Breeding vs. genetic engineering: Debating the best approach for livestock

Last year Stuart Brand, principal author of the back-to-the-land, simplicity-is-god, do-it-yourself, hippie bible known as The Whole Earth Catalogue, opined about the irreversibility of technological progress. "Good old stuff sucks. Sticking with the fine old whatevers is like wearing 100 per cent cotton in the mountains; it's just stupid," he said.

Brand — who was also one of the promoters of the mind-changing powers of LSD — then urged the truly smart modern citizen to buy the latest laptop, the newest drug and "genetically modified food [from a farmers' market, preferably]."

I mention this remark both because it captures the absolutism that has characterized the debates over genetically modified food, and because of something quite surprising. Without most of the public being aware of it, geneticists have begun to believe that disdain for the "good old stuff" when it comes to traditional livestock food production is somewhere between chuckleheaded and — incredible as it sounds — a technological backward step. 

'We can use the new techniques of genomics selection and marker-assisted selection in breeding better animals, and in so doing we can avoid making transgenic animals — at least into the foreseeable future.' —Max Rothschild, Center for Integrated Animal Genomics

Without resorting to any genetic engineering whatsoever, "it looks like we can increase breeding efficiency in the dairy cow by 50 per cent," is how Graham Plastow, a University of Alberta professor and director of the Alberta Bovine Genomics Program, quantifies the potential for change in one animal.

"We can use the new techniques of genomics selection and marker-assisted selection in breeding better animals, and in so doing we can avoid making transgenic animals — at least into the foreseeable future," adds Max Rothschild, director of the Center for Integrated Animal Genomics at Iowa State University (he's also the U.S. coordinator for the mapping of the pig genome).

The 'genetic wall'

What has happened is that perhaps the greatest anxiety of modern livestock production, the fear that we are about to hit a genetic wall when it comes to improving animals through selective breeding, has been fading if not quite disappearing.

To understand the "genetic wall" fear, you have to appreciate that what we have done in the past 50 years in terms of livestock improvement sounds not like Darwinian evolution, but daily, monthly, yearly revolution.

"The amount of fat in a pig is probably down 60 per cent from what it was 40 years ago — maybe more. Feed efficiencies are up by about 20 to 25 per cent; growth rates show 20 to 25 per cent improvements," says Rothschild. "Also keep in mind reducing fat content and increasing growth rate are antagonistically correlated, so what we see is truly amazing."

But then it turns out it is not just pigs that in the last few decade or so have been turned into super animals. It took 85 days of growth in 1957 for a chicken to reach the same size that it did in 43 days in 2001.

Every year the average production of milk in a cow increases by two per cent. To put this in context, estimates are that the wild cow ancestors produced somewhere between 400 and 500 pounds of milk a year. (There are approximately 2.27 pounds of milk to a litre). That has risen to what Jacques Chesnais, senior geneticist at The Semex Alliance, a Canadian artificial insemination company, says is 24,000 pounds a year. And in 1998, a U.S.-bred cow named Lucy produced 75,275 pounds of milk in a single year.

While some of this improvement has come from improved animal feeds — Rothschild points out that you need to provide cattle with amino acids if you want more muscled meat — much of it also has come from a more "scientific" mating of animals with desirable features.

However, given the forecasts that human population would hit nine billion before the end of the century and UN projections which suggest that between 1983 and 2020 world meat demand will have jumped 2½ times, the revolution in livestock improvement had to continue to prevent humans from going hungry. And how could they do that without genetic engineering if we had effectively gone as far as we could with the breed-animals-with-desirable-traits technique farmers had been using since domestication began about 10,000 years ago?

Selective breeding

The first answer is that we now know we haven't gotten anywhere near to having exhausted the possibilities of improving animals through traditional selective breeding techniques.

Harris Lewin, director of the Institute for Genomic Biology at the University of Illinois who wrote a recent comment piece in Science magazine about what the sequencing of the cow genome means, points out that even after the seemingly intense genetic pressures of modern selective breeding, "cows remain more genetically diverse than humans." Only about five or six per cent of their 22,000 genes give an indication of having been selected by humans.

Moreover, few operations in a body are controlled by just a single gene. So to increase milk production, you might have to move dozens of genes from other creatures into farm animals. And the effect of these genes would not be just on milk production but on many other processes, since we now know that genes produce proteins which participate in the regulation of various operations in the body.

Ergo you have to worry that selecting for one gene or set of genes could unintentionally throw off some other system in an animal's body.

To date, the only way you can be certain you are getting an animal which has a variety of beneficial traits is by using traditional breeding techniques. And even that process remains so complex that, "the biggest boost in production is by mixing genes — that is, through cross-breeding," says Lewin.

But maybe the most profound change for traditional breeding is that new genomic tests now can identify specific gene-based traits in individual animals.

Today, every bull selected by Semex is first tested with a genomic panel that contains 50,000 genetic markers. Based on the results, a "genomic evaluation" is computed for each bull and trait, and only those with superior genomic evaluations are used to sire the next generation.

With this in mind, Australian scientists have estimated that because we can look at a cow's genome and see what traits it is carrying, as opposed to waiting until the animal matures, we can knock five years off the time it takes for a better animal to make it to market.

Selective breeding vs. genetic engineering

So where does this leave us when it comes to the politics of genetically engineering animals? (Note I have purposely removed plants from my conclusions.)

I think the best way to understand where we stand is by imagining scientists had found the reverse set of principles at work in animal genomes. Imagine that we could see we were almost at the end of the line in terms of improving species through classical breeding. Imagine we had learned that if we just added a gene from this animal or a gene from that bacterium, we could ramp milk production up tremendously without any other effect.

If that was the case, I think people who were dead against genetically modified livestock would have to accept that that meant they were either for more people going hungry as world population grew or their opposition to GM food would have to soften.

Only today anti-GM people don't have to change their views, because modern genomics is telling us the old way of creating more productive animals doesn't suck — at least for the foreseeable future — and genetic engineering (at least as far as we understand it) generally does.   It is a truth, which if he knew it, Stuart Brand might find not only didn't suck but was positively psychedelic.