The evaluation of new biotechnologies in livestock production is often done with reductionist, biological criteria such as liveweight gain or feed efficiency, and in isolated, artificial and homogeneous experimental conditions. Too often, the recommendation to apply a specific technology, tested under experimental conditions, assumes that the introduction will not affect other parts in the system. Looking at the animal, farm and farming community as a system provides a more comprehensive framework for assessing the potentials and risks of applying biotechnology in animal nutrition.

A high demand for food and restricted access of many poor farmers to external inputs leaves the production of animal feeds often at a low priority for farmers in many developing countries. Particularly on the better soils and in densely populated areas, the production of cash and food crops is a first priority. As a result, poor quality crop residues such as straw and rangeland grasses are left as major sources of animal feed. Limitations for animal production in developing countries are, therefore, the lack of feed in terms of quality, quantity and seasonal availability.
The lack of quality of ruminant feeds is caused by a high content of lignified, poorly and slowly digestible cell walls in crop residues and mature grasses, usually associated with a low content of nitrogen (N), phosphorus (P) and sulphur (S). Attempts to use biotechnology through pre-fermentation of straws with aerobic fungi have not been successful. The digestibility could be improved in some cases, but almost invariably at the expense of organic matter losses, costly process control and at the risk of mycotoxin formation due to contaminant organisms.
Possible alternatives to this approach could be the pre-treatment of fibrous feeds with enzymes, such as cellulases. Such enzymes are abundantly present in the rumen but they are cell-bound and they tend to be too large to penetrate pores in the cell walls. In other words, they are not very effective in breaking down the cell wall. Fungal cellulases usually do not associate in large complexes, but are secreted freely in the medium. If the transit through the digestive tract is too fast, or the rate of degradation too slow for the successful use of these feeds, fortified with enzymes, then the ensiling of fodder with fungal cellulases together with a stabilization period of weeks to months may be an option.
The conservation of high quality feed by ensiling fodder or hay making also seems an option in situations with periodically sufficiently available high quality material. In temperate regions this has been a very successful way of maintaining quality, but the method is little used in developing countries. Possible technical explanations could be the difficulty to maintain anaerobic conditions. Improvements seem possible by using additives (energy, N, P, S), enzyme producers (microbes) or enzymes (cellulases). However, this way of securing feed quality requires skills which may not always be available at the right moment. Also, the cost of ensilage is often higher than the expected benefits; particularly the cost of airtight cover material and additives may be prohibitive.
By-product feeds for non-ruminants (pigs, poultry) in rural areas are often available in limited quantities, for example, rice bran or oil seed cakes. Quite often they are exported or sold to urban centres. Their protein and phosphorus (P) content seems adequate to support reasonable levels of animal production. However, a large proportion of the P is present in the compound inositol, a form of P not available to monogastric animals (such as pigs and poultry). The addition of the enzyme phytase has proven to be successful in developed countries, and some scope could exist for developing countries as well.

Converter: the animal
A wide array of biotechnological techniques are now available to enhance the breeding, (re-)production and digestion of domesticated animals. The use of antibiotics in animal nutrition, for example, has been applied since the 1960s. More recent possibilities to manipulate mammalian metabolism are or will soon be available through the development of transgenic animals, the use of hormones and beta-antagonists, structural analogues of the hormones adrelin and nor-adrenalin which block their receptors. Their potential in developing countries is less than in developed countries, mainly because the successful application usually requires better feed quality and management.
In addition to antibiotics, a wide variety of feed additives, many of biotechnological origin, are known to modify rumen fermentation. They include components that can reduce methanogenesis, enhance propionic acid production, reduce protein degradation, improve microbial protein synthesis and inhibit protozoa. Among such additives are antibiotics, microbes (probiotics), and specific substrates like oligosaccharides (prebiotics). Probiotics are live, microbial feed supplements that improve the intestinal microbial balance. The term prebiotic refers to substrates selectively stimulating probiotics.
Successful attempts in the use of biotechnology at animal level include the elimination of plant toxins in feed. The transfer of rumen microbes from Indonesian goats to Australian sheep enabled the Australian ruminants to degrade hydroxypyridone (HDP), a metabolite from mimosine, that is commonly present in the tropical legume Leucaena leucocephala.
Another successful application from a technological point of view is the development of the recombinant bovine somatotropin (rBST) hormone. Its effectiveness and safety were confirmed by organizations such as the Food and Drug Administration (FDA) in the USA, but are still disputed by various groups. Consumers tend to be wary of the use of hormones, and widespread application in developing countries is not without danger, particularly because its effective application requires high-quality feed and proper management. If this is not available, the treated animals will be more susceptible to metabolic disorders, to diseases, and may even face the problem of "burn out".

Assessing the impact
The common reductionist, disciplinary approach to improving the world’s food supply by technological solutions, such as the Green Revolution, has yielded impressive results but has been accompanied by unexpected and undesirable side-effects. As a result, particularly in the past two decades an increased awareness has developed of the need to assess a technology from a system’s point of view.
In spite of the semantics on the definitions of what a system approach really is, we distinguish two common principles that serve as a guideline for the discussion in this article:
(1) A system is generally considered to be a set of interrelated parts that together transform inputs into outputs. All parts of a system are interrelated, i.e., an action upon one of the parts affects the functioning of other parts.
(2) Systems can be defined at different levels, at the level of a cell, an organism, an animal, a farm, a region etc. The perceived success of an intervention depends, therefore, on the definition of the system boundaries.
The problems that led to the adoption of the systems approach are still existent and form the basis for a tentative discussion about biotechnology, taking three biotechnology applications as examples: (1) the use of the "mimosine-bug" at rumen level, (2) the use of rBST at the level of the animal metabolism, and (3) the use of enzymes to improve feed intake and digestibility .
An intervention in one part of the system does affect the functioning of other parts. For example, the application of rBST allows an animal to produce more milk, only if enough feed of sufficient quality is available. If this is not the case, the application of rBST for the production of more milk will "suck away" resources from elsewhere in the animal body, for instance from the reproductive organs, with possible negative effects on fertility.
Another often ignored aspect is that the production of high-quality feed usually requires a higher input of fossil energy. What seems to be gained by the application of rBST, can be offset by an increased input of fossil energy. If a system is defined to comprise a farming community (principle 2), and if resources are limited, the application of rBST on one farm may "suck away" resources from other parts of the community. Several studies have shown that a higher sub-system production can even be at the expense of total system output, when resources are limited.
As another example, the mimosine bug may allow a farmer to utilize a so far unused resource such as the Leucaena leucocephala. At the farm level, such a crop may become profitable to produce. However, this means that less land is available for food crop production. A different issue may occur in the use of enzymes to improve feed digestibility or conservation. This may help an animal to survive better a period of feed shortage and reduce the loss of body reserves. The extra feed available for the animal, however, cannot be used by other components in the system, such as earth worms, and other animals with a lower direct commercial value.
In the context outlined above, whether a technology is considered to be beneficial depends entirely on the local conditions, on the definition of the systems boundaries and on the perception of the observer (farmer, consumer, policy maker, etc.). Not many studies are known that address the issues of trade-offs within and between systems. Some deal with the application of technology in general, some focus only on the negative aspects of technology introduction. To date, only a few aim to determine the specific conditions under which the technologies are useful or not.
J.B. Schiere/S. Tamminga

Wageningen Institute of Animal Sciences (WIAS), Marijkeweg 40, 6709 PG Wageningen, the Netherlands. Phone (+31) 317 484594; Fax (+31) 317 485006. E-mail Hans.Schiere@DPS.VH.WAU.NL

Sources
J. Coombs and Y.R. Alston (1995), The international biotechnology directory1995: Products, companies, research organisations. Basingstoke: MacMillan.

A.M. Craig (1995), "Detoxification of Plant and Fungal Toxins by Ruminant Microbiota". In: W. Von Engelhardt, S. Leonhardt-Marek, G. Breves and D. Giesecke (eds.), Ruminant physiology: Digestion, metabolism, growth and reproduction. Stuttgart: Ferdinand Enke Verlag. pp. 271-288.

R.J. Jones and R.G. Megarrity (1983), "Comparative Toxicity Responses of Goats Fed Leucaena leucocephala in Australia and Hawaii". Australian Journal for Agricultural Research, Vol.34, pp. 781.

J.L. Van Os (1993), "Prospects of Feed Additives Produced by Biotechnology". In: D.H. Vuijk, J.J. Dekkers and H.C. van der Plas (eds), Developing agricultural biotechnology in The Netherlands. Wageningen: Pudoc Scientific Publishers. pp. 186-194.