April 1992: (11)21-29
Essential Fatty Acids: The Importance of 12-3 Fatty Acids in the Retina and Brain William E. Connor, M.D., Martha Neuringer, Ph.D., and Sydney Reisbick, Ph.D.
Humans need fat in the diet in order to survive, grow, and prosper. Over 60 years ago, Burr and Burr' demonstrated that fat was an essential component of the diet. Rats reared on a fat-free diet failed to grow and reproduce and also developed renal disease, fatty liver, dermatitis, and necrosis of the tail. Later studies identified the deficient components of the diet as polyunsaturated fatty acids with two or more double bonds.* Until recently, linoleic acid, an n-6 fatty acid* commonly found in vegetable oils and many other foods, was deemed the primary essential fatty acid, together with its derivative n-6 fatty acids, of which arachidonic acid (20:4) was the most important. However, it now appears that two classes of fatty acids are both essential for health (Figure 1). The second series of highly polyunsaturated fatty acids includes the n-3 fatty acids, a-linolenic acid (18:3) and its longerchained, more polyunsaturated derivative, docosahexaenoic acid (22:6). n-3 fatty acids are also contained in a wide variety of natural foodstuffs, including some but not all liquid vegetable oils, particularly soybean and rapeseed oils, and green leafy vegetables. The higher derivative, docosahexaenoic acid, is found only in the plants of the sea, phytoplankton, or, as one ascends the food chain, in shellfish, fish, and sea mammals (Figure 2). Docosahexaenoic acid is especially prominent in brain, retina, and spermatozoa'but is also found in phospholipid membranes throughout the body, including the myocardium. Dr. Connor is a Professor of Medicine and Section Head, Dr. Neuringer is a Research Associate Professor, and Dr. Reisbick is a Research Assistant Professor at the Section of Clinical Nutrition, Department of Medicine, L-465, Oregan Health Sciences University, 3181 South West Sam Jackson Park Road, Portland, OR 97201, USA. 'In fatty acid nomenclature, the terms n-6 or n-3 may be used interchangeably with omega-6 or omega-3.
Nutrition Reviews, Vol. 50, No. 4
Aside from essential amino acids, n-3 fatty acids constitute the largest component of the cerebral cortex and retina that can be obtained only from the diet. The body cannot synthesize either the n-3 or the n-6 structure, as shown in Figure 1. Once the basic n-3 structure is consumed in the form of linolenic acid (18:3), the body can synthesize the longer-chain and highly polyunsaturated fatty acid docosahexaenoic acid (22:6) (Figure 3). Additional double bonds are inserted through the action of desaturase enzymes, and the 18-carbon chain of the precursor, linolenic acid, is elongated to 20 and then to 22 carbons. Similarly, in the n-6 series, linoleic acid (18:2) is converted to arachidonic acid (20:4), the predominant n-6 fatty acid in most tissues, and also to 22:4 and 225. Longer-chained fatty acids of the n-6 series (225) have become of interest in the n-3 dietary deficiency syndrome, which will be described subsequently. n-3 and n-6 fatty acids compete for the desaturase enzymes; high levels of linoleic acid inhibit the conversion of linolenic acid in the n-3 series to the next form, 18:4. Thus, a high ratio of linoleic acid to linolenic acid is likely to produce the greatest depletion of longer-chain n-3 fatty acids such as docosahexaenoic acid (22:6). It is this latter fatty acid, with its extreme degree of unsaturation, that is present in particularly high concentrations in the outer segment membranes of the rods and cones of the retina and in the synaptic membranes of the brain. The essentiality of n-3 fatty acids has been fully demonstrated in fish.3 There is also evidence of their importance in retinal and brain function in rats4" and especially in the nonhuman primate, the rhesus monkey.6 A diet deficient in n-3 fatty acids leads to a triad of signs in the rhesus monkey: visual impairment, abnormalities of the electroretinogram, and p o l y d i p ~ i aProfound .~~ biochemical changes in the fatty-acid composition of the membranes of the retina, brain, and other organs accompany these other disturbance^.^*'^ Low concentrations of n-3 fatty acids, especially 22:6, occur at birth in the
21
FATTY ACID NOMENCLATURE &Y
DIETARY SOURCES
18:2 (w6)
STRUCTURE
FATTY ACID
18:3-18:4-20:4
wRCOOH
w3
Eicoaovantoenoic Acid ICZ0:S w31
ru6
Linolalc Acid (C18:2w6)
w9
Olalc Acid
H,C
Marine Oils, Fish
6
Vegetable Oils; Animal Fats
Figure 1. The three major series of unsaturated fatty acids and the structure of the terminal end of the fatty acid chain for each series.
plasma, red blood cells, and neural tissues of rhesus infants born from mothers fed an n-3-deficient diet.6*7.9These concentrations become even lower as the deficient diet is continued postnatally. Visual impairment can be demonstrated by four weeks of age, and abnormalities of the electroretinogram occur by three months of age.639Polydipsia develops later in life.* In contrast to the more obvious clinical stigmata of n-6 fatty acid deficiency, the findings of n-3 deficiency are far more subtle, as might be expected for a fatty acid class required for neural membranes. n-3 fatty acid-deficient rhesus monkeys appear grossly normal, even after long-term depletion. The fur and skin do not show the conspicuous dermatitis found in n-6 fatty acid deficiency, nor does fatty liver occur. Table 1 illustrates the characteristic differences, clinical and biochemical, between essential fatty acid deficiencies of the n-6 and n-3 series. As noted above, n-6 fatty acid deficiency has been well-known and described in animals and humans.*"' Until recently, n-3 fatty acid deficiency, because of its less conspicuous symptomatology , has not been well-categorized and has at times oc-
Phytoplankton Algae
I
Terrestrial (18:3)
Plants (leaves, some nuts & seeds)
Krill, shellfish
+I
Fish
tt
Humans
Humans Figure 2. The dietary sources of n-3 fatty acids according to the food chains in the waters and on land. Note that the only rich source of the very polyunsaturated n-3 fatty acids is from phytoplankton and the sea life that directly or indirectly feeds on phytoplankton. 22
-=
Vegetable Oils 9
(20:5,22:6)
--205-225
' RCOOH
""
Marine
y~ 3 Serieq
w 6 Series 18:2-18:3-20:3-=-=4--
Figure 3. The major pathways for synthesis of longerchain n-3 and n-6 fatty acids from linolenic (18:3) and linoleic (18:2) acids. As indicated by the dotted line, high levels of linoleic acid block the desaturation of linolenic acid. In tissues, docosahexaenoic acid is the major n-3 fatty acid and arachidonic acid (20:4) is the major n-6 fatty acid.
curred in combination with n-6 fatty acid deficiency. '*-I6 Methods of Study
Rhesus monkeys were studied after a combination of maternal and postnatal deprivation of dietary n-3 fatty acids. Throughout pregnancy, adult females were given a semipurified diet with safflower oil as the only fat source.6 Their infants were fed a liquid version of the same diet from birth onwards. Safflower oil is particularly low in n-3 fatty acids, with less than 0.3% of the total fatty acids as linolenic acid (18:3, n-3). It is especially high in n-6 fatty acids, with a ratio of n-6 to n-3 fatty acids of about 250: 1. Mothers and infants in the control group received diets with the same composition except that the fat source was soybean oil, which provides 7.7% of total fatty acids as linolenic acid and which has an n-6 to n-3 ratio of 7: 1. The more highly polyunsaturated fatty acids of the n-3 series, such as eicosapentaenoic and docosahexaenoic, were not present in either of these diets. To test the reversibility of the deficiency, five of the deficient offspring were repleted with very longchain and highly polyunsaturated n-3 fatty acids from fish oil beginning at 10-24 months of age." In the repletion diet, fish oil replaced 80% of the safflower oil, the remaining safflower oil providing ample n-6 linoleic acid (4.5% of energy) and the fish oil supplying large amounts of the n-3 fatty acids, docosahexaenoic acid and eicosapentaenoic acid (205). Plasma, erythrocyte, and tissue samples, including whole retina and cerebral cortex, were analyzed for fatty acid composition by capillary column gasliquid chromatography after separation of total phospholipids or individual phospholipid classes by thin-layer chromatography. l7 In addition, in order to follow the time course of the biochemical Nutrition Reviews, Vol. 50, No. 4
Table 1. Differing Characteristics of n-3 and n-6 Essential Fatty Acid Deficiencies
Clinical Features
Biochemical Markers
n-3
n-6
Normal skin, growth and reproduction Reduced learning Abnormal electroretinogram Impaired vision Polydipsia Decreased 18:3 n-3 and 22:6 n-3 Increased 22:4 n-6 and 2 2 5 n-6 Increased 20:3 n-9 (only if n-6 also low)
Growth retardation Skin lesions Reproductive failure Fatty liver Polydipsia Decreased 18:2 n-6 and 20:4 n-6 Increased 20:3 n-9 (only if n-3 also low)
changes, serial biopsies of the frontal cortical gray matter were obtained for fatty acid analysis after craniotomy . Visual acuity was measured by the preferentiallooking method as described previously.6 The specific physiological effects of n-3 fatty acid deficiency on the function of the retina were examined by the e l e c t r ~ r e t i n o g r a m . ~ ~ ' ~ ResuIts Faffy Acid Composition of Plasma and Tissues Dietary deprivation of n-3 fatty acids resulted in low plasma and red blood cell concentrations of all n-3 fatty acids (Figure 4). The plasma concentrations of these fatty acids were low even at birth in the deficient monkeys, and their mothers had plasma n-3 fatty acid concentrations less than 50% of controls. In the deficient animals, the n-3 fatty acids were barely detectable in the phospholipids of plasma Adult females u)
and erythrocytes by 24 weeks of age, in contrast to the concentrations detected in the control animals. The brain and retina also had very low levels of n-3 fatty acids, particularly docosahexaenoic acid. In deficient animals at or near birth, docosahexaenoic acid concentrations in phosphatidylethanolamine were reduced by 50% in the retina and by 75% in the cerebral cortex compared to the control values (Figure 5). The proportion of docosahexaenoic acid in both tissues doubled between birth and two years of age in control monkeys but failed to increase in the deficient group, so that by two years docosahexaenoic acid concentrations in deficient monkeys were reduced to 15 to 20% of control values. It is notable that n-6 fatty acids, 2 2 5 particularly, compensatorily increased in response to the reduced docosahexaenoic acid, so that polyunsaturation of the phospholipid membranes was preserved as much as possible. This n-6 fatty acid comprised less than 1% of total fatty acids in the tissue Infants
1
2
0
a
c
c
Q,
$
n
Figure 4. Total n-3 fatty acids (weight % of total fatty acids) in plasma phospholipids of adult female rhesus monkeys and their offspring fed control (soybean oil) and n-3 acid-deficient (safflower oil) diets. Red blood cells date shown in f w r e 6 . Nutrition Reviews, Vol. 50, No. 4
23
Cerebral Cortex
/
Retina
/
/
/
Control/ /
/
/
/
/
/
/
fl Deficient
+ At Birth
Deficient
I
At 2 Yrs.
Figure 5. The concentrations of docosahexaenoic acid in the phosphatidylethanolamine of the cerebral cortex and retina of control animals and deficient animals at birth and at 2 years. Note that at birth both retina and cerebral
At Birth
At 2 Yrs.
cortex had much higher levels of docosahexaenoic acid in control than in deficient animals, and these disparities increased greatly by 2 years of age.
phospholipids of normal animals but in two-yearold deficient animals rose to approximately 20% in the phosphatidylethanolamineof the cerebral cortex and to nearly 30% in the retina. After dietary repletion with fish oil, these changes in fatty acid composition were rapidly reversed, as indicated by changes in plasma phospholipids, red blood cells, and biopsy specimens of frontal cortex (Figures 6 and 7)." The changes in the fatty acid composition of the frontal cortex also occurred rapidly, as early as one week after fish oil supplementation. By 24 to 28 weeks, the docosahexaenoic acid in phosphatidylethanolamine increased from 4.2 to 29.3% of total fatty acids (Figure 7), compared to the 22.3% in soybean oil-fed control animals. It is particularly interesting that 2 0 5 n-3 and 2 2 5 n-3 both increased from 0% to approximately 3% in the cerebral cortex while at the same time, the concentrations of 2 2 5 n-6 and other longer-chain n-6 fatty acids decreased. Visual Function and the Electroretinogram The visual acuity of n-3 fatty acid-deficient infant monkeys was reduced by at eight and l 2 weeks of age (Figure 8 ) ~as Previously Furthermore, deficient monkeys developed a number of abnormalities in the electroretinogram (ERG).9*'8The 24
Figure 6. The fatty acids of plasma and erythrocyte phospholipids in n-3 fatty acid-deficient monkeys and mankeys repleted with fish oil. Note the reciprocal relationships between the n-3 and n-6 fatty acids: n-3 fatty acids increase after repletion with the fish oil diet whereas n-6 fatty acids are diminished. Nutrition Reviews, Vol. 50, No. 4
40-
(I)
4 0
30-
4
3. L
m
-m LL
5
I1
20-
I.+-
0
c
C
Q,
2
Q,
a
10-
I
P -0 Qo 0 'h0, @%*
0 1
3
6
8
Weeks on Repletion Diet
/ '
Autopsy
Figure 7. The time course of fatty acid changes in phosphatidylethanolamine of the cerebral cortex of five juvenile monkeys fed fish oil for 43-129 weeks. As docosahexaenoic acid (DHA) increased, 2 2 5 n-6 decreased reciprocally. Levels of DHA and 2 2 3 n-6 in phosphatidylethanolamine of the frontal cortex of monkeys fed
control (soybean oil) and deficient diets from a previous study' are given for comparison. DHA and 225 n-6 in control monkeys were 22.3 2 0.3% and 1.4 2 0.3% of total fatty acids, respectively. DHA and 2 2 5 n-6 in the deficient monkeys were 3.8 ? 0.4% and 18.3 2 2.5%, respectively.
timing of the ERG response was altered, with significant delays in the peak latency (time to the B-wave peak) of both cone and rod responses (Figure 9). In contrast to previous studies in n-3 fatty aciddeficient rats, differences in response amplitudes were not detected at seven to 24 months of age.
However, more recent recordings of younger infants at three to four months of age have demonstrated clear differences in the A-wave amplitudes of both rod and cone responses. The reason for the transient nature of this effect is unknown. Deficient animals also showed a specific abnormality in the rate of recovery of the ERG response after an initial bright flash. This effect was present at three months but increased in magnitude with age. With an interval of 3.2 seconds between flashes, response amplitude in the deficient animals was reduced nearly twice as much as in controls, relative to the maximal amplitude seen in response to the first flash or to flashes presented at intervals of 20 seconds or more (Figure 10). Thus, recovery of the capacity to generate a full ERG response was significantly slowed. In deficient monkeys repleted with fish oil, ERGS were recorded at three, six, and nine months after the beginning of the repletion phase. Despite the increase in n-3 fatty acid levels in tissues, no improvement was seen in either peak latencies or in the ERG recovery function."
l,ol,o 20130
4 weeks
8 weeks
12 weeks
20150 20160 20175
2
.-
20/100 201120 4. 201150 201200
-h
:
(I)
201300
201600
@o /> ' 0"5 0 ,
@ / "
/ '
Figure 8. Visual acuity thresholds (mean t SEM) as determined by the preferential-looking method for control and n-3 fatty acid-deficient infant monkeys. Thresholds are expressed in cycles per degree of visual angle and in the equivalent Snellen values. The p values for statistical significance were determined by Student's t-test. Nutrition Reviews, Vol. 50, No. 4
Polydipsia in Deficient Monkeys Cage behavior in n-3 fatty acid-deficient and control monkeys was monitored by videotaping. It was noticed that the deficient monkeys visited the water spouts of their cages much more often than did con25
trol monkeys, suggesting the possibility of greater water intake. A study of 24-hours water intakes was then camed out to confirm and quantify this difference.8 Figure 11 depicts the mean water intake in milliliters per kilogram of body weight over 24 hours (A) and the combined excretion of urine and feces over the same period (B). Water intake was more than double for deficient compared to control monkeys (264 vs. 70 mVkg) as was excretion (268 vs. 121 g/kg). The excretion was necessarily a combined output of feces and water because there was mixing of the soft stools and urine. These studies were repeated on several different occasions with similar results in monkeys varying in age from 18 months to young adults. The mechanism of the polydipsia is not yet understood, but several possible explanations can be eliminated. The effect is probably not caused by an osmotic imbalance. Dietary electrolytes and serum electrolytes were similar in both control and deficient animals. Fasting glucose levels were normal, indicating that polydipsia did not result from fluid loss as a result of diabetes. The blood urea nitrogen and creatinine levels were also equal and normal, with no indication of renal insufficiency in either group. Polydipsia does occur in dietary n-6 fatty acid deficiency because of increased skin permeability and the resulting loss of water by evaporation. However, the amount of n-6 fatty acids in the diet was actually 50% higher in deficient than in control
1.oo
a, .80 U
-3
c,
. I
f .60
a
T
a,
.? c,
-aa,
a
.40
.20
0
C
D
Figure 10. Relative B-wave amplitude (mean 5 SEM) of the ERG elicited by flashes at 3.2-seconds intervals, as a percent of the maximal amplitude produced at intervals of 20 seconds or more. Relative amplitude is significantly reduced in the deficient group (p < 0.01).
monkeys. In addition, the skin and fur of the n-3 fatty acid-deficient animals appeared visually and histologically normal up to two years of age. Investigations are now in progress to describe possible hormonal and prostaglandin mechanisms of the polydipsia.
Discussion
1 1
tJ
\
Deficient
I
I
I
0
100
200
I
300 msec
Figure 9. Representative ERG waveforms from control and n-3 fatty acid-deficient monkeys under conditions selectively stimulating the rod system. The vertical arrow indicates time of flash. B-wave peak latencies were delayed in the deficient group; vertical lines have been drawn through the peaks to aid comparison. 26
This study has shown that dietary n-3 fatty acid deficiency leads to severe and progressive depletion of n-3 fatty acids from the plasma and from all tissues analyzed, including red blood cells, liver, skin, fat, cerebral cortex, and retina. In particular, the very long-chain n-3 fatty acid, docosahexaenoic (22:6), was selectively depleted from neural and retinal phospholipids and was replaced by n-6 fatty acids, in particular 2 2 5 . Associated with these biochemical changes were a significant impairment in the development of visual acuity, abnormalities in the electroretinogram, and an increase in fluid intake and excretion. In the initial experiments, linolenic acid (18:3) was the only dietary source of n-3 fatty acids. In the control animals, docosahexaenoic acid was synthesized by successive and multiple steps of desaturation and elongation from linolenic acid. It then selectively accumulated in brain and retinal phosphoNutrition Reviews, Vol. 50, No. 4
40
LA
Deficient Diet
H
(n=7)
Control Diet
0-4
(n=7)
if4
acid in phospholipids of the retina, particularly in the outer segment membranes of the photoreceptor cells. This fatty acid is thought to be responsible for the special biophysical properties of the outer segment membranes, which contain the visual pigment. Depletion of docosahexaenoic acid from these membranes would be expected to alter their physical properties and, therefore, the efficiency of the visual process. It is also possible that docosahexaenoic acid has a more specific biochemical function in the retina. In addition to the retina, docosahexaenoic acid is also a major fatty acid of brain gray matter, especially synaptic membranes, which, like photoreceptor membranes, are excitable and highly fluid. Thus, the change in membrane phospholipid composition produced by n-3 fatty acid deficiency might alter the transmission of information through the brain’s visual pathways as well as affecting the photoreceptive process in the retina. A few cases of n-3 fatty acid deficiency have been described in humans. Holman et al.” recently reported a clinical case of peripheral neuropathy and blurred vision in a child receiving total parenteral nutrition. The symptoms were attributed to n-3 fatty acid deficiency because safflower oil, as in our study, was the sole fat source. Replacement of the safflower oil emulsion with linolenic acid-rich soybean oil was associated with recovery. However, it is difficult to be certain that the symptoms were due to n-3 fatty acid deficiency rather than to some other metabolic disturbance induced by long-term parenteral nutrition. More recently Bjerve et al. I 3 * l 4 described ten cases of presumed linolenic acid deficiency occurring in a number of nursing-home patients in Norway, some semicomatose, who had been fed over several years by gastric tubes. The diet of these patients was based on a commercially available powder supplement containing small amounts of corn oil ( 1 . 3 g / l O O g) and mixed with skim milk. The authors reported that the patients developed very low plasma levels of n-3 fatty acids and a scaly dermatitis. Corn oil has an especiaily low content of linolenic acid (0.3%) but, if supplied in quantity, would have ample n-6 fatty acid (i.e., linoleic acid or 18:2) to meet the needs for n-6 essential fatty acids. However, the diet furnished only 0.5% of energy as 18:2 n-6, the plasma 18:2 n-6 was low (99 to 15% of total plasma fatty acids vs. 35% in healthy control subjects’s-’5), and levels of eicosatrienoic acid (20:3 n-9) were substantially elevated. These observations suggest that the patients were also deficient in n-6 fatty acid and that this was the cause of their skin lesions, since dermatitis has not been found in animals having pure n-3 fatty acid deficiency. It is likely that these patients had a com-
’*
s c
8
c5
w
13
10
l A
ifi
0, v
5s
22-7
t
,
0
20
18
e
8
,
,
10 AM
,
l
12
1
,
1
14
,
,
16
,
,
,
,
20
18
PM
1
22-7 AvelHour in Dark
Hours of the D a y
Figure 11. The mean hourly water intake (mL/kg) over a 24-hours period and the mean hourly output of combined urine and feces (g/kg) over 24 hours. The AveIHour in dark refers to the measurement taken at the end of the 10-hours dark period divided by 10.
lipids, particularly in the phosphatidylethanolamine and phosphatidyl-serine fractions.’ The conversion of linolenic acid to docosahexaenoic acid may occur in the liver, in the brain or retina, or in the placenta during fetal development. In the deficient monkeys, the dietary supply of linolenic acid was insufficient to support the synthesis of adequate levels of docosahexaenoic acid. High levels of dietary linoleic acid also may have suppressed synthesis of docosahexaenoic acid from the available linolenic acid. l9 However, during intrauterine life our experimental monkeys received some docosahexaenoic acid directly from their mothers via the placenta. Although deficient mothers had been receiving the n-3 fatty acid-deficient diet for at least two months before conception, their stores of docosahexaenoic acid were probably depleted very slowly. The plasma of newborn deficient infants and their mothers revealed the presence of some docosahexaenoic acid, the levels being consistently higher in the infants’ plasma than in their mothers’, a process termed biomagnification.20 However, plasma concentrations of docosahexaenoic acid were much lower in the deficient infants than in control infants. Once the infants were removed from any maternal sources of n-3 fatty acids, their plasma levels fell rapidly. They were separated from their mothers at birth and, therefore, did not have access to maternal milk, a natural source of docosahexaenoic acid. Docosahexaenoic acid is the predominant fatty Nutrition Reviews, Vol. 50, No. 4
27
bined essential fatty acid deficiency which included both n-6 and n-3 fatty acid deficiencies.l6 The findings of our study provide the first experimental evidence for a dietary requirement for n-3 fatty acids in primates and add to the extensive studies in rats emphasizing the essentially of at least one of the n-3 fatty acids, linolenic acid.4 n-3 fatty acids are essential nutrients for retinal and brain function, especially during fetal and postnatal development. Both biochemical and functional stigmata are characteristic of the n-3 fatty acid deficient state. A recent review details the experimental background of n-3 fatty acid deficiency in different species of a n i m a ~ s . ~ Summary and Implications
In summary, n-3 fatty acid deficiency is characterized in rhesus monkeys by a triad of problems: impaired vision, abnormal electroretinograms, and polydipsia. Concomitant with the visual and ERG defects is the disturbed biochemistry of the retina. Docosahexaenoic acid (22:6 n-3) in the retinal phospholipids is decreased, whereas the n-6 fatty acids, particularly 2 2 5 , are increased. Similar biochemical abnormalities occur in the cerebral cortex. In the absence of n-3 fatty acids, the body produces the most similar polyunsaturated fatty acids possible from precursors, so that the amount of total polyunsaturation of the tissue membranes is largely maintained. The cause of the polydipsia may also be related to the same altered brain biochemistry, but does not seem to be mediated by renal disease or by abnormalities in the posterior pituitary hormones or in osmotic regulation. The diagnosis of n-3 fatty acid deficiency can be made at birth and later in life by determination of fatty acid concentrations in plasma and red cells of cord and peripheral venous blood. In deficiency, there will be low levels of the n-3 fatty acids 18:3 and 22:6 and high levels of the n-6 fatty acid 2 2 5 , which is normally very low in both plasma and tissues of primates. The deficiency state can be prevented at any age by the provision of adequate amounts of n-3 fatty acids in the diet. It i s not yet known whether both linolenic and docosahexaenoic acids should be present in the diet for optimal development; both are found in human milk. Once overt deficiency develops past infancy, biochemical correction of tissue levels may not necessarily restore all the functional defects. In our studies, the electroretinogram continued to be abnormal. Adequate provision of n-3 fatty acids is critical during three periods of life: pregnancy, lactation, and infancy. During pregnancy, these fatty acids are transferred across the placenta to the fetus. 19*20
28
During lactation, the diet of nursing mothers affects the n-3 content of their milk.21 If a formula is fed, adequate and balanced sources of both n-3 and n-6 fatty acids are required in the formula to provide these fatty acids postnatally. Moreover, in infancy, n-3 fatty acids continue to accumulate in both the brain and retina, and an adequate dietary source is important. In all probability, requirements for n-3 fatty acids continue during childhood and even in adult life. Elderly and other patients fed special diets may be at risk for n-3 fatty acid deficiency. We suggest that the dietary requirements for n-3 fatty acids in infancy are from 0.5 to 1.0% of total energy and ideally might include both 18:3 and 22:6, as does human milk. The suggested requirement for n-6 fatty acids should be from 2 to 6% of total energy and should include arachidonic acid, 20:4, as well as linoleic acid. The ratio of n-6 to n-3 fatty acids is important and should range from about four to ten.22Too high a ratio, as is present in the n-3 fatty acid-deficient safflower oil, might further accentuate the deficient state. Some infant formulas, particularly the dry powdered formulas containing corn and coconut oils, provide marginal amounts of n-3 fatty acids, which are as low as 0.1 to 0.2% of total energy. The ratio of n-6 to n-3 fatty acids, at 75: 1 , is also high (Connor WE, Van Winkle S, unpublished data). Recent research in preterm human infants has shown, as in monkeys, vision and electroretinogram abnormalities after feeding formulas low in n-3 fatty Further dietary studies in fullterm human infants are needed. These studies support the conclusion that there should be adequate amounts of both n-3 and n-6 fatty acids in the diet throughout life and that their ratio is of great importance. This is particularly important for infants fed artificial formulas whose fat sources include coconut oil, corn oil, or safflower oil. Such formulas may not meet the criteria of having adequate n-3 fatty acids and may have a relative excess of n-6 fatty acids. Ideally, the fat content and fatty acid composition of infant formulas, including the content of both n-3 and n-6 fatty acids, should resemble human milk. This objective seems reasonable and technologically feasible. Acknowledgments. Supported by the National Institute of Diabetes, Digestive and Kidney Disease Research Grants DK29930 and the Clinical Nutrition Unit Grant DK40566; the General Clinical Research Center Grant RR00334; and the Oregon Regional Primate Research Center Core Grant RR00063. This is publication number 1800 of the Oregon Regional Primate Research Center. Portions of this article were previously published in World Review of Nutrition and Dietetics, Vol. 66, 1990, Health Effects of n-3 Polyunsaturated Fatty Acids in Seafoods.
Nutrition Reviews, Vol. 50, No. 4
1. Burr GO, Burr MM. A new deficiency disease produced by the rigid exclusion of fat from the diet. J Biol Chem 1929;82:345-67 2. Holman RT. Essential fatty acid deficiency. Prog Chem Fats Other Lipids 1968;9:275-348 3. Sinnhuber RO, Castell JD, Lee DJ. Essential fatty acid requirement of the rainbow trout, Salmo gairdneri. Fed Proc Fed Am SOCExp Biol 1972;31: 1436-41 4. Tinoco J. Dietary requirements and functions of a-linolenic acid in animals. Prog Lipid Res 1982; 21 : I 4 5 5. Neuringer M, Anderson GJ, Connor WE. The essentiality of n-3 fatty acids for the development and function of the retina and brain. Ann Rev Nutr 1988;8:517-41 6. Neuringer M, Connor WE, Van Petten C, Barstad L. Dietary omega-3 fatty acid deficiency and visual loss in infant rhesus monkeys. J Clin Invest 1984; 73:272-6 7. Connor WE, Neuringer M, Barstad L, Lin DS. Dietary deprivation of linolenic acid in rhesus monkeys: Effects on plasma and tissue fatty acid composition and visual function. Trans Assoc Am Phys Vol. XCVII, 1984 8. Reisbick S, Neuringer M, Hasnain R, Connor WE. Polydipsia i n rhesus monkeys deficient i n omega-3 fatty acids. Physiol Behav 1990;47:31523 9. Neuringer MD, Connor WE, Lin DS, Barstad L, Luck S. Biochemical and functional effects of prenatal and postnatal omega-3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc Natl Acad Sci USA 1986;83:4021-25 10. Lin DS, Connor WE, Anderson GJ, Neuringer M. The effects of dietary n-3 fatty acids upon the phospholipid molecular species of monkey brain. J Neurochem 1990;55:1200-7 11. Brown WR, Hansen AE, Burr GD, McQuarrie I. Effects of prolonged use of extremely low-fat diet on adult human subjects. J Nutr 1938;16:511-24 12. Holman RT, Johnson SB, Hatch TF. A case of human linolenic acid deficiency involving neurological abnormalities. Am J Clin Nutr 1982;35:617-23 13. Bjerve KS, Mostad IL, Thoresen L. Alpha-linolenic acid deficiency in patients on long-term gastric tube feeding: Estimation of linolenic acid and long-chain unsaturated n-3 fatty acid requirement in man. Am J Clin Nutr 1987;45:66-77
Nutrition Reviews, Vol. 50, No. 4
14. Bjerve KS, Fischer S, Alme K. Alpha-linolenic acid deficiency in man: Effect of ethyl linolenate on plasma and erythrocyte fatty acid composition and biosynthesis of prostanoids. Am J Clin Nutr 1987;46:570-6 15. Wene JD, Connor WE, DenBesten L. The development of essential fatty acid deficiency in healthy men fed fat-free diets intravenously and orally. S Clin Invest 1975;56:127-34 16. Anderson GJ, Connor WE. On the demonstration of 17-3 essential-fatty-acid deficiency in humans. Am J Clin Nutr 1989;49:585-7 17. Connor WE, Neuringer M, Lin DS. Dietary effects upon brain fatty acid composition: the reversibility of n-3 fatty acid deficiency and turnover of docosahexaenoic acid in the brain, erythrocytes, and plasma of rhesus monkeys. J Lipid Res 1990;31: 237-48 18. Neuringer M, Connor WE. Omega-3 fatty acids in the retina. In: Galli C, Simopoulos AP, eds. Dietary n-3 and n-6 fatty acids: biological effects and nutritional essentiality. New York: Plenum Publishing, 1989377-90 19. Ruyle M, Connor WE, Anderson GJ, Lowensohn RI. Placental transfer of essential fatty acids in humans: venous arterial difference for docosahexaenoic acid in fetal umbilical erythrocytes. Proc Natl Acad Sci USA 1990;87:7902-6 20. Crawford MA, Hassam AG, Williams G. Essential fatty acids and fetal brain growth. Lancet 1976;l: 452-3 21. Harris WS, Connor WE, Lindsey S. Will dietary omega-3 fatty acids change the composition of human milk? Am J Clin Nutr 1984;40:780-5 22. FAO/WHO. Dietary fats and oils in human nutrition: a joint FAO/WHO report, Rome, FAO, 1978 23. Carlson SE, Rhodes PG, Ferguson MG. Docosahexaenoic acid status of preterm infants at birth and following feeding with human milk or formula. Am J Clin Nutr 1986;44:798-804 24. Carlson S, Cooke R, Werkman S, et al. Docosahexaenoic (DHA) and eicosapentaenoate (EPA) supplementation of preterm infants: Effects on phospholipid DHA and visual acuity. Fed Proc 1989;3:A1056 25. Uauy RD, Birch DG, Birch EE, Tyson JE, Hoffman DR. Effect of dietary omega-3 fatty acids on retinal function of very-low-birth weight neonates. Pediatr Res 1990;28:485-92
29
Essential Fatty Acids: The Importance of 12-3 Fatty Acids in the Retina and Brain William E. Connor, M.D., Martha Neuringer, Ph.D., and Sydney Reisbick, Ph.D.
Humans need fat in the diet in order to survive, grow, and prosper. Over 60 years ago, Burr and Burr' demonstrated that fat was an essential component of the diet. Rats reared on a fat-free diet failed to grow and reproduce and also developed renal disease, fatty liver, dermatitis, and necrosis of the tail. Later studies identified the deficient components of the diet as polyunsaturated fatty acids with two or more double bonds.* Until recently, linoleic acid, an n-6 fatty acid* commonly found in vegetable oils and many other foods, was deemed the primary essential fatty acid, together with its derivative n-6 fatty acids, of which arachidonic acid (20:4) was the most important. However, it now appears that two classes of fatty acids are both essential for health (Figure 1). The second series of highly polyunsaturated fatty acids includes the n-3 fatty acids, a-linolenic acid (18:3) and its longerchained, more polyunsaturated derivative, docosahexaenoic acid (22:6). n-3 fatty acids are also contained in a wide variety of natural foodstuffs, including some but not all liquid vegetable oils, particularly soybean and rapeseed oils, and green leafy vegetables. The higher derivative, docosahexaenoic acid, is found only in the plants of the sea, phytoplankton, or, as one ascends the food chain, in shellfish, fish, and sea mammals (Figure 2). Docosahexaenoic acid is especially prominent in brain, retina, and spermatozoa'but is also found in phospholipid membranes throughout the body, including the myocardium. Dr. Connor is a Professor of Medicine and Section Head, Dr. Neuringer is a Research Associate Professor, and Dr. Reisbick is a Research Assistant Professor at the Section of Clinical Nutrition, Department of Medicine, L-465, Oregan Health Sciences University, 3181 South West Sam Jackson Park Road, Portland, OR 97201, USA. 'In fatty acid nomenclature, the terms n-6 or n-3 may be used interchangeably with omega-6 or omega-3.
Nutrition Reviews, Vol. 50, No. 4
Aside from essential amino acids, n-3 fatty acids constitute the largest component of the cerebral cortex and retina that can be obtained only from the diet. The body cannot synthesize either the n-3 or the n-6 structure, as shown in Figure 1. Once the basic n-3 structure is consumed in the form of linolenic acid (18:3), the body can synthesize the longer-chain and highly polyunsaturated fatty acid docosahexaenoic acid (22:6) (Figure 3). Additional double bonds are inserted through the action of desaturase enzymes, and the 18-carbon chain of the precursor, linolenic acid, is elongated to 20 and then to 22 carbons. Similarly, in the n-6 series, linoleic acid (18:2) is converted to arachidonic acid (20:4), the predominant n-6 fatty acid in most tissues, and also to 22:4 and 225. Longer-chained fatty acids of the n-6 series (225) have become of interest in the n-3 dietary deficiency syndrome, which will be described subsequently. n-3 and n-6 fatty acids compete for the desaturase enzymes; high levels of linoleic acid inhibit the conversion of linolenic acid in the n-3 series to the next form, 18:4. Thus, a high ratio of linoleic acid to linolenic acid is likely to produce the greatest depletion of longer-chain n-3 fatty acids such as docosahexaenoic acid (22:6). It is this latter fatty acid, with its extreme degree of unsaturation, that is present in particularly high concentrations in the outer segment membranes of the rods and cones of the retina and in the synaptic membranes of the brain. The essentiality of n-3 fatty acids has been fully demonstrated in fish.3 There is also evidence of their importance in retinal and brain function in rats4" and especially in the nonhuman primate, the rhesus monkey.6 A diet deficient in n-3 fatty acids leads to a triad of signs in the rhesus monkey: visual impairment, abnormalities of the electroretinogram, and p o l y d i p ~ i aProfound .~~ biochemical changes in the fatty-acid composition of the membranes of the retina, brain, and other organs accompany these other disturbance^.^*'^ Low concentrations of n-3 fatty acids, especially 22:6, occur at birth in the
21
FATTY ACID NOMENCLATURE &Y
DIETARY SOURCES
18:2 (w6)
STRUCTURE
FATTY ACID
18:3-18:4-20:4
wRCOOH
w3
Eicoaovantoenoic Acid ICZ0:S w31
ru6
Linolalc Acid (C18:2w6)
w9
Olalc Acid
H,C
Marine Oils, Fish
6
Vegetable Oils; Animal Fats
Figure 1. The three major series of unsaturated fatty acids and the structure of the terminal end of the fatty acid chain for each series.
plasma, red blood cells, and neural tissues of rhesus infants born from mothers fed an n-3-deficient diet.6*7.9These concentrations become even lower as the deficient diet is continued postnatally. Visual impairment can be demonstrated by four weeks of age, and abnormalities of the electroretinogram occur by three months of age.639Polydipsia develops later in life.* In contrast to the more obvious clinical stigmata of n-6 fatty acid deficiency, the findings of n-3 deficiency are far more subtle, as might be expected for a fatty acid class required for neural membranes. n-3 fatty acid-deficient rhesus monkeys appear grossly normal, even after long-term depletion. The fur and skin do not show the conspicuous dermatitis found in n-6 fatty acid deficiency, nor does fatty liver occur. Table 1 illustrates the characteristic differences, clinical and biochemical, between essential fatty acid deficiencies of the n-6 and n-3 series. As noted above, n-6 fatty acid deficiency has been well-known and described in animals and humans.*"' Until recently, n-3 fatty acid deficiency, because of its less conspicuous symptomatology , has not been well-categorized and has at times oc-
Phytoplankton Algae
I
Terrestrial (18:3)
Plants (leaves, some nuts & seeds)
Krill, shellfish
+I
Fish
tt
Humans
Humans Figure 2. The dietary sources of n-3 fatty acids according to the food chains in the waters and on land. Note that the only rich source of the very polyunsaturated n-3 fatty acids is from phytoplankton and the sea life that directly or indirectly feeds on phytoplankton. 22
-=
Vegetable Oils 9
(20:5,22:6)
--205-225
' RCOOH
""
Marine
y~ 3 Serieq
w 6 Series 18:2-18:3-20:3-=-=4--
Figure 3. The major pathways for synthesis of longerchain n-3 and n-6 fatty acids from linolenic (18:3) and linoleic (18:2) acids. As indicated by the dotted line, high levels of linoleic acid block the desaturation of linolenic acid. In tissues, docosahexaenoic acid is the major n-3 fatty acid and arachidonic acid (20:4) is the major n-6 fatty acid.
curred in combination with n-6 fatty acid deficiency. '*-I6 Methods of Study
Rhesus monkeys were studied after a combination of maternal and postnatal deprivation of dietary n-3 fatty acids. Throughout pregnancy, adult females were given a semipurified diet with safflower oil as the only fat source.6 Their infants were fed a liquid version of the same diet from birth onwards. Safflower oil is particularly low in n-3 fatty acids, with less than 0.3% of the total fatty acids as linolenic acid (18:3, n-3). It is especially high in n-6 fatty acids, with a ratio of n-6 to n-3 fatty acids of about 250: 1. Mothers and infants in the control group received diets with the same composition except that the fat source was soybean oil, which provides 7.7% of total fatty acids as linolenic acid and which has an n-6 to n-3 ratio of 7: 1. The more highly polyunsaturated fatty acids of the n-3 series, such as eicosapentaenoic and docosahexaenoic, were not present in either of these diets. To test the reversibility of the deficiency, five of the deficient offspring were repleted with very longchain and highly polyunsaturated n-3 fatty acids from fish oil beginning at 10-24 months of age." In the repletion diet, fish oil replaced 80% of the safflower oil, the remaining safflower oil providing ample n-6 linoleic acid (4.5% of energy) and the fish oil supplying large amounts of the n-3 fatty acids, docosahexaenoic acid and eicosapentaenoic acid (205). Plasma, erythrocyte, and tissue samples, including whole retina and cerebral cortex, were analyzed for fatty acid composition by capillary column gasliquid chromatography after separation of total phospholipids or individual phospholipid classes by thin-layer chromatography. l7 In addition, in order to follow the time course of the biochemical Nutrition Reviews, Vol. 50, No. 4
Table 1. Differing Characteristics of n-3 and n-6 Essential Fatty Acid Deficiencies
Clinical Features
Biochemical Markers
n-3
n-6
Normal skin, growth and reproduction Reduced learning Abnormal electroretinogram Impaired vision Polydipsia Decreased 18:3 n-3 and 22:6 n-3 Increased 22:4 n-6 and 2 2 5 n-6 Increased 20:3 n-9 (only if n-6 also low)
Growth retardation Skin lesions Reproductive failure Fatty liver Polydipsia Decreased 18:2 n-6 and 20:4 n-6 Increased 20:3 n-9 (only if n-3 also low)
changes, serial biopsies of the frontal cortical gray matter were obtained for fatty acid analysis after craniotomy . Visual acuity was measured by the preferentiallooking method as described previously.6 The specific physiological effects of n-3 fatty acid deficiency on the function of the retina were examined by the e l e c t r ~ r e t i n o g r a m . ~ ~ ' ~ ResuIts Faffy Acid Composition of Plasma and Tissues Dietary deprivation of n-3 fatty acids resulted in low plasma and red blood cell concentrations of all n-3 fatty acids (Figure 4). The plasma concentrations of these fatty acids were low even at birth in the deficient monkeys, and their mothers had plasma n-3 fatty acid concentrations less than 50% of controls. In the deficient animals, the n-3 fatty acids were barely detectable in the phospholipids of plasma Adult females u)
and erythrocytes by 24 weeks of age, in contrast to the concentrations detected in the control animals. The brain and retina also had very low levels of n-3 fatty acids, particularly docosahexaenoic acid. In deficient animals at or near birth, docosahexaenoic acid concentrations in phosphatidylethanolamine were reduced by 50% in the retina and by 75% in the cerebral cortex compared to the control values (Figure 5). The proportion of docosahexaenoic acid in both tissues doubled between birth and two years of age in control monkeys but failed to increase in the deficient group, so that by two years docosahexaenoic acid concentrations in deficient monkeys were reduced to 15 to 20% of control values. It is notable that n-6 fatty acids, 2 2 5 particularly, compensatorily increased in response to the reduced docosahexaenoic acid, so that polyunsaturation of the phospholipid membranes was preserved as much as possible. This n-6 fatty acid comprised less than 1% of total fatty acids in the tissue Infants
1
2
0
a
c
c
Q,
$
n
Figure 4. Total n-3 fatty acids (weight % of total fatty acids) in plasma phospholipids of adult female rhesus monkeys and their offspring fed control (soybean oil) and n-3 acid-deficient (safflower oil) diets. Red blood cells date shown in f w r e 6 . Nutrition Reviews, Vol. 50, No. 4
23
Cerebral Cortex
/
Retina
/
/
/
Control/ /
/
/
/
/
/
/
fl Deficient
+ At Birth
Deficient
I
At 2 Yrs.
Figure 5. The concentrations of docosahexaenoic acid in the phosphatidylethanolamine of the cerebral cortex and retina of control animals and deficient animals at birth and at 2 years. Note that at birth both retina and cerebral
At Birth
At 2 Yrs.
cortex had much higher levels of docosahexaenoic acid in control than in deficient animals, and these disparities increased greatly by 2 years of age.
phospholipids of normal animals but in two-yearold deficient animals rose to approximately 20% in the phosphatidylethanolamineof the cerebral cortex and to nearly 30% in the retina. After dietary repletion with fish oil, these changes in fatty acid composition were rapidly reversed, as indicated by changes in plasma phospholipids, red blood cells, and biopsy specimens of frontal cortex (Figures 6 and 7)." The changes in the fatty acid composition of the frontal cortex also occurred rapidly, as early as one week after fish oil supplementation. By 24 to 28 weeks, the docosahexaenoic acid in phosphatidylethanolamine increased from 4.2 to 29.3% of total fatty acids (Figure 7), compared to the 22.3% in soybean oil-fed control animals. It is particularly interesting that 2 0 5 n-3 and 2 2 5 n-3 both increased from 0% to approximately 3% in the cerebral cortex while at the same time, the concentrations of 2 2 5 n-6 and other longer-chain n-6 fatty acids decreased. Visual Function and the Electroretinogram The visual acuity of n-3 fatty acid-deficient infant monkeys was reduced by at eight and l 2 weeks of age (Figure 8 ) ~as Previously Furthermore, deficient monkeys developed a number of abnormalities in the electroretinogram (ERG).9*'8The 24
Figure 6. The fatty acids of plasma and erythrocyte phospholipids in n-3 fatty acid-deficient monkeys and mankeys repleted with fish oil. Note the reciprocal relationships between the n-3 and n-6 fatty acids: n-3 fatty acids increase after repletion with the fish oil diet whereas n-6 fatty acids are diminished. Nutrition Reviews, Vol. 50, No. 4
40-
(I)
4 0
30-
4
3. L
m
-m LL
5
I1
20-
I.+-
0
c
C
Q,
2
Q,
a
10-
I
P -0 Qo 0 'h0, @%*
0 1
3
6
8
Weeks on Repletion Diet
/ '
Autopsy
Figure 7. The time course of fatty acid changes in phosphatidylethanolamine of the cerebral cortex of five juvenile monkeys fed fish oil for 43-129 weeks. As docosahexaenoic acid (DHA) increased, 2 2 5 n-6 decreased reciprocally. Levels of DHA and 2 2 3 n-6 in phosphatidylethanolamine of the frontal cortex of monkeys fed
control (soybean oil) and deficient diets from a previous study' are given for comparison. DHA and 225 n-6 in control monkeys were 22.3 2 0.3% and 1.4 2 0.3% of total fatty acids, respectively. DHA and 2 2 5 n-6 in the deficient monkeys were 3.8 ? 0.4% and 18.3 2 2.5%, respectively.
timing of the ERG response was altered, with significant delays in the peak latency (time to the B-wave peak) of both cone and rod responses (Figure 9). In contrast to previous studies in n-3 fatty aciddeficient rats, differences in response amplitudes were not detected at seven to 24 months of age.
However, more recent recordings of younger infants at three to four months of age have demonstrated clear differences in the A-wave amplitudes of both rod and cone responses. The reason for the transient nature of this effect is unknown. Deficient animals also showed a specific abnormality in the rate of recovery of the ERG response after an initial bright flash. This effect was present at three months but increased in magnitude with age. With an interval of 3.2 seconds between flashes, response amplitude in the deficient animals was reduced nearly twice as much as in controls, relative to the maximal amplitude seen in response to the first flash or to flashes presented at intervals of 20 seconds or more (Figure 10). Thus, recovery of the capacity to generate a full ERG response was significantly slowed. In deficient monkeys repleted with fish oil, ERGS were recorded at three, six, and nine months after the beginning of the repletion phase. Despite the increase in n-3 fatty acid levels in tissues, no improvement was seen in either peak latencies or in the ERG recovery function."
l,ol,o 20130
4 weeks
8 weeks
12 weeks
20150 20160 20175
2
.-
20/100 201120 4. 201150 201200
-h
:
(I)
201300
201600
@o /> ' 0"5 0 ,
@ / "
/ '
Figure 8. Visual acuity thresholds (mean t SEM) as determined by the preferential-looking method for control and n-3 fatty acid-deficient infant monkeys. Thresholds are expressed in cycles per degree of visual angle and in the equivalent Snellen values. The p values for statistical significance were determined by Student's t-test. Nutrition Reviews, Vol. 50, No. 4
Polydipsia in Deficient Monkeys Cage behavior in n-3 fatty acid-deficient and control monkeys was monitored by videotaping. It was noticed that the deficient monkeys visited the water spouts of their cages much more often than did con25
trol monkeys, suggesting the possibility of greater water intake. A study of 24-hours water intakes was then camed out to confirm and quantify this difference.8 Figure 11 depicts the mean water intake in milliliters per kilogram of body weight over 24 hours (A) and the combined excretion of urine and feces over the same period (B). Water intake was more than double for deficient compared to control monkeys (264 vs. 70 mVkg) as was excretion (268 vs. 121 g/kg). The excretion was necessarily a combined output of feces and water because there was mixing of the soft stools and urine. These studies were repeated on several different occasions with similar results in monkeys varying in age from 18 months to young adults. The mechanism of the polydipsia is not yet understood, but several possible explanations can be eliminated. The effect is probably not caused by an osmotic imbalance. Dietary electrolytes and serum electrolytes were similar in both control and deficient animals. Fasting glucose levels were normal, indicating that polydipsia did not result from fluid loss as a result of diabetes. The blood urea nitrogen and creatinine levels were also equal and normal, with no indication of renal insufficiency in either group. Polydipsia does occur in dietary n-6 fatty acid deficiency because of increased skin permeability and the resulting loss of water by evaporation. However, the amount of n-6 fatty acids in the diet was actually 50% higher in deficient than in control
1.oo
a, .80 U
-3
c,
. I
f .60
a
T
a,
.? c,
-aa,
a
.40
.20
0
C
D
Figure 10. Relative B-wave amplitude (mean 5 SEM) of the ERG elicited by flashes at 3.2-seconds intervals, as a percent of the maximal amplitude produced at intervals of 20 seconds or more. Relative amplitude is significantly reduced in the deficient group (p < 0.01).
monkeys. In addition, the skin and fur of the n-3 fatty acid-deficient animals appeared visually and histologically normal up to two years of age. Investigations are now in progress to describe possible hormonal and prostaglandin mechanisms of the polydipsia.
Discussion
1 1
tJ
\
Deficient
I
I
I
0
100
200
I
300 msec
Figure 9. Representative ERG waveforms from control and n-3 fatty acid-deficient monkeys under conditions selectively stimulating the rod system. The vertical arrow indicates time of flash. B-wave peak latencies were delayed in the deficient group; vertical lines have been drawn through the peaks to aid comparison. 26
This study has shown that dietary n-3 fatty acid deficiency leads to severe and progressive depletion of n-3 fatty acids from the plasma and from all tissues analyzed, including red blood cells, liver, skin, fat, cerebral cortex, and retina. In particular, the very long-chain n-3 fatty acid, docosahexaenoic (22:6), was selectively depleted from neural and retinal phospholipids and was replaced by n-6 fatty acids, in particular 2 2 5 . Associated with these biochemical changes were a significant impairment in the development of visual acuity, abnormalities in the electroretinogram, and an increase in fluid intake and excretion. In the initial experiments, linolenic acid (18:3) was the only dietary source of n-3 fatty acids. In the control animals, docosahexaenoic acid was synthesized by successive and multiple steps of desaturation and elongation from linolenic acid. It then selectively accumulated in brain and retinal phosphoNutrition Reviews, Vol. 50, No. 4
40
LA
Deficient Diet
H
(n=7)
Control Diet
0-4
(n=7)
if4
acid in phospholipids of the retina, particularly in the outer segment membranes of the photoreceptor cells. This fatty acid is thought to be responsible for the special biophysical properties of the outer segment membranes, which contain the visual pigment. Depletion of docosahexaenoic acid from these membranes would be expected to alter their physical properties and, therefore, the efficiency of the visual process. It is also possible that docosahexaenoic acid has a more specific biochemical function in the retina. In addition to the retina, docosahexaenoic acid is also a major fatty acid of brain gray matter, especially synaptic membranes, which, like photoreceptor membranes, are excitable and highly fluid. Thus, the change in membrane phospholipid composition produced by n-3 fatty acid deficiency might alter the transmission of information through the brain’s visual pathways as well as affecting the photoreceptive process in the retina. A few cases of n-3 fatty acid deficiency have been described in humans. Holman et al.” recently reported a clinical case of peripheral neuropathy and blurred vision in a child receiving total parenteral nutrition. The symptoms were attributed to n-3 fatty acid deficiency because safflower oil, as in our study, was the sole fat source. Replacement of the safflower oil emulsion with linolenic acid-rich soybean oil was associated with recovery. However, it is difficult to be certain that the symptoms were due to n-3 fatty acid deficiency rather than to some other metabolic disturbance induced by long-term parenteral nutrition. More recently Bjerve et al. I 3 * l 4 described ten cases of presumed linolenic acid deficiency occurring in a number of nursing-home patients in Norway, some semicomatose, who had been fed over several years by gastric tubes. The diet of these patients was based on a commercially available powder supplement containing small amounts of corn oil ( 1 . 3 g / l O O g) and mixed with skim milk. The authors reported that the patients developed very low plasma levels of n-3 fatty acids and a scaly dermatitis. Corn oil has an especiaily low content of linolenic acid (0.3%) but, if supplied in quantity, would have ample n-6 fatty acid (i.e., linoleic acid or 18:2) to meet the needs for n-6 essential fatty acids. However, the diet furnished only 0.5% of energy as 18:2 n-6, the plasma 18:2 n-6 was low (99 to 15% of total plasma fatty acids vs. 35% in healthy control subjects’s-’5), and levels of eicosatrienoic acid (20:3 n-9) were substantially elevated. These observations suggest that the patients were also deficient in n-6 fatty acid and that this was the cause of their skin lesions, since dermatitis has not been found in animals having pure n-3 fatty acid deficiency. It is likely that these patients had a com-
’*
s c
8
c5
w
13
10
l A
ifi
0, v
5s
22-7
t
,
0
20
18
e
8
,
,
10 AM
,
l
12
1
,
1
14
,
,
16
,
,
,
,
20
18
PM
1
22-7 AvelHour in Dark
Hours of the D a y
Figure 11. The mean hourly water intake (mL/kg) over a 24-hours period and the mean hourly output of combined urine and feces (g/kg) over 24 hours. The AveIHour in dark refers to the measurement taken at the end of the 10-hours dark period divided by 10.
lipids, particularly in the phosphatidylethanolamine and phosphatidyl-serine fractions.’ The conversion of linolenic acid to docosahexaenoic acid may occur in the liver, in the brain or retina, or in the placenta during fetal development. In the deficient monkeys, the dietary supply of linolenic acid was insufficient to support the synthesis of adequate levels of docosahexaenoic acid. High levels of dietary linoleic acid also may have suppressed synthesis of docosahexaenoic acid from the available linolenic acid. l9 However, during intrauterine life our experimental monkeys received some docosahexaenoic acid directly from their mothers via the placenta. Although deficient mothers had been receiving the n-3 fatty acid-deficient diet for at least two months before conception, their stores of docosahexaenoic acid were probably depleted very slowly. The plasma of newborn deficient infants and their mothers revealed the presence of some docosahexaenoic acid, the levels being consistently higher in the infants’ plasma than in their mothers’, a process termed biomagnification.20 However, plasma concentrations of docosahexaenoic acid were much lower in the deficient infants than in control infants. Once the infants were removed from any maternal sources of n-3 fatty acids, their plasma levels fell rapidly. They were separated from their mothers at birth and, therefore, did not have access to maternal milk, a natural source of docosahexaenoic acid. Docosahexaenoic acid is the predominant fatty Nutrition Reviews, Vol. 50, No. 4
27
bined essential fatty acid deficiency which included both n-6 and n-3 fatty acid deficiencies.l6 The findings of our study provide the first experimental evidence for a dietary requirement for n-3 fatty acids in primates and add to the extensive studies in rats emphasizing the essentially of at least one of the n-3 fatty acids, linolenic acid.4 n-3 fatty acids are essential nutrients for retinal and brain function, especially during fetal and postnatal development. Both biochemical and functional stigmata are characteristic of the n-3 fatty acid deficient state. A recent review details the experimental background of n-3 fatty acid deficiency in different species of a n i m a ~ s . ~ Summary and Implications
In summary, n-3 fatty acid deficiency is characterized in rhesus monkeys by a triad of problems: impaired vision, abnormal electroretinograms, and polydipsia. Concomitant with the visual and ERG defects is the disturbed biochemistry of the retina. Docosahexaenoic acid (22:6 n-3) in the retinal phospholipids is decreased, whereas the n-6 fatty acids, particularly 2 2 5 , are increased. Similar biochemical abnormalities occur in the cerebral cortex. In the absence of n-3 fatty acids, the body produces the most similar polyunsaturated fatty acids possible from precursors, so that the amount of total polyunsaturation of the tissue membranes is largely maintained. The cause of the polydipsia may also be related to the same altered brain biochemistry, but does not seem to be mediated by renal disease or by abnormalities in the posterior pituitary hormones or in osmotic regulation. The diagnosis of n-3 fatty acid deficiency can be made at birth and later in life by determination of fatty acid concentrations in plasma and red cells of cord and peripheral venous blood. In deficiency, there will be low levels of the n-3 fatty acids 18:3 and 22:6 and high levels of the n-6 fatty acid 2 2 5 , which is normally very low in both plasma and tissues of primates. The deficiency state can be prevented at any age by the provision of adequate amounts of n-3 fatty acids in the diet. It i s not yet known whether both linolenic and docosahexaenoic acids should be present in the diet for optimal development; both are found in human milk. Once overt deficiency develops past infancy, biochemical correction of tissue levels may not necessarily restore all the functional defects. In our studies, the electroretinogram continued to be abnormal. Adequate provision of n-3 fatty acids is critical during three periods of life: pregnancy, lactation, and infancy. During pregnancy, these fatty acids are transferred across the placenta to the fetus. 19*20
28
During lactation, the diet of nursing mothers affects the n-3 content of their milk.21 If a formula is fed, adequate and balanced sources of both n-3 and n-6 fatty acids are required in the formula to provide these fatty acids postnatally. Moreover, in infancy, n-3 fatty acids continue to accumulate in both the brain and retina, and an adequate dietary source is important. In all probability, requirements for n-3 fatty acids continue during childhood and even in adult life. Elderly and other patients fed special diets may be at risk for n-3 fatty acid deficiency. We suggest that the dietary requirements for n-3 fatty acids in infancy are from 0.5 to 1.0% of total energy and ideally might include both 18:3 and 22:6, as does human milk. The suggested requirement for n-6 fatty acids should be from 2 to 6% of total energy and should include arachidonic acid, 20:4, as well as linoleic acid. The ratio of n-6 to n-3 fatty acids is important and should range from about four to ten.22Too high a ratio, as is present in the n-3 fatty acid-deficient safflower oil, might further accentuate the deficient state. Some infant formulas, particularly the dry powdered formulas containing corn and coconut oils, provide marginal amounts of n-3 fatty acids, which are as low as 0.1 to 0.2% of total energy. The ratio of n-6 to n-3 fatty acids, at 75: 1 , is also high (Connor WE, Van Winkle S, unpublished data). Recent research in preterm human infants has shown, as in monkeys, vision and electroretinogram abnormalities after feeding formulas low in n-3 fatty Further dietary studies in fullterm human infants are needed. These studies support the conclusion that there should be adequate amounts of both n-3 and n-6 fatty acids in the diet throughout life and that their ratio is of great importance. This is particularly important for infants fed artificial formulas whose fat sources include coconut oil, corn oil, or safflower oil. Such formulas may not meet the criteria of having adequate n-3 fatty acids and may have a relative excess of n-6 fatty acids. Ideally, the fat content and fatty acid composition of infant formulas, including the content of both n-3 and n-6 fatty acids, should resemble human milk. This objective seems reasonable and technologically feasible. Acknowledgments. Supported by the National Institute of Diabetes, Digestive and Kidney Disease Research Grants DK29930 and the Clinical Nutrition Unit Grant DK40566; the General Clinical Research Center Grant RR00334; and the Oregon Regional Primate Research Center Core Grant RR00063. This is publication number 1800 of the Oregon Regional Primate Research Center. Portions of this article were previously published in World Review of Nutrition and Dietetics, Vol. 66, 1990, Health Effects of n-3 Polyunsaturated Fatty Acids in Seafoods.
Nutrition Reviews, Vol. 50, No. 4
1. Burr GO, Burr MM. A new deficiency disease produced by the rigid exclusion of fat from the diet. J Biol Chem 1929;82:345-67 2. Holman RT. Essential fatty acid deficiency. Prog Chem Fats Other Lipids 1968;9:275-348 3. Sinnhuber RO, Castell JD, Lee DJ. Essential fatty acid requirement of the rainbow trout, Salmo gairdneri. Fed Proc Fed Am SOCExp Biol 1972;31: 1436-41 4. Tinoco J. Dietary requirements and functions of a-linolenic acid in animals. Prog Lipid Res 1982; 21 : I 4 5 5. Neuringer M, Anderson GJ, Connor WE. The essentiality of n-3 fatty acids for the development and function of the retina and brain. Ann Rev Nutr 1988;8:517-41 6. Neuringer M, Connor WE, Van Petten C, Barstad L. Dietary omega-3 fatty acid deficiency and visual loss in infant rhesus monkeys. J Clin Invest 1984; 73:272-6 7. Connor WE, Neuringer M, Barstad L, Lin DS. Dietary deprivation of linolenic acid in rhesus monkeys: Effects on plasma and tissue fatty acid composition and visual function. Trans Assoc Am Phys Vol. XCVII, 1984 8. Reisbick S, Neuringer M, Hasnain R, Connor WE. Polydipsia i n rhesus monkeys deficient i n omega-3 fatty acids. Physiol Behav 1990;47:31523 9. Neuringer MD, Connor WE, Lin DS, Barstad L, Luck S. Biochemical and functional effects of prenatal and postnatal omega-3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc Natl Acad Sci USA 1986;83:4021-25 10. Lin DS, Connor WE, Anderson GJ, Neuringer M. The effects of dietary n-3 fatty acids upon the phospholipid molecular species of monkey brain. J Neurochem 1990;55:1200-7 11. Brown WR, Hansen AE, Burr GD, McQuarrie I. Effects of prolonged use of extremely low-fat diet on adult human subjects. J Nutr 1938;16:511-24 12. Holman RT, Johnson SB, Hatch TF. A case of human linolenic acid deficiency involving neurological abnormalities. Am J Clin Nutr 1982;35:617-23 13. Bjerve KS, Mostad IL, Thoresen L. Alpha-linolenic acid deficiency in patients on long-term gastric tube feeding: Estimation of linolenic acid and long-chain unsaturated n-3 fatty acid requirement in man. Am J Clin Nutr 1987;45:66-77
Nutrition Reviews, Vol. 50, No. 4
14. Bjerve KS, Fischer S, Alme K. Alpha-linolenic acid deficiency in man: Effect of ethyl linolenate on plasma and erythrocyte fatty acid composition and biosynthesis of prostanoids. Am J Clin Nutr 1987;46:570-6 15. Wene JD, Connor WE, DenBesten L. The development of essential fatty acid deficiency in healthy men fed fat-free diets intravenously and orally. S Clin Invest 1975;56:127-34 16. Anderson GJ, Connor WE. On the demonstration of 17-3 essential-fatty-acid deficiency in humans. Am J Clin Nutr 1989;49:585-7 17. Connor WE, Neuringer M, Lin DS. Dietary effects upon brain fatty acid composition: the reversibility of n-3 fatty acid deficiency and turnover of docosahexaenoic acid in the brain, erythrocytes, and plasma of rhesus monkeys. J Lipid Res 1990;31: 237-48 18. Neuringer M, Connor WE. Omega-3 fatty acids in the retina. In: Galli C, Simopoulos AP, eds. Dietary n-3 and n-6 fatty acids: biological effects and nutritional essentiality. New York: Plenum Publishing, 1989377-90 19. Ruyle M, Connor WE, Anderson GJ, Lowensohn RI. Placental transfer of essential fatty acids in humans: venous arterial difference for docosahexaenoic acid in fetal umbilical erythrocytes. Proc Natl Acad Sci USA 1990;87:7902-6 20. Crawford MA, Hassam AG, Williams G. Essential fatty acids and fetal brain growth. Lancet 1976;l: 452-3 21. Harris WS, Connor WE, Lindsey S. Will dietary omega-3 fatty acids change the composition of human milk? Am J Clin Nutr 1984;40:780-5 22. FAO/WHO. Dietary fats and oils in human nutrition: a joint FAO/WHO report, Rome, FAO, 1978 23. Carlson SE, Rhodes PG, Ferguson MG. Docosahexaenoic acid status of preterm infants at birth and following feeding with human milk or formula. Am J Clin Nutr 1986;44:798-804 24. Carlson S, Cooke R, Werkman S, et al. Docosahexaenoic (DHA) and eicosapentaenoate (EPA) supplementation of preterm infants: Effects on phospholipid DHA and visual acuity. Fed Proc 1989;3:A1056 25. Uauy RD, Birch DG, Birch EE, Tyson JE, Hoffman DR. Effect of dietary omega-3 fatty acids on retinal function of very-low-birth weight neonates. Pediatr Res 1990;28:485-92
29
