The gene that settles a sheep's stomach

In future, sheep, cows and goats may be able to thrive on otherwise poisonous plants by virtue of genetically engineered bacteria living in their digestive tracts which break down the poisons.

Keith Gregg and colleagues at the University of New England in Armidale, New South Wales, have engineered a bacterium that destroys a poison which kills substantial numbers of livestock every year in Australia, Africa and Central America. Introduced into the rumen (the first chamber of the stomach in ruminant animals), it should protect sheep against fluoroacetate, a lethal toxin found in 40 types of trees and shrubs.

The research, described in this month's Trends in Biotechnology, could presage a new dimension in the dependence of ruminants on microbes. In contrast to humans, sheep, goats, cows and deer have a rumen containing a rich population of microbes. These not only produce B vitamins but also digest cellulose and other plant substances, releasing energy and materials necessary for the animals' growth. The microbes in some ruminants also have a limited ability to break down certain poisons the animals may ingest while grazing.

Some 20 years ago, Australian scientists began to explore the idea of extending this protective role of rumen microbes. Their first success came in 1984, when they devised a method of preventing livestock being poisoned by a legume known as Leucaenia leucephala, which had recently been introduced into their country. It contained mimosine, an amino acid similar to those that form the building blocks of two proteins. When sheep and goats ate the plant, their digestive juices converted the relatively harmless mimosine into a chemical that damaged the thyroid gland.

Indonesian goats, however, munched Leucaenia with impunity, and tests on their rumen microbes showed why. One of the organisms was the bacterium Synergistes jonesii, which broke down and thus rendered safe the thyroid poison. When researchers transferred some rumen contents from the Indonesian goats into native Australian sheep and goats, the microbes thrived and these animals, too, became able to live on Laecaenia. More recent observations have revealed that the tolerance to mimosine spreads throughout sheep flocks as the rumen microbes pass from one animal to another and become established in their rumens.

However, not all poisons found in plants can be rendered safe by rumen microbes that occur naturally in one species of ruminant or another. For this reason, and because many of these poisons at least impair the growth and productivity of livestock, Keith Gregg and his collaborators decided to investigate the feasibility of using genetic engineering to make "detoxifying" bacteria. Despite the great variety of microbes in the rumen, there are astronomically more diverse populations in the soil and elsewhere in the environment. The Australian scientists reasoned that if they could isolate a non-ruminant microbe capable of attacking a particular plant poison, they might be able to introduce that ability into a ruminant microbe.

A few months ago, they announced their initial success. They had found a bacterium, living in the soil, which produced an enzyme that degraded fluoroacetate, a plant poison lethal to sheep and cattle even in low doses. They then transferred the gene responsible for making the enzyme to a rumen bacterium, Botyrivibrio fibrosolvens, in the laboratory. It, too, was then able to attack fluoroacetate.

Gregg and his colleagues have now demonstrated that the genetically altered microbe, introduced into the rumen of sheep, persists and forms a new but stable part of the microbial population. This reflects the fact that B. fibrosolvens is itself a normal inhabitant of the rumen rather than a totally foreign organism (which would probably be rapidly eliminated). The scientists are now assessing the capacity of the modified B. fibrosolvens to protect the animals against fluoroacetate poisoning. Early indications are that the genetically altered microbe does become established in sufficient numbers to reduce the likelihood of this happening.

As the Australian team concedes, more work will be required before their initial experimental success can be translated into everyday farming practice. One concern is that toxin-resistant animals will damage pastures because they can consume much more vegetation than before.

Another anxiety is that the newly engineered B. fibrosolvens might pass into other species from the sheep into which it was first introduced. Australian farmers use fluoroacetate as a pesticide to control rabbits and foxes. Could these animals become resistant to the pesticide if they were colonised by a bacterium able to break it down? The chances seem remote, simply because rabbits and foxes do not have a rumen. But could B. fibrosolvens invade the lower end of their digestive tracts, with their large microbial populations? Although theoretically possible, this would probably have little or no effect on the fluoroacetate, which would be absorbed higher up the tract and thus retain its poisonous power.

These are typical of the questions to be confronted as we enter an era rich with possibilities for applying genetic manipulation in agriculture.

BERNARD DIXON