indicates that the amount of long-chain n-3 polyunsaturated fatty acids required in the diet is less than the requirement for a-linolenic acid alone (reviewed by Innis, 1991; Greiner et al., 1997; Su et al. 1999a).
The patterns of brain development in rhesus monkeys show a growth spurt in the last trimester of fetal development; the brain of a newborn weighs nearly 70% as much as that of an adult (Venkatraman et al., 1992). The presence of n-3 fatty acids, and particularly the long chain docosahexaenoic and eicosapentaenoic acids, in the diet of pregnant females (Greiner et al., 1996), is therefore critical for sustaining normal brain development. The work of Conner and associates (Connor et al., 1984; Neuringer et al., 1984; Neuringer et al., 1986; Lin et al., 1990; Reisbick et al., 1991; Lin et al., 1994; Reisbick et al., 1994) has demonstrated that a deficiency of n-3 fatty acids in diets of rhesus monkeys can result in demonstrable abnormalities in brain and retinal function. In a series of studies, two diets were fed to pregnant and lactating females, one with about 1% of dietary MEas a-linolenic acid (control) and one with less than 0.1% of dietary ME as a-linolenic acid (deficient). The infants raised on the n-3 fatty acid-deficient diet showed reduced visual acuity by the age of 4 weeks (Neuringer et al., 1984). Deficient monkeys also showed a tendency toward increased intake of water and other fluids (Reisbick et al., 1991) and more stereotypical behavior than the control monkeys (Reisbick et al., 1994).
Observed biochemical changes included reduced docosahexaenoic acid levels in the phospholipids of brain and retina and replacement with long-chain n-6 fatty acids, principally 22:5 n-6 (Lin et al., 1990). Replenishment of the deficient diet with long-chain n-3 fatty acids from fish oil for 14 months resulted in complete reversal of the patterns of n-6 and n-3 fatty acids in brain phospholipids. The remodeling of brain phospholipids appeared to occur normally without significant loss of n-3 fatty acids (Innis, 1991). Furthermore, observations of Kanazawa et al. (1991) in cynomolgus monkeys and in Japanese macaques showed that the ability of the brain tissue to convert a-linolenic acid into docosahexaenoic acid is age-dependent, being essentially zero in newborn primates and increasing maximally in young adults. Nevertheless, apparently adequate amounts of docosahexaenoic acid are deposited in the brains of monkeys fed diets in which essentially the only n-3 fatty acid is a-linolenic acid (Lin et al. 1990). That indicates that other tissues, predominantly the liver, have the desaturases and elongases needed for conversion of a-linolenic acid to the docosahexaenoic acid required for lipid deposition in the gray matter of developing brain and in the retina. In the case of monkeys, in which much of brain development occurs in utero, the transfer of n-3 fatty acids across the placenta into the fetus supplies a major portion of the requirements for early life (Innis, 1991).
The amounts of n-3 fatty acids that must be consumed for adequate deposition of docosahexaenoic acid in the developing nonhuman primate brain can be estimated by extrapolation (Kanazawa et al., 1995) from human data (Clandinin et al., 1980a, 1980b). However, it is difficult to define an exact dietary requirement because much of the needed n-3 fatty acid will be derived in utero from the mother, and the efficiency of this transfer process is unknown (Greiner et al., 1996). Subsequent studies by the same group (Greiner et al., 1997) gave the estimate of the requirement as 0.45% of ME as a-linolenic acid or 0.30% of ME as docosahexenoic acid (22:6 n-3) in fetal baboons for normal brain development. Other data show that diets of rhesus monkey mothers with 1% of ME as a-linolenic acid were adequate to maintain normal fetal brain development, as were infant diets that contained about 2% of ME as a-linolenic acid (Lin et al., 1990; Neuringer et al., 1984). Furthermore, the data derived from studies of African green monkeys showed that diets for mothers and infants containing about 0.25% of ME as long-chain n-3 fatty acids (eicosapentaenoic and docosahexaenoic acids), with another 0.2% of a-linolenic acid, resulted in normal development (Wolfe et al., 1993). In the mother’s milk, about 0.6% of ME was found as 22:5 n-3 and 22:6 n-3 fatty acids, and 0.2% as a-linolenic acid. Therefore, in the absence of dose-range studies, those data form the basis of the minimal amounts of n-3 fatty acids recommended for nonhuman primate diets. It is recommended that 0.5% (by weight) of dietary dry matter (about 1% of ME) be present as n-3 fatty acids to support normal development and maintenance of the brain and nervous system.
Research showing that n-6 fatty acids are dietary essentials for nonhuman primates was published by Greenberg and Moon (1961), who documented changes in blood fatty acids in rhesus monkeys fed a linoleic acid-deficient diet. Subsequently, Greenberg (1970) showed that diets containing corn oil at about 2% by weight (about 4% of ME) prevented the deficiency. Portman et al. (1959, 1961) demonstrated linoleic acid deficiency in cebus monkeys and described the changes in physical appearance of the animals and many biochemical changes in fatty acid composition and concentration in their tissues. Substantial pathophysiologic changes in cebus monkeys fed a fat-free diet for 19 months were limited to scaly skin, hyperplastic bone marrow, erythrophagocytosis by the reticuloendothelial system, and undersized gonads. The link between requirements for polyunsaturated fatty acids and vitamin E was studied in rhesus monkeys made vitamin E-deficient (Fitch et al., 1961, 1963). It was shown that vitamin E deficiency could be induced by using diets either unsupplemented