Long-Chain Polyunsaturated Fatty Acids in Neurodevelopment

Introduction

The brain and retina contain large quantities of long-chain (20 and 22 carbon) n -3 and n -6 polyunsaturated fatty acids (LCPUFAs), particularly docosahexaenoic acid (DHA; 22:6 n -3) and arachidonic acid (AA; 20:4 n -6). ^, During the last intrauterine trimester ^, and the first 18 months of postnatal life, DHA and AA accumulate in neural tissue at a high rate, supported by selective placental transfer of DHA and AA from mother to fetus, ^, transfer of preformed DHA and AA from human milk consumption, ^, and synthesis of DHA and AA from the dietary essential fatty acids, α-linolenic acid (18:3 n -3) and linoleic acid (18:2 n -6), respectively. The first studies of infants fed formulas, which, at that time, did not contain DHA or AA, found lower levels of DHA and AA in red blood cell (RBC) and plasma lipids ^, and lower levels of DHA in brain, ^, compared with infants fed human milk. This finding suggested that synthesis of these fatty acids might not meet the needs for DHA and AA in the developing retina and brain and that a dietary source was important. Furthermore, infants born preterm had lower brain DHA than infants born at term, reinforcing the importance of the last intrauterine trimester for postnatal DHA status.

Docosahexaenoic Acid Accumulation and Synthesis

These studies are evidence that formulas containing α-linolenic acid, the 18-carbon precursor of DHA, did not compensate for the DHA accumulation that occurs in RBCs, plasma, and brains of infants fed even small amounts of DHA with functional consequences. ^, In the months after birth, infants fed formulas that contain α-linolenic acid but no DHA have progressively lower RBC DHA, whereas infants fed formulas with DHA have higher RBC DHA. ^, Preterm-born infants are especially vulnerable to receiving a diet without DHA. The third-trimester fetus is estimated to accumulate approximately 40 to 60 mg of DHA per kg each day, whereas Carnielli and colleagues quantified DHA synthesis in 1-month-old infants born preterm and showed that they could synthesize only approximately 13 mg/kg/day. Their synthesis fell dramatically to approximately 3 mg/kg/day by 3 months of age. The rate of endogenous synthesis of LCPUFA from their precursors is further dependent on genetics, balance of fatty acids, and amount of DHA consumed.

Early Clinical Trials in Infants

Randomized studies of DHA and AA supplementation in preterm infants were conducted early in the history of the field to test the hypotheses that (1) erythrocyte DHA and AA were biomarkers for DHA and AA status, (2) lower status could have functional consequences, and (3) neurodevelopmental function could be improved or optimized by DHA and AA supplementation. The functional outcomes chosen for early study were those linked to lower brain DHA in animals: retinal electrophysiology, visual acuity, and various measures of cognition. We now have evidence that the effects of DHA on development include positive effects on blood pressure, response to stress, and autonomic nervous system (ANS) development. The latter is discussed in this chapter. The beneficial effects observed in these realms are most certainly related to increased DHA accumulation in the central nervous system and ANS.

A critical difference between preclinical animal models and human studies was that animals were fed diets lacking α-linolenic acid, the essential fatty acid precursor for DHA, whereas infant formula contained α-linolenic acid. In most animal models, brain DHA was extremely depleted over several generations and replaced by n -6 docosapentaenoic acid (22:5 n -6). ^, In contrast, brain DHA in studies of human term and preterm infants fed formulas without LCPUFA were respectively reduced to only 20% and 50% compared with human milk–fed infants. ^,

In the first studies of preterm infants, researchers found that DHA-supplemented formula (there were no sources of AA available at the time) produced higher visual acuity and more mature retinal physiology. ^, These results led to larger randomized trials in which the growth and neurodevelopment of term and preterm infants fed formulas supplemented with DHA and AA were compared with those fed formulas without DHA and AA. These studies have been reviewed previously elsewhere.

Single-cell oil sources of DHA and AA were added to commercially available term formula in the United States beginning in 2002, after the Food and Drug Administration gave single-cell oils generally recognized as safe (GRAS) status. DHA and AA were added to preterm formulas somewhat later. Before DHA and AA were added to formulas, infants had to rely on synthesis for any additional postnatal DHA accumulation; as noted previously, synthesis is likely inadequate as a source for optimal levels of DHA.

The Docosahexaenoic Acid Intake and Measurement of Neural Development (DIAMOND) trial, the only randomized, controlled dose-response study of infant formula LCPUFA, was initiated in 2002 soon after the addition of DHA to infant formula in the United States became routine. All three DHA-containing formulas included the same amount of AA. The primary outcome of this study was to determine the effects of LCPUFA on visual acuity of term infants at 12 months of age. The study included additional aims to study child development at the individual study sites in Dallas and Kansas City. Results from the trial out to age 9 years are discussed in the section on LCPUFA and the developing human infant.

Maternal Influence on Offspring Docosahexaenoic Acid Status

Intrauterine Docosahexaenoic Acid Transfer

Intrauterine DHA accumulation is variable, as evidenced by (1) a large range of normative values for DHA in cord RBC phospholipids of infants born preterm, (2) a progressive increase in maternal phospholipid DHA throughout the last intrauterine trimester, (3) variable brain DHA content among infants born at the same gestational age, and (4) variable fetal concentration of DHA in adipose tissue.

Variable maternal DHA intake is one well-known reason for this variability, which in turn influences the amount of DHA transferred to the fetus. A review by Haggarty illustrates the influence of maternal DHA intake and gestation duration on fetal DHA accumulation, with most accumulation in fetal adipose tissue at term. However, Kuipers and colleagues question the basis for some of Haggarty’s estimates of adipose tissue DHA concentrations. Although these authors confirm an overall trend to adipose DHA accumulation in the last trimester of gestation, they found extreme variability in individual DHA accumulation at all gestational ages from 25 and 42 weeks. The size of the adipose tissue pool of DHA at birth could be a variable influencing postnatal DHA status.

A large multicountry comparative study included pregnant women from the Netherlands, Hungary, Finland, England, and Ecuador. In the results of this study, researchers reported significant differences among countries in the concentrations of DHA in maternal plasma, most likely due to differences in DHA intake across these locations. As mentioned previously, ^, there is selective transfer of DHA and AA across the placenta and much is known about the specifics of fatty acid transport. Despite the relationship between maternal DHA status and newborn DHA status ^, ^, ; however, only 25% of the variance in DHA status of newborns is predicted by maternal DHA status. Gestational age is also a predictor of cord blood DHA, but no predictor has been found for the majority of variance in cord blood DHA.

Pregnancy itself has been shown to influence the amount of DHA in circulating blood lipids and RBC phospholipids. Al and colleagues observed that phospholipid DHA in maternal plasma and RBCs increases in early pregnancy, regardless of the initial levels of RBC DHA. These data suggest that DHA is mobilized during pregnancy for transfer to the fetus. However, the same group of investigators reported increased maternal 22:5 n -6 relative to DHA, from which they concluded that pregnant women’s systems lag behind with respect to DHA production and transfer to the fetus. Van Houwelingen and colleagues observed similar relative increases in the ratio of 22:5 n -6 to DHA in infants born preterm, further suggesting that an increase in the ratio of 22:5 n -6 to DHA indicates inadequate DHA synthesis or accumulation. Thus, even though it appears that DHA is mobilized during pregnancy, the amount transferred may not provide for optimal fetal development when maternal DHA status is low. The number of prior pregnancies is inversely related to maternal DHA status, suggesting that maternal DHA stores are depleted by repeated pregnancy and/or lactation.

The FADS complex contains the genes that govern the rate-limiting enzymes of the fatty acid metabolic pathways. Studies show the relation between single nucleotide polymorphisms (SNPs) in the FADS gene complex and AA and/or DHA status. ^, Maternal DHA supplementation improves DHA status in all women; however, minor allele carriers of FADS 1rs174553 and FADS 2rs174575 experienced a significant drop in AA status with DHA supplementation.

Human Milk Provides Docosahexaenoic Acid Postnatally

Just as women’s DHA status during pregnancy affects transfer of DHA to the fetus, it influences DHA transfer in human milk. Connor and colleagues and Harris and colleagues first showed that higher maternal DHA intake increases the amount of DHA in breast milk and, in turn, increases the DHA status of their infants. The DHA content of human milk as a percentage of total fatty acids varies within and among populations. The extremes of milk DHA that have been reported are 0.02% in vegans and 2.4% in women from China who consume a diet high in ocean fish. Human milk DHA in most groups tends to fall on the lower side of this range. ^,

Like DHA, the reported average amount of AA in human milk varies across cultural groups, from 0.2% to 1.2%. However, the average AA in the milk of most populations falls within the range of 0.4% to 0.6%, and the average linoleic acid content between 10% to 17% of total fatty acids. ^, Women homozygous for minor alleles of some FADS SNPs appear to have limited transfer of DHA to human milk. Molto-Puigmarti and colleagues linked lower milk DHA content to genetic polymorphisms of two FADS SNPs, FADS 1 rs174561 and FADS 2 rs174575. In a randomized controlled study of DHA supplementation, women assigned to DHA and homozygous for the minor allele FADS 2 rs174575 had lower DHA in milk compared with other genotypes, although their circulating DHA was increased by supplementation. ^,

We have reported that DHA supplementation reduces RBC phospholipid AA in women with minor alleles of FADS 1/ FADS 2, suggesting that alterations in the balance of n -3 and n -6 LCPUFA with DHA intake are greater in minor allele carriers. The implications of this are not known and deserve further study.

Clinical Trials of Docosahexaenoic Acid Supplementation in Pregnancy

Because maternal DHA status during pregnancy and lactation can influence fetal and infant DHA accumulation, studies of the effects of DHA supplementation during pregnancy and lactation on cognitive and other brain-related functions in childhood have been needed. Several trials of supplementation during pregnancy completed since the last edition have measured offspring neurodevelopment. The results are included in the next section.

Lcpufa and the Developing Human Infant

Effect of Dietary Lcpufa on Infant Lcpufa Status

The addition of DHA from fish oil, ^, ^, ^, egg phospholipids, and single-cell oils increases DHA in circulating phospholipids of infants, evidence that these sources increase DHA availability for accumulation in brain and other tissues. As noted previously, infant formulas with α-linolenic acid (but no DHA) do not prevent a gradual postnatal decline in RBC DHA; infants consuming such formulas show low DHA in circulating phospholipids for at least a year. ^,

After birth, there is an apparent physiologic decline in AA in plasma and RBC phospholipids. AA is higher in preterm than in term cord blood, and even after 4 to 6 months of breast-feeding, infants born at term had much lower RBC AA than seen at preterm delivery (8.8% vs. 15.7%). AA in the brain and liver also declines during the last intrauterine trimester. However, breast-fed infants have more AA in RBC and plasma phospholipids than do infants fed formulas without LCPUFAs, and infants fed formulas with n -3 LCPUFAs have even lower phospholipid AA levels than those fed formulas without n -3 LCPUFAs.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here