HIGH FLYERS THINK TANK

Biotechnology and the future of Australian agriculture

The Shine Dome, Canberra, 26 July 2005

Biotechnology: Livestock
by Dr Peter Willadsen, Chief Scientist, CSIRO Livestock Industries, Brisbane

(Click on image for a larger version)

I have on this slide a couple of points about background and where we are at the moment, in terms of livestock biotechnology.

To start with agriculture: the point is often made that agriculture is a decreasing part of Australia's GDP. It is, but it still supports major manufacturing industries and, quite critically, it is still extremely important in terms of our exports, providing a quarter of our merchandise exports.

The major livestock industries are very strongly export-oriented, and that of course plays back into what we can do with biotech and the targets we choose. Of all our export commodities, beef is the largest, and beef, dairy, wool and sheep are all in the top seven of our exports.

The next point is less pleasing. Productivity growth for grains over a 25-year period averaged more than 3 per cent per annum, and most of that was attributed to R&D, giving a total productivity increase of 240 per cent. In livestock, it was 30 per cent.

In terms of the ag-biotech industry in Australia, as opposed to agriculture, the numbers are there on the slide. We would probably put, I would guess, about $100 million per annum, in aggregate, into livestock biotech-related R&D.

(Click on image for a larger version)

So what have we delivered so far? The biggest business to date is probably in DNA-based parentage assignments, and a number of small companies do deliver that. The DNA-based diagnostics for disease, both endemic and exotic, is absolutely critical nowadays but is something that the industry and probably the public are not really aware of. But it is an important contribution.

We are starting to see DNA-based diagnostics for productivity and product quality traits. In cattle, as an example, the two major price determinants for a piece of meat are tenderness and intramuscular fat distribution. Increasingly we have genetic tests for those in beef cattle, and they have been commercialised successfully.

Also, biotech has been critical to a number of vaccines – I have listed here some which either have been totally produced within Australia or else have had a critical Australian contribution – and we have products in the pipeline, including things like viral vectors to deliver genes that are bioactive and often substitute for chemical treatment. So there is some record of achievement.

(Click on image for a larger version)

I think what is very obvious there is that, as Jim Peacock mentioned, the whole transgenic approach to biotechnology has probably been much less evident and less important in livestock biotechnology than in the plant domain.

It is important to recognise that there are always livestock–human links, in a number of aspects. Most of the emerging human diseases that people are really concerned about are, in fact, zoonoses: bat lyssavirus, Nipah, Hendra, SARS, avian influenza and other examples. Also, there are examples – though, I suspect, too few – of the way technologies that have been developed for livestock have actually moved into the human field. I have given two examples here. Interferon-γ assays were first developed in Australia for the eradication of bovine TB, and are now an FDA-approved diagnostic test for human tuberculosis. And viral delivery vectors, developed for livestock, are used in human application.

(Click on image for a larger version)

In terms of the science, the most exciting thing that is happening for us at the moment is the Bovine Genome Project, which to my mind as an old guy is happening with breathtaking speed. We have an international consortium. The major players are the US, where most of the funding has come from, with critical contributions also from Canada, New Zealand and Australia. We already have six-fold coverage, and the first draft bovine genome sequence was released a couple of months ago. We currently have 20,000 SNPs in validation, and the first whole genome scans using 10,000 SNPs have been done. So it is all systems go.

(Click on image for a larger version)

In terms of science future, then, there are some things that it is important to keep in mind. Many of these may be self-evident, but I think we do tend to forget them. First, a cow is as complex as a human being, plus its nutrition is dependent on the rumen, which is arguably one of the most complex microbial systems on Earth.

We do, as I said, have a draft bovine genome sequence, and that will improve within the next year. The sheep genome will follow; the pig genome is currently at only one-fold coverage but is being improved; the chicken genome is complete.

We have more and more microbial genomes. We have them for pathogens and we are starting to get genome sequences for rumen microorganisms. And we have the whole panoply of, if you like, companion technologies that have been developed and are important throughout our whole industry and our whole science.

These genomes are already finding application in genetics and gene discovery programs, but our bigger challenge for the coming years, as with human medicine, as with rodent models, is to understand the complexity of these gene and metabolic networks, with a tiny fraction of the investment in human resources and money to do it.

(Click on image for a larger version)

In fact, it is pretty straightforward. All we need to do is actually understand the black box. I think what we are seeing, in essence, is the serious application of engineering to biology. And the engineering is the key to all of these inputs, and it is producing the flood of data which, in essence, threatens to overwhelm us at the moment. It is exciting, it is stimulating, but I doubt that we really know how to deal with it.

(Click on image for a larger version)

The science issues, then, can conveniently be considered as at least three.

We know that we can go out today and do a microarray experiment or a proteomics experiment, and end up with a list of 100 genes that are of interest. So what? (to use that great American phrase). How do we validate these genes, how do we prioritise them, what is our approach? We know steps in that, but I think we are caught in a bottleneck in terms of making full use of the information we can generate.

Secondly, scientific fields tend not to overlap. Livestock is de facto quarantined from the rodent work, which is de facto quarantined from the human work, to a great degree. And so I think as livestock biologists we are simply not drawing on the information and the knowledge resources that are there in general mammalian genomics to the extent we should. And I think increasingly the reverse will also apply.

Finally, I have a conviction that we will simply not understand biology without a major mathematical input. Here I am not talking about doing a rigorous statistical analysis of, say, a microarray experiment, but the much bigger problem of modelling and then conveying the model information about how the cell or a component of a cell actually works, and ultimately a whole organism. That will require a massive mathematical input.

We have, I believe, a problem, in that most of the data and the people that want to understand the data are coming from molecular biologists and such like, who would have to be the most amathematical scientists on the planet. So as scientists we have, I believe, a real communication issue.

(Click on image for a larger version)

A word on the research scene in Australia: we have at least seven CRCs that have a significant part of their research in livestock biotech; most state departments of agriculture are involved one way or another; and the largest single research effort in one organisation is probably in CSIRO Livestock Industries, but that is still only a fraction of the whole. So is there an issue for us, as a nation, in the fragmentation of effort in Australia? And is there anything we can do about it?

(Click on image for a larger version)

I can't put together a coherent statement – nor, given the people here, would I even to try to do so – about public acceptance of livestock biotech. So I have got some incoherent dot points instead.

I think it is worthwhile keeping the first point in mind. If a piece of technology is really, really successful, by and large people are not aware of it. And that is almost a criterion for success in technology. So, for example, to look at the gene marker work, we are having a lot of success but it is rapidly reaching a stage of complexity that a farmer will find quite difficult to deal with.

We are dazzled by our own technology. We could lose credibility with an excess of genomic hubris.

I think it is worth while keeping in mind that most biotech, particularly in the livestock area, has been accepted by the public without difficulty or contentious debate. While dealing with the difficult issues we need to keep in mind that there is a lot of acceptance as well. And that has been reiterated by other speakers.

The final three points I think have been made by others, and made very well. One of the strengths of biotechnology is the degree to which it substitutes for less acceptable practice, and often as a society we have difficulty in talking about those less acceptable practices.

(Click on image for a larger version)

In terms of environmental impact, the gains that we can make through livestock biotechnology have the potential to have beneficial environmental impacts, but whether they do or don't will depend very much on the market and industry structures and the incentives that society puts in place. They won't, of themselves, automatically produce environmental benefits. And the environment-scale application of some biotechnology, for example in the control of feral pests, could be beneficial but it will and does raise public concern.

(Click on image for a larger version)

In terms of the markets, unfortunately economics is critical. There are many more products we could deliver than are ever likely to be delivered. Public perception is part of that; a major issue, though, is the constraint between the size of the market and the cost of entry.

Arguably, with $100 million a year going to Australian livestock R&D, we should be seeing more products and more companies than we are. Is there a problem, and if so, what is it?

A final speculation or challenge, if you like: we know that a lot of new technologies are not accepted or driven by the technology per se, but by what in computer jargon would be called a 'killer application' – something that everybody needs and can't get any other way. If you go back three decades, personal computers were either niche products or hobbyist things. A number of people have argued, at least, and I think quite justifiably, that what really drove the desktop computer revolution was a program called Visicalc, which was the first spreadsheet program that actually worked. People suddenly found they 'needed' that product, and therefore they bought a computer. So in terms of biotech in general, or livestock biotechnology in particular, is there a 'killer application' out there for, for example, RNAi technology, and what will it be?