The fat-1 gene of C. elegans encodes an n-3 fatty-acid desaturase enzyme that converts n-6 to n-3 fatty acids and which is absent in most animals, including mammals2,3. We transferred this fat-1 gene into mice and raised them alongside wild-type mice maintained on an identical diet that was high in n-6 but deficient in n-3 fatty acids. However, the fatty-acid profiles of the two groups turned out to be quite different (Fig. 1). The tissues of wild-type animals contain polyunsaturated fatty acids that are mainly (about 98%) n-6 linoleic acid (designated an 18:2 n-6 fatty acid as it has 18 carbon atoms and 2 double bonds, one at position n-6) and arachidonic acid (AA, 20:4 n-6), with very little n-3 fatty acid (from a dietary source). By contrast, the transgenic animal tissues are rich in n-3 polyunsaturated fatty acids, including linolenic acid (18:3 n-3), eicosa-pentaenoic acid (EPA, 20:5 n-3), docosa-pentaenoic acid (DPA, 22:5 n-3) and docosahexaenoic acid (DHA, 22:6 n-3).

Figure 1: Partial gas chromatograph traces showing the polyunsaturated fatty-acid profiles of total lipid extracted from mouse skeletal muscle.
figure 1

a, b, Traces from lipid from a, a wild-type mouse, and b, a fat-1 transgenic mouse (heterozygote). The expression vector used for microinjection contained the humanized fat-1 sequence (with modification of codon usage) and a chicken β-actin promoter and cytomegalovirus enhancer, which allow high and broad expression of the transgene in mice6,7. Both the wild-type and transgenic mice were 8-week-old females that were fed on the same diet, which was high in n-6 but low in n-3 fatty acids. The lipid profiles show that concentrations of n-6 polyunsaturated acids (18:2 n-6, 20:4 n-6, 22:4 n-6 and 22:5 n-6) are lower and levels of n-3 fatty acids (asterisks) are markedly higher in transgenic (b) than in wild-type (a) muscle. (Homozygotes and heterozygotes have a similar phenotype.)

The concentrations of n-6 linoleic and arachidonic acids in the tissues of the transgenic mice are significantly reduced, indicating that n-6 fatty acids have been converted to n-3, causing the ratio of n-6 to n-3 to drop from 20–50 to almost 1. This n-3 enrichment at the expense of n-6 gives a balanced ratio of n-6 to n-3 and of AA/(EPA+DPA+DHA) in all of the organs and tissues tested (Table 1). Transgenic skeletal muscle contains more EPA than DHA, but DHA is the dominant n-3 fatty acid in other organs.

Table 1 Fatty-acid ratios in WT and fat-1 mice

We have examined the tissue fatty-acid profiles in four generations of transgenic mouse lines (homozygote or heterozygote) and find consistently raised n-3 fatty acids, indicating that the transgene is functionally active in vivo and transmittable. The transgenic mice appear to be normal and healthy.

Efforts have been made to incorporate n-3 fatty acids into the food supply1,4 because of their health benefits and concern over the high n-6:n-3 ratio in Western diets. Our findings suggest a new strategy for producing food that is enriched in n-3 fatty acids from livestock carrying an n-3 desaturase trans-gene. At present, farm animals are fed fishmeal and other marine products, but this is time-consuming and costly, and is limited by the quantity of the source5. Production of n-3 fatty acids by the animals themselves would be a cost-effective and sustainable way of meeting the increasing demand; the ideal n-6:n-3 ratio of about 1 could be achieved by consuming foods containing this ratio and without introducing stringent dietary changes.

Our transgenic mice also offer a model for investigating the biological functions of n-3 fatty acids and the importance of the ratio of n-6:n-3 in disease prevention and treatment. Specific effects of n-3 fatty acids and of the n-6:n-3 ratio can be tested in different organs and tissues — for example, they may alter gene expression or physiological activity during the life cycle. Our mouse lines could be genetically backcrossed with mouse disease models to test the effects of n-3 fatty acids on the pathogenesis and treatment of those diseases.