regulate its own gene expression.
By refining that technology, we now know that actually one can use much smaller pieces of DNA to silence genes and create viral resistance to almost any virus – and also with greater efficacy and greater specificity.
So why isn't this happening? Why aren't we deploying lots and lots of viral-resistant plants?
(Click on image for a larger version)
Basically, when Dennis Gonsalves drove the commercialisation of papayas, it was in an era when there was not a heightened media attention on GM crops, and so it went through. And also it was in the US, which I suppose was an easy and fertile land for deploying this technology. These days, GM technology is under much greater regulation and the costs for developing GM crops have become quite high. According to published estimates it is now taking about 12 years to bring a GM crop to market in a global sense, at a cost of around $100 million.
A lot of that cost is not actually at the discovery and developing the plant end of the process but at the regulatory and product supply end, and in managing the liability issues and the value chain – and the consumer.
So I think that in the short term it is very likely that any GM deployment from a plant biosecurity point of view will be used in big crops where we can take a global perspective, and for major pathogens. For example, in Australia wheat is our biggest crop, wheat is a large crop globally, and I think that we are likely to find that wheat is going to be a target. And rusts and necrotrophic pathogens like Fusarium diseases are likely to be targets.
It creates issues, I think, for people in small crops who may have a particularly strong genetic resistance they want to deploy but who cannot actually manage the capital expense of putting it through the system.
(Click on image for a larger version)
Time is short, but I will talk just very quickly about climate change.
(Click on image for a larger version)
We do know that climate affects pathogens and disease development. This slide shows the results of a 100+ years experiment in Rothamsted, where wheat is grown in the same paddock over and over again. You can go back to that material, and using a DNA method now, PCR, measure pathogen incidence. You see here a ratio of two pathogens over time, plotted with the SO2 (sulphur dioxide) emissions over that period. On interpretation, this shows quite clearly that climatic effects are going to affect pathogenesis.
(Click on image for a larger version)
There are ways of testing this, using carbon dioxide-type in-field systems, where actually in the field you maintain the carbon dioxide at levels that we predict will occur in a climate change scenario. Although the results of these studies are just emerging, some of the results that are coming out of them suggest that plants will in fact change in their sensitivity to pests and pathogens. The data at the moment is primarily with insect pests, where we have increased herbivory on plants grown in these environments.
This, I think, is an important aspect of the future, considering the impacts of the environment.
Just to finish: plant pests and pathogen threats have long and deep impacts, and are important. We have a strong, vigilant plant biosecurity system. Any preparedness, I think, has to include genetic preparedness, and this has to include understanding the pathogen, how to use that knowledge to reduce pathogen diversity and increase host diversity, and then to develop robust, pre-emptive breeding strategies and ultimately to develop strategically the GM tools which can provide us with highly robust and durable protection.
Also, we have to bear it in mind that the environment that we actually are breeding for now may be different from the environment in the future.
(Click on image for a larger version)
As the Red Queen said, we have to run as fast as we can to stay still.
(Click on image for a larger version)
Most of what I have presented has been provided by a range of people, some of whom are in the audience here. I acknowledge Jeff Ellis's group, with Peter Dodds, Mick Ayliffe, Evans Lagudah, Sukumar Chakraborty, in Brisbane, and Ian Dry, in Adelaide; Barry Croft for the slides on smut incursion; and Richard Oliver, Bruce McDonald and Dennis Gonsalves for their part of the story. I would refer people to the web sites of DAFF [Department of Agriculture, Forestry and Fisheries] and PHA (Plant Health Australia) for more detail on some of the other threats and diseases.
Thank you.
Discussion
Chair (Jim Peacock): Thanks very much, John.
Question: Just picking up on your comment about the Red Queen, if we do invest in developing GM-based resistance in crops, what are the lifetimes of those crops likely to be?
John Manners: It is difficult for me to predict the future but I think we can talk about the present, in terms of the Bt gene that is deployed at the moment against cotton bollworm and boll weevil. You have to manage the lifetime of those genes, and there are actually some papers in Nature Biotechnology at the moment which indicate that in countries where that is being managed, through correct use of refugia, minimising the use of single-gene deployments, then you can actually manage for a more durable future.
I don't think GM resistance is going to be any different from any other resistance. It is like the rust system: we actually have to have processes in place to manage that resistance into the future.
Question (Adrian …): Could I ask the same question, slightly in reverse – it is a question of timing back rather than timing forward. Is there any evidence of exactly how fast some of these fungal genes are evolving? And as a result of that knowledge can you predict then what is happening out there in nature?
For example, the sugarcane smut, which arose in Africa, clearly is not an old disease, because sugarcane arose in Papua New Guinea, whereas papaya ringspot is a very diverse plant pathogen and is likely therefore to be re-emerging. Whenever you manage to build up resistance in one particular country it is likely to come in again from elsewhere.
The question I am really asking is: have you got any idea from ‘back there' what the rate of evolution of these genes is?
John Manners: Well, I guess it depends which particular genes you are talking about.
Question (cont.): Particularly the ones that Jeff Ellis is working on.
John Manners: Maybe Jeff can answer that, but my understanding is that there is fairly rapid diversification in the resistance genes in response to also rapid diversification in the effector and avirulence genes. I am unsure of the time frames of those, but I would have thought that, in terms of our experience in agriculture, those time frames must be very short. Basically, you can deploy a resistance gene for a rust pathogen in Australia and you know it will be defeated fairly quickly, in a matter of years.
For some of the other genes, such as the horizontal transfer events, these are occasional events but what we do know is that they are happening far more frequently than we previously expected. So, in the necrotrophs, we know that the same cluster of genes exists in different genetic backgrounds, so several different species have the same cluster, we know that there are the same strain-specific chromosomes, which we can also interpret as horizontal transfer accoms and in our lab we have demonstrated that you can actually shift them between incompatible types.
I think that the plasticity of fungal genomes is extremely high. I can't give you any specifics at the moment, in terms of time frame. But in response to selection it will be high.
Jim Peacock: Jeff, do you want to add to that?
Jeff Ellis: The comment that I would want to make is that clearly the resistance genes, and things that they recognise, in natural systems have evolved as part of a host-pathogen co-evolutionary arms race. Agriculture is somewhat different from natural systems, in so far as the plant is attempting to survive by producing enough seed for the next generation, whereas in agriculture we have the added challenge of not producing enough seed for the reproduction of the plant but the continuation of feeding the human race. So the demand on that system to produce food is much higher.
We are going to have to adapt what is occurring to the agricultural system by better understanding: first of all, how does it work in nature, in terms of an evolutionary process, and secondly, how does it work at the biochemical level, so we can make modifications? The hope is that through this process we will be able to extend the life of individual resistance genes, and then extend that life by another factor by pyramiding these genes, using some of the technologies that John has mentioned tonight.
John Manners: Just coming back to your comment on sugarcane smut, Adrian: it is quite clear now that a lot of diseases in that period were actually moving with the biological material that was being taken and being discovered. I think the Stagonospora pathogen I indicated is another example of a South African emergent disease, although it probably came from Europe in plant material – the environment revealed the pathogen there.
Question (Bill Scowcroft): Thank you very much, John. That was a wonderful view of where the science is taking us. But having spent the last 10 years of my life in a commercial area, I believe we must sit back and think a little bit more about the commercial aspect. The reason GM technology has worked so well is that there is money to be made out of selling individual seeds – canola and maize, for example, particularly in North America.
But when we come to the large-volume plant crops, like wheat and rice, where the individual value of a planting seed is not very high, we have to have a different philosophy. In other words, it has to be driven by public funding. I cannot see the wheat industry or the rice industry worldwide actually being prepared to pay the cost that goes into developing GM-based crops in these two species. That has always given me a problem of how we get the concern of the public purse behind it. Would you comment on that?
John Manners: I guess I can comment on two aspects of that, Bill. One is that yes, I agree. Perhaps the growth we have seen recently in large foundations and large corporations in developing research programs which are going to benefit developing countries, particularly – some of those are using GM technologies – is a step towards getting finance, not government finance but actually private finance, into that sector to develop GM crops for developing countries.
I think the severity of the threats is driving a rethinking by the large corporations of their view of wheat. I think the Fusarium head blight problem in North America has caught the attention of some of the large agribusiness and they are actually developing GM products. They will accompany those with other traits such as herbicide resistance, for example, where they can leverage financially.
Effectively, also, over time as we get more comfortable with GM technologies in the agribusiness group of crops, and the record of the technology becomes more familiar to us, the record of safety becomes extremely robust, then I think that some of these costs that we have to meet at the moment will reduce.
I think your comment is right for where we stand now, but perhaps in a future scenario that may not be the case.
Question (cont.): Just a point following from that: the biggest global threat to the use of GM crops is the EU. How do we confront that?
John Manners: The EU is very diverse. It is seen as a bloc, but it is actually a very diverse group of countries. It is not the whole of the EU that is actually the issue, in my mind; it is a few powerful countries in the EU. And I think that as we see a greater diversification and a greater shift in the power base in Europe, this will erode.
Question: Mine is a practical question and I am sure you have got an answer, because you have been doing this stuff for a long time. When you are breeding new plants to counter some new threat, whether you are using conventional techniques or GM techniques, and you say, ‘Eureka!' in the lab, you have found the one plant that is going to do it but you have got to replant however many millions of acres and produce however many millions of tonnes of seed, how do you do that from your one plant in time?
John Manners: I can use the sugarcane smut example to answer that one, because it is a problem which has got some very specific needs which meet that question. Sugarcane is a vegetative crop. So you don't actually make seed; you just cut the stem up into small nodes and then plant those. Each sugarcane plant probably produces only about eight or so progeny, rather than hundreds, so production of planting material is a big issue.
At the moment the way that is being addressed is, first, by diverting a lot of cane from sugar production now to plant cane, so rather than being taken to the mill it is taken between different regions and replanted. But the major innovation that we have had in sugarcane for vegetative reproduction is through micropropagation. That is a technique where we actually use thin sections of the apex of the sugarcane and generate many, many plants. Using the right conditions, you can do that and those many, many plants become exactly the same as the parent plant they came from, and you can make hundreds from a single plant.
That is actually being used in providing enough smut-resistant material to distribute around the industry. If we just waited for the stem section method, it would take too long.
Jim Peacock: I am sorry I have to cut this off. You obviously agree that this has been a really wonderful lecture by John. To me, the way you developed the story, John, showed us how we have moved from the era when plant pathologists used to use rather mystical terms to indicate what was going on without truly understanding, to the present day when we are beginning to understand, at a much greater level of detail, the molecular interaction, the way in which the fungus talks to and challenges the plant when it attempts an infestation. This I think is the great hope, whether it is GM, non-GM or, most likely, a combination of both, with careful management of the two, for a safer and more robust future in crop management.
Thank you once again, John.
