![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Health Studies and Gerontology, University of Waterloo, Waterloo, ON, N2L 3G1, Canada and * Ross Laboratories, Columbus, OH 43215–1724
3To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-linolenic acid (g/100 g fat); with the use of a 2 x 2 design,
they were fed one of two levels of arachidonic acid (AA) and
docosahexaenoic acid (DHA) (0.0 and 2.5 g/100 g of fatty acids). A
fifth artificially reared group was fed a diet high in saturated fat,
and a sixth group was reared by dams fed a standard AIN-93M diet. The
pups were weaned onto modified AIN-93G diets, with a fat composition
similar to that fed during the artificial rearing period. Behavioral
testing was conducted between 6 and 9 wk of age; brain lipid
composition was then assessed. Relative to the unsupplemented group
(0.0 g/100 g AA and DHA), dietary supplementation resulted in a wide
range of AA (84–103%) and particularly DHA (86–119%) levels in
forebrain membrane phospholipids. AA supplementation increased AA
levels and decreased DHA levels, and DHA supplementation increased DHA
levels and decreased AA levels, with the magnitude of these effects
dependent on the level of the other fatty acid. DHA levels were very
low in the saturated fat group. The groups did not differ on the place
or cued version of the Morris water-maze, but on a test of working
memory, the saturated fat group was impaired relative to the suckled
control group. Further correlational analyses in the artificially
reared animals did not support a relationship between the wide range of
DHA and AA levels in the forebrain and working-memory
performance.
KEY WORDS: • rat • brain development • water-maze learning • arachidonic acid • docosahexaenoic acid
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-linolenic acid [LNA, 18:3 (n-3)] are essential
dietary fatty acids (EFA) because they cannot be synthesized de novo by
animals and must therefore be provided in the diet. These 18-carbon
compounds are then converted to longer-chain polyunsaturated fatty
acids (LCPUFA) through a series of desaturation and elongation
reactions (Cook 1991
). The LCPUFA that predominate in
the brain are arachidonic acid [AA, 20:4(n-6)] and docosahexaenoic
acid [DHA, 22:6(n-3)] (Sastry 1985
). In addition to
serving a structural role as components of membrane phospholipids, the
20-carbon compounds also make an important contribution to regulatory
function by serving as precursors of the eicosanoids, which include the
prostaglandins and leukotrienes; AA has itself been implicated as a
putative second messenger in neural signaling pathways (reviewed in
Wainwright 1997
).
The importance of the availability of (n-6) fatty acids (FA) for growth
has been clearly established (reviewed in Innis 1991
).
Elucidation of the physiologic role for the (n-3) FA, on the other
hand, has been based largely on functional outcomes, including
behavioral effects (Reisbick and Neuringer 1997
,
Wainwright 1997
). Levels of DHA are high in the testes,
retina and gray matter of the brain compared with levels in most other
tissues. Dietary restriction of (n-3) FA (i.e., LNA, DHA and metabolic
intermediates) typically leads to a decrease in membrane levels of DHA,
with a reciprocal increase in 22:5(n-6) (e.g., Galli et al. 1971
, Mohrhauer and Holman 1963
, Tinoco et al. 1978
). Supplementation of preterm infant formulas with
marine oil has been associated with indications of reduced growth that
was more pronounced if the supplemented oils contained high levels of
eicosapentaenoic acid [EPA, 20:5(n-3)] as well as DHA (Carlson et al. 1992, 1993b and 1996
). In this case, the supplemented
groups also showed lower blood levels of AA, suggesting that the
reduced growth may have resulted from reduced availability of AA due to
competitive inhibition by EPA (Carlson et al. 1993b
).
This emphasizes the importance of taking the metabolic interactions
between the (n-3) and (n-6) FA into account in dietary formulations.
Numerous animal studies have addressed the effects of dietary (n-3) FA
deficiency on retinal and behavioral function. These studies include
work on rhesus monkeys (reviewed by Connor et al. 1992
),
as well as a large number of studies that have been conducted in
rodents, particularly rats and mice. Although this work provides
considerable evidence for the importance of adequate levels of DHA in
normal retinal function, a corresponding role for DHA in brain with
respect to cognitive capacity remains to be firmly established. Some,
but not all of these studies show alterations in performance in
learning tasks in animals fed (n-3) FA–restricted diets, but in many
cases, because of methodological limitations, these findings are
difficult to interpret (reviewed in Reisbick and Neuringer 1997
, Wainwright 1992
and 1997
,
Wainwright and Ward 1997
). Although these studies
encompass various tests of so-called "learning ability," very
often the study designs do not separate what might be considered truly
cognitive differences from related effects on task performance, such as
those on the animals' sensory or motor capacities, for example, or on
their motivation to perform the task. It is also important to realize
that questions about a putative role for specific nutrients in
cognitive function cannot be answered by considering the findings on
any one cognitive task, but rather by consideration of the pattern of
results obtained over a series of different behavioral tests (discussed
in Wainwright 1997
).
In our previous work with mice, we addressed the question whether
manipulations of dietary FA supply affect species-typical learning,
i.e., social learning of a food preference, as well as instrumental
learning, i.e., spatial water escape learning in the Morris
water-maze (e.g., Wainwright et al. 1994a, 1994b and 1997
). In none of these studies did we show effects of (n-3) FA
deficiency on learning ability per se, although there were
some indications of performance effects. It may be, however, that
acquisition and reversal of place-learning in the Morris maze, in
which the animal is required to learn the fixed location of a hidden
platform relative to extra-maze cues, constitiute too easy a task
to show subtle differences in function. Strupp and Levitsky (1995)
have suggested that rather than simple tests of
learning, tests of executive function may be better indicators of early
nutritional insult. One such function is the so-called "working
memory"; the animal is presented with a series of problems, and in
order to solve each problem, it is required to remember details that
are specific to that problem only, and to then start afresh on the next
problem. This type of memory is distinguished from "reference
memory," which is memory of the general procedural requirements of
the task that remain constant across testing. Thus we recently
conducted a study in rats to ascertain whether there were effects of
two generations of dietary (n-3) FA restriction on an adaptation of the
Morris water-maze that specifically measured working memory. In
this study, we used rats rather than mice because of evidence that rats
are more proficient at this task than mice (Whishaw 1995
). The two groups did not differ in learning the position
of the hidden platform in the Morris water-maze, but the
(n-3)–deficient rats showed deficits in performance on the
working-memory task (Wainwright et al. 1998
).
Not only has dietary restriction of (n-3) FA (diets devoid or very low
in LNA) been associated with effects on behavior, but there is also
evidence that diets containing very high levels of long-chain (n-3)
FA and extremes in the (n-6)/(n-3) ratio can also affect development.
One example is work showing that newborn guinea pigs that were fed
diets containing large amounts of fish oil, which had high levels of
(n-3) FA as EPA and DHA [(n-6)/(n-3) ratio 2.54] and which in turn
resulted in high levels of DHA in the retina, had electroretinograms
that resembled those of animals fed (n-3) FA–restricted diets
(Weisinger et al. 1996
). Another recent study that used
a maternal supplementation model in rats reported that the offspring of
dams fed a diet containing fish oil [(n-6)/(n-3) 
0.125)] showed
delayed appearance of the auditory startle response and longer
brainstem auditory conduction times (Saste et al. 1998
).
Furthermore, we have shown that mice fed diets containing very high
levels of DHA with low levels of (n-6) FA [(n-6)/(n-3) 0.3]
during development showed impaired growth and delayed sensorimotor
development (Wainwright et al. 1996
) and swam more
slowly in a water-maze unless some of the (n-6) FA were provided in
the form of AA, rather than LA alone (Wainwright et al. 1997
).
A question currently of great interest is whether dietary provision of
LA and LNA is sufficient to support development of the central nervous
system, or whether it is also necessary to provide preformed sources of
AA and DHA. This is of particular concern in the case of preterm
infants, in whom AA and DHA would have normally been provided through
the placenta. It is here that the artificial rearing model in rats can
make an important contribution (Ward and Wainwright
1997
). In addition to allowing for the control of extraneous
variables not possible in studies with human infants, it allows for
systematic manipulations of dietary FA over a wide range of dietary
levels. The aim therefore of the present series of studies was to use
this model to supplement a diet containing the 18-carbon (n-6) and
(n-3) FA within the ranges found in rat milk (10% LA and 1% LNA) with
both AA and DHA. In the first of these studies (Ward et al. 1998
), rat pups were fed rat milk substitutes containing oils
providing 10% LA and 1% LNA through gastrostomy tubes from postnatal
d 5 to 18. With the use of a 3 x 3 factorial design, these diets
were supplemented with microbial cell oils providing one of three
levels of both AA and DHA (0.0, 0.4 and 2.4% of fatty acids). When
expressed as a percentage of the unsupplemented group, DHA in the
forebrain on d 18 ranged from 82 to 142%, whereas the range for AA was
less, from 86 to 110%; for the cerebellum the values were 77–153%
and 75–112%, respectively. The extreme values were obtained only when
either DHA or AA was provided alone at high levels. Thus, on the basis
of these findings, the intent of this study was to produce a wide range
of DHA and AA levels in the brain by the use of a similar preweaning
feeding protocol, followed by weaning the rats onto diets of similar
PUFA composition until behavioral testing, beginning at 6 wk. At the
end of behavioral testing at 9 wk, the FA composition of the brain was
measured to determine the relationship between the resulting range in
brain FA composition and performance on the working-memory task in
the Morris water-maze. Besides a suckled control group, the design
of this study also included an artificially reared group that was fed a
diet deficient in (n-3) FA and high in saturated fat. The literature
supports adverse effects of saturated fat diets on performance in
learning tasks in rats (Greenwood and Winocur 1996
), and
we have shown in addition that a saturated fat diet in mice results in
very low DHA levels in the brain (Wainwright et al. 1994a
). Given the previous findings cited above, associating
(n-3) deficiency in rats with effects on working memory, we expected
this group to serve as a "positive" control for negative
behavioral effects. Overall, we hypothesized that the findings would
indicate a relationship between levels of brain AA and DHA, as well as
of the (n-6)/(n-3) FA ratio and performance on the behavioral task.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The rats studied were the male offspring of timed-pregnant,
Long-Evans rats (Harlan Sprague Dawley, Indianapolis, IN). The pregnant
dams were obtained at 10–12 d gestation and were housed individually
with free access to food [AIN-93M (Reeves et al. 1993
)
supplied by Dyets, Bethlehem, PA]. They were maintained at 22 ± 1°C under a reversed –12-h light:dark cycle (lights off at 0600 h). Litters were culled when necessary to 10 pups within 24 h of
birth. The male pups from each litter were assigned randomly to groups,
with no more than one pup from any litter being assigned to a single
group. The sample size ranged from 19 to 21 rats per group. All
procedures used in this study were approved by the Animal Care
Committee at the University of Waterloo, in compliance with the Animals
for Research Act of Ontario (Revised Statutes of Ontario) and the Guide
for the Care and Use of Experimental Animals from the Canadian Council
on Animal Care.
Diets.
The details of the composition and mode of preparation of the rat milk
substitute have been described previously (Ward et al. 1998
). The dietary oils were comprised of medium-chain
triglyceride (MCT) oil, coconut, soy and high oleic safflower oils
(supplied by Ross Products Division, Abbott Laboratories, Columbus,
OH); AA and DHA were provided in the form of single-cell microbial
oils, ARAsco and DHAsco (Martek Biosciences, Columbia, MD). It is
important to emphasize that, apart from the saturated fat group, which
had minimal LA levels, all diets were formulated to have a ratio of
LA/LNA of ~10:1. The postweaning diet consisted of a custom-made
basal powder [AIN 93G (Reeves et al. 1993
) prepared
without added fat, Dyets] to which the appropriate dietary oil
mixtures were added. The FA composition of the dietary oil mixtures for
the rat milk substitute and powdered diet are shown in Tables 1 and
2 respectively; it should be noted that the differences between the
compositions of these two oils are related mainly to the short- and
medium-chain compounds. The values for rat milk are published in
our previous paper (Ward et al. 1998
). The dietary oils
were stored frozen under nitrogen, as were the prepared milk diets;
powdered diets were prepared three times a week and stored refrigerated
under nitrogen. The rat milk substitute was replaced daily before
weaning and, after weaning, the powdered diet was replaced every 2-3
d. All behavioral testing and brain fatty acid analyses were conducted
with the experimenter unaware of the dietary treatment.
|
|
The basic design was a 2 x 2 factorial, with two levels of AA and of DHA (0.0 and 2.5%). A fifth artificially reared group, the "positive" control group, was fed a diet high in saturated fat, but with sufficient LA to avoid adverse effects on growth and overt signs of EFA deficiency, i.e., high (n-6)/(n-3) ratio and trace levels of 18: 3(n-3). The sixth group, the suckled control group, consisted of pups that were fostered to nursing dams fed AIN-93M diet on the day their littermates were gastrostomized. This group functioned as the "normal" behavioral control group; it should be noted that it differed from the artificially reared groups in terms of both diet and early environment (rearing conditions). There were between 19 and 21 rats in each group, resulting in 121 rats in the study.
Artificial rearing procedure.
This has been described in detail previously (Ward et al. 1998
). Briefly, on d 27 postconception (~d 5 after birth),
the rat pups were anesthetized with methoxyfluorane inhalant (Metofane)
and the gastrostomy tube inserted. The gastrostomy tube itself was an
~15-cm length of Intramedic tubing (PE 10, Clay Adams, Parsippany,
NJ) with a small plastic flange at one end. This tube was attached to a
short wire contained within silastic tubing and lubricated with MCT
oil, inserted into the mouth, down the esophagus and out through the
stomach wall. Pups were housed individually in plastic cups floating in
a water bath maintained at 36 ± 1°C and were fed one of the
experimental diets via PE tubing attached at one end to their
gastrostomy tubes and at the other end to syringes attached to an
infusion pump (model #55–4143, Harvard Apparatus, South Natick, MA).
The pumps were programmed to deliver the formulas for 10 min every
hour, and the pups were fed an amount of diet representing 29% of
their body weight (adjusted daily) at the start of the procedure,
increasing to 36% of body weight after ~1 wk. The pups were weighed
each day, their gastrostomy tubes were flushed with 0.1 mL of deionized
water, and their anogenital regions were rubbed gently with a wet
tissue to stimulate urination and defecation. Suckled control pups were
also weighed daily. On d 40 postconception (13 d after gastrostomy) the
pups were weaned onto the AIN-93G diet as described previously and
housed in groups of 3–5. Behavioral testing began at 6 wk of age; at
the end of the testing period, the rats were killed under Halothane
anesthesia, and their brains removed and stored at -80°C.
Water-maze testing.
This test, which was developed by Morris (Stewart and Morris 1993
), uses a circular tank (152 cm diameter, 50 cm deep)
constructed of opaque white plastic. It is filled with water
(21–22°C) to a depth of 28 cm, and the water is rendered opaque by
the addition of soluble, nontoxic white latex paint. In the place
version of the maze, the rat develops a spatial map of the
extra-maze cues, which it then uses to locate the platform. Thus
the distance swum to the platform and the time taken in doing so should
decrease over testing sessions (Days) as the rat learns the location of
the platform. Moreover, it is expected that if the rat has learned the
location of the platform in relation to the extra-maze cues, its
initial response on the probe trial will be to swim directly to the
quadrant in which it expects to find the platform. Thus the distance
swum (and time spent) in the target quadrant should be greater than
that in the other two quadrants (excluding the start quadrant). The
distance swum to the platform as well as the latency to reach the
platform were monitored using the Videomex V tracking system (Columbus
Instruments, Columbus, OH). Distance can be considered a better measure
of learning than latency because it will not be affected by possible
differences in performance factors such as swimming speed (discussed in
Wainwright 1997a
). The behavioral testing was conducted
during the dark cycle, when rats are normally most active.
Place version (wk 1). The pool was located in a test room in which there were many extra-maze spatial cues. On the first 4 d, the rats were required to locate the hidden platform (9 cm x 9 cm) situated 1 cm below the surface of the water. There was one testing session per day, with four trials per session. On each trial, the rat was placed, facing the wall, in one of the four quadrants in the tank, and allowed to swim for a maximum of 45 s. Once the rat found the platform, it remained there for 10 s before being returned to the holding cage, which was kept warm on a heating pad. If the rat failed to find the platform in that time, it was placed on it for 10 s before being returned to the holding cage. Each of the four trials conducted each day was started from a different quadrant, with the order determined pseudorandomly (not twice from the same quadrant) and varying from day to day. The intertrial interval (ITI) was 60 s, counted from the end of one trial to the beginning of the next. On d 5, the platform was removed from the tank, and a probe trial was conducted by placing the rat in the quadrant opposite to that of the platform and then allowing it to swim for 15 s.
Working-memory version (wk 2). After a 2-d interval following the probe trial, the rats were tested with the maze set up as previously, but with the following variation in procedure. The animals were tested twice a day, morning and afternoon, for 5 d. In each testing session, the rat received a pair of trials in which the start position was varied pseudorandomly, but with the platform remaining in the same place. However, in contrast to the place-learning protocol, the location of the platform was varied in a pseudorandom order among the 10 testing sessions, and the 8 locations of the platform were selected so that it would not be easy for the rats to encounter them accidentally from the start positions. As before, the ITI was 60 s, with the rat swimming for a maximum of 45 s on each trial and being allowed to remain on the platform for 10 s before being returned to the holding cage. In this version of the task, each of the testing sessions can be considered a separate "problem" in which the first of the two trials is a search trial and the second a test trial that measures the rat's ability to remember the platform location from the immediately preceding trial. A significantly shorter distance (and latency) on the second trial is considered evidence of working memory.
Cued version (wk 3). After a further 2-d interval, the rats were tested for another 2 d, morning and afternoon, on the cued version of the maze. Under this condition, the tank was surrounded by a white curtain to obscure extra-maze cues, and the location of the platform was indicated by the attachment of a 10 cm high, 1.5 cm diameter pole, painted with black and white horizontal stripes, that served as a prominent visual cue. The location of the platform remained the same within sessions, but varied between testing sessions. Each session consisted of four trials, with the order of the four starting positions determined pseudorandomly as in the place version. Differences in performance on this task would be indicative of alterations in the sensory, motor or motivational attributes of the animals.
Brain FA composition.
At the conclusion of the behavioral testing, the rats were killed by an overdose of Halothane.
The brains were removed and stored at -80°C. Fatty acids in the
forebrain were extracted using the method of Folch et al. (1957)
. Aliquots of total tissue lipid extracts were separated
into different phospholipid fractions by thin layer chromatography with
chloroform/methanol/water/triethylamine (4:5:1:4, v/v/v/v) as the
developing system. The FA in the phosphatidylcholine (PC),
phosphatidylethanolamine (PE) and phosphatidylinositol + phosphatidylserine (PI + PS) fractions were methylated under
nitrogen according to the method of Morrison and Smith (1964)
. Heptadecaenoic acid was added as the internal standard.
Fatty acid methyl esters were analyzed by capillary gas chromatography
(Hewlett-Packard 5890 11 Plus, Hewlett-Packard, Palo Alto, CA)
equipped with a 30-m (0.32-mm i.d.) capillary column (Omegawax,
0.25-µm film thickness, Supelco, Bellefonte, PA) and
integrated by a Hewlett-Packard Chem Station. The oven temperature
was programmed to increase from 120 to 200°C at 4°C/min and to hold
a final 15 min. The identification of each fatty acid was made with
authentic standard mixtures (Nu-Chek-Prep, Elysian, MN).
Statistical analysis.
The data were analyzed by using SAS version 6.0.9 (SAS Institute, Cary,
NC) to perform ANOVA. Brain fatty acid composition was analyzed using
preplanned (a priori) comparisons to address the following:
1) the main effects of DHA and AA; 2) the
interaction between DHA and AA (with significant interactions
interpreted by further analysis of the simple main effects); and
3) the difference between the two control groups, i.e.,
the suckled group and the saturated fat group (positive control).
Tukey's test was used to conduct further post-hoc comparisons
among all of the groups. The place and cued versions of the
water-maze were analyzed by a two-factor ANOVA, with Diet as a
between-groups factor and Days as a within-groups (repeated)
factor. For the working-memory task, the analyses were conducted by
first averaging, for each rat, the difference between the score on
Trial 1 and Trial 2 across the 10 problems, and then analyzing these
difference scores with the use of ANOVA and planned comparisons as
described above, followed by post-hoc Tukey's tests on all groups.
The significance level was set at = 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There were no significant differences among the groups in body weight
at 6 wk, with a range of 189–198 g (data not shown). Forebrain fatty
acid composition of the PE, PC and PI + PS fractions are shown in
Tables 3
,
4 and
5, respectively; the data for selected 20- and 22-carbon FA for the
PE fraction are depicted graphically in Figure 1
. The tables also present a summary of the statistical analyses,
which show that the general pattern of results for the long-chain
compounds appeared to be consistent across all phospholipid fractions.
Generally, these were interactive, with the effects of AA depending on
the level of DHA, and the effects of DHA depending on the level of AA.
For example, DHA decreased 20:4 (n-6), and AA increased 20:4 (n-6), but
only at 2.5% DHA. A similar trend was seen for 22:4 (n-6). DHA
decreased 22:5 (n-6), and AA increased it only at 0% DHA. DHA
increased 22:6 (n-3) and AA decreased 22:6 (n-3), with a larger effect
of AA at 0% DHA. Differences between the suckled and saturated fat
groups with respect to 20:4 (n-6) and 22:4 (n-6), if present, were
small. However, there were large differences between these groups with
respect to 22:5 (n-6) and 22:6 (n-3). These were reciprocal in size and
direction, such that the saturated fat group had very high levels of
22:5 (n-6) and very low levels of 22:6 (n-3). In comparing the overall
FA profile, including the (n-6)/(n-3) ratio, the artificially reared
group that was the most similar to the suckled control was the
unsupplemented group, i.e., 10% LA and 1% LNA with 0% AA and 0%
DHA.
|
|
|
|
On the acquisition of place learning, shown in Figure 2A
, there was a main effect of Days, F (3, 345) = 208.07,
P > 0.001, with all groups swimming shorter
distances over days to locate the platform. There was no effect of
Diet, F (5, 115) = 0.8073, P > 0.05, nor a
Diet x Days interaction, F (15, 345) = 0.44,
P > 0.05 on distance swum. Swimming speed tended
to increase over days. F (3, 345) = 2.45, P < 0.07, but there was no effect of Diet, F (5, 115) = 1.09,
P > 0.05, nor a Diet x Days interaction, F
(5, 345) = 0.62, P > 0.05 (data not shown). Thus,
the effects on latency to find the platform (data not shown) were
similar to those on distance swum. On the probe trial, there was a main
effect of Quadrant, F (2, 230) = 46.94, P < 0.001
but no effect of Diet, F (5, 115) = 0.31, P > 0.05, nor a Diet x Quadrant interaction. As seen in Figure 2
B, the distance swum by all groups in the target
quadrant was greater than that swum in the other quadrants in all
groups. There were no significant differences on the cued trials in
distance swum across Days, F (1, 115) = 0.5439, P
> 0.05, nor an effect of Diet, F (5, 115) = 0.5462, nor a
Diet x Days interaction, F (1, 115) = 0.5830, P
>0.05. These data are shown in Figure 2
C.
|
The data on working memory are presented as difference scores in
Figure 3
, with the score representing the improvement in performance from
Trial 1 to Trial 2. Because one concern with a difference score is that
the same score could be achieved in different ways, preliminary
analyses were done on Trial 1 to ascertain whether the groups were
performing at the same level on the search trial, and this was indeed
the case (data not shown). With respect to the difference scores, the
suckled control group and the saturated fat group were the highest and
lowest scoring groups, respectively, and the a priori comparison
between these two groups was significantly different F(1, 115) = 4.44,
P < 0.05. Although the improvement also appeared
to be somewhat lower in both groups receiving 2.5% AA, compared with
those receiving none, the main effect of AA was not significant, F (1,
115) = 0.61, P > 0.05, nor was there a main effect
of DHA, F (1, 115) = 0.13, p P > 0.05 or an
interaction, F (1, 115) = 0.10, P > 0.05.
Post-hoc comparisons among all the groups using Tukey's
t test did not support any significant differences.
Further correlational analyses were conducted between the rats' brain
levels of selected FA and the behavioral improvement score in
individual rats. Because the suckled controls differed from the other
groups not only with respect to diet but also rearing environment,
these latter analyses were conducted only on the rats in the
artificially reared groups. The data for the AA and DHA levels, and the
(n-6)/(n-3) ratio in the PE fraction are shown in Figure 4
A,
B and
C), respectively. There was no relationship apparent between the fatty
acid levels and the behavioral score. The same was true of the PC and
PI + PS fractions (data not shown).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The groups did not differ in their performance on
the place and cued versions of the water-maze. The only significant
difference among the groups with respect to working memory was that
predicted a priori between the suckled control and the positive control
(saturated fat) groups, with the saturated fat group showing poorer
performance. Furthermore, despite the large sample size, there were no
significant correlations in the artificially reared rats between levels
of AA, DHA or the (n-6)/(n-3) ratio in the various phospholipid
fractions and the behavioral measure.
The advantage of the artificial rearing method is that it allows
precise control over the delivery of nutrients to the rat pup during
the time when its brain is growing most rapidly. The timing of this
brain growth spurt varies among mammalian species, occurring
postnatally in the rat, but during the third trimester in human infants
(Dobbing 1990
). In contrast to our previous study in
which the FA composition of the brain was measured at 18 d of age,
the rats in this study were maintained on diets of similar oil
composition until the conclusion of the behavioral testing, when they
were 9 wk. On the basis of the previous results (Ward et al. 1998
), the level of supplementation of AA and DHA that was
implemented in this study (2.5%) was expected to produce a wide range
of FA values. This was supported by the results, which show that when
expressed as a percentage of the unsupplemented group (0% AA and DHA),
AA in forebrain PE varied between 84 and 103%, and DHA varied between
86 and 119% (compared with the previous values of 86–110% and
82–142%). The previous study also included an intermediate level of
supplementation (0.5% AA and DHA), and it was this group that most
closely resembled the suckled controls in long-chain PUFA levels,
with the unsupplemented group (0% AA and DHA) being somewhat lower. It
should be noted that in this study, the suckled control group was very
similar to the unsupplemented group, with AA values of 99% and DHA
values of 105% of those of the unsupplemented group. Because both
groups were weaned onto diets that did not contain AA and DHA (suckled
group fed AIN 93G semipurified diet, which contains soybean oil), the
capacity in both groups to manufacture the necessary long-chain
PUFA during the postweaning period is supported. An additional feature
of this design was the inclusion of the saturated fat group as a
positive control for the behavioral measures. This group was fed oils
containing sufficient LA to prevent growth retardation (preweaning
2.5% LA, postweaning 3.0% LA), and AA levels in the brain were
similar to those of the unsupplemented control (100%); however, DHA
levels were much lower (64%). These low DHA levels, accompanied by a
reciprocal increase in 22:5 (n-6), are consistent with our previous
work in mice fed a diet high in saturated fat (Wainwright et al. 1994a
).
It is important to note that in this study, the effects of supplementation of AA and DHA could be evaluated against a background of sufficient precursor (n-6) and (n-3) FA in the diet. Moreover, by providing a range of brain FA levels, the nature of the relationship between brain FA composition and behavior could be assessed directly. In contrast, a behavioral difference in a design using only two groups, for example (n-3) deficient and adequate, is more difficult to interpret because the diet could be affecting a wide range of biochemical functions other than those related to DHA or AA levels. Furthermore, although the effects of extreme deficiency and excess are important, a thorough understanding of the role of FA in brain function must include consideration of the dose-response relationship which, as in this study, encompasses effects within the normal range. The results indicate that despite the large variation in brain FA composition across the artificially reared groups, none of these groups differed significantly on behavioral performance in the water-maze. Although supplementation with AA appeared to be associated with slightly lower scores on the measure of working memory, this was not supported by the factorial analysis or by post-hoc comparisons. The correlations did not support a relationship between the FA composition of forebrain phospholipids and behavioral performance. The only significant difference on the working-memory task was that predicted a priori between the control groups, with the saturated fat group being less proficient than the suckled control group. It is important to bear in mind, however, that this effect could be due as much to rearing as to dietary condition. Comparison of each of these groups with the unsupplemented group can be informative in this regard because the unsupplemented group was similar to the suckled control with respect to brain FA composition (discussed above), and it shared rearing conditions with the saturated fat group. Interestingly, although the differences were not significant, the value of the working-memory score of the unsupplemented group was between those of the two control groups, suggesting that both rearing and dietary treatment were contributory factors.
The relationship between dietary PUFA, brain DHA and AA levels
and functional development is of particular interest in the case of
formula-fed premature infants, in whom AA and DHA would have
normally been provided through the placenta. Several studies with
preterm infants randomly assigned to receive standard or supplemented
formulas have been published. Studies with preterm infants fed formulas
supplemented with DHA from marine oil report higher electroretinogram
responses at 36 wk postconception (Birch et al. 1992
)
and transient increases in visual acuity (2 and/or 4 mo)
(Carlson et al. 1993a and 1996
). Similar studies with
formulas containing both DHA and AA are not currently available. There
are also studies in term infants that have shown effects on visual and
developmental outcomes of supplementing formula with DHA or both AA and
DHA (Birch et al. 1998
, Makrides et al. 1996
), although these findings are not consistent
(Auestad et al. 1997
). The relationship between the
effects seen on visual function and overall cognitive development
remains controversial (discussed in Dobbing 1997
). Some
studies of preterm infants do report some differences associated with
dietary supplementation on Bayley Scales of Infant Development
(Carlson et al.1994
), as well as measures of looking
behavior on the Fagan Test of Infant Intelligence (Carlson and Werkman 1996
, Werkman and Carlson 1996
); these
effects are similar to those reported in monkeys fed
(n-3)–supplemented, compared with (n-3)–deficient formula
(Reisbick et al. 1997
). There is also a report in term
infants of improvements on the Brunet-Lézine test of psychomotor
development at 4 mo (Agostoni et al. 1995
), but the
dietary groups no longer differed in a follow-up study at 24 mo
(Agostoni et al. 1997
). A recently published study by
Willatts et al. (1998)
showed improved problem solving
in 10-mo-old term infants fed diets supplemented with AA and DHA
compared with infants fed an unsupplemented formula that had a high
(n-6)/(n-3) ratio and very low levels of - linolenic acid.
Where does this leave us with respect to the question of a
putative role for (n-3) FA in cognitive performance? Maybe what is
needed here is for us to reexamine the basic assumptions that inform
our experimental approach when we try to determine the relationship
between nutritional interventions and cognitive outcomes. One approach
to such questions is based on the premise that if we can show that
dietary manipulations have an effect on biochemical measures that we
know to be associated with alterations in physiologic functions in the
brain, then these must necessarily be implicated in behavioral changes.
For example, it was reported recently that rats fed diets restricted in
(n-3) FA showed decreased dopamine levels and decreased
D2 receptor binding in frontal cortex
(Delion et al. 1994
). This led Reisbick and Neuringer (1997)
to postulate that (n-3) FA deficiency might be
associated with alterations in the function of the mesocorticolimbic
pathway. Our current understanding of the role of this pathway in
behavioral functions would thus lead us to expect changes associated
with dopamine release in the nucleus accumbens in the midbrain, as well
as in prefrontal cortex. On the other hand, the considerable redundancy
seen in various systems in the brain suggests that it might be possible
for the brain to maintain normal function despite considerable
biochemical variation. In Parkinson's disease, for example, only when
the loss of dopaminergic neurons in the substantia nigra exceeds
70–80% do the overt motor symptoms become apparent (Agid et al. 1987
). Thus, the fact that we have been unable to show
effects on a task related to working memory in rats in this study,
despite the wide range of FA levels, may be an indication of such
plasticity. It appears that there is an optimum range of activity in
different neurochemical systems that supports normal function, and it
is only when these homeostatic limits are challenged by extreme dietary
manipulations, disease, or possibly aging, that symptoms will become
evident. Furthermore, on the basis of research on animal cognition, it
is becoming recognized increasingly that there are different systems of
learning and memory, and that a manipulation that impairs performance
on one type of learning task may have no effect on another. In studies
done in rats, for example, excitotoxic lesions of the hippocampus
impaired aversive contextual conditioning, but left aversive
conditioning to an explicit conditioned stimulus intact (Everitt and Robbins 1997
). Findings such as these caution against the
expectation that manipulation of neural systems, either by drugs or
nutrition, will necessarily be manifest in changes in behavior, and,
should such behavioral changes indeed be present, that interpretation
will be simple. They also emphasize the importance of basing dietary
recommendations, not on any one behavioral outcome, such as
"learning ability," but rather on a clear understanding of the
basic neural mechanisms involved, derived from the interrelationships
among a variety of biochemical and behavioral measures.
In summary, by feeding artificially reared rat pups diets with adequate levels of LA and LNA, supplemented with AA and DHA as microbial oils, we varied levels of AA and DHA in forebrain phospholipids over a wide range. We did not, however, show a relationship between brain fatty acid composition and performance on a working-memory task in the Morris water-maze.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by funds provided by Ross Laboratories
together with matching funds from the Natural Sciences and Engineering
Research Council of Canada through a Collaborative Research and
Development grant. The single cell oils were provided by Martek
Biosciences Corporation.
4 Abbreviations used: ARA, arachidonic acid; EFA, essential
fatty acids; EPA, eicosapentaenoic acid; FA, fatty acids; LA, linoleic
acid; LNA, linolenic acid: LCPUFA, long-chain polyunsaturated fatty
acid; MCT, medium-chain triglyceride; PC, phosphatidylcholine; PE,
phosphatidylethanolamine; PI, phosphatidylinositol; PS,
phosphatidylserine; PUFA, polyunsaturated fatty acid.
Manuscript received July 10, 1998. Initial review completed November 5, 1998. Revision accepted January 28, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.
Agostini C., Trojan S., Bellu R., Riva E., Bruzzese M. G., Giovannini M. Developmental quotient at 24 months and fatty acid composition of diet in early infancy: a follow up study. Arch. Dis. Child. 1997;76:421-424
2. Agostoni C., Trojan S., Bellu R., Riva E., Giovannini M. Neurodevelopmental quotient of healthy term infants at 4 mos. and feeding practice: the role of long-chain polyunsaturated fatty acids. Pedatr. Res. 1995;38:262-266[Medline]
3. Agid Y., Javoy-Agid F., Ruberg M. Biochemistry of neurotransmitters in Parkinson's disease. Mov. Disord. 1987;2:166-230
4. Auestad N., Montalto M. B., Hall R. T., Fitzgerald K. M., Wheeler R. E., Connor W. E., Neuringer M., Connor S. L., Taylor J. A., Hartmann E. E. Visual acuity, erythrocyte fatty acid composition, and growth in term infants fed formulas with long chain polyunsaturated fatty acids for one year. Ross Pediatric Lipid Study. Pediatr. Res. 1997;41:1-10[Medline]
5.
Birch E. E., Birch D. G., Hoffman D. R., Uauy R. Dietary essential fatty acid supply and visual acuity development. Invest. Opthalmol. Vis. Sci. 1992;33:3242-3253
6. Birch E. E., Hoffman D. R., Uauy R., Birch D. G., Prestidge C. Visual acuity and the essentiality of docosahexaenoic acid and arachidonic acid in the diet of term infants. Pediatr. Res. 1998;44:201-209[Medline]
7. Carlson S. E., Cooke R. J., Werkman S. H., Tolley E. A. First year growth of preterm infants fed standard compared to marine oil (n-3) supplemented formula. Lipids 1992;27:901-907[Medline]
8. Carlson S. E., Werkman S. H. A randomized trial of visual attention of preterm infants fed docosahexaenoic acid until two months. Lipids 1996;31:85-90[Medline]
9.
Carlson S. E., Werkman S. H., Peeples J. M., Cooke R. J., Tolley E. A. Arachidonic acid staus correlates with first year growth in preterm infants. Proc. Natl. Acad. Sci. U.S.A. 1993;90:1073-1077
10. Carlson S. E., Werkman S. H., Peeples J. M., Cooke R. J., Tolley E. A. Growth and development of premature infants in relation to omega 3 and omega 6 fatty acid status. World Rev. Nutr. Diet 1994;75:3-69
11.
Carlson S., Werkman S. H., Rhodes P. G., Tolley E. A. Visual acuity development in healthy preterm infants: effects of marine oil supplementation. Am. J. Clin. Nutr. 1993;58:35-42
12.
Carlson S. E., Werkman S. H., Tolley E. A. Effect of long-chain (n-3) fatty acid supplementation on visual acuity and growth of preterm infants with and without bronchopulmonary dysplasia. Am. J. Clin. Nutr. 1996;63:687-697
13. Connor W. E., Neuringer M., Reisbick S. Essential fatty acids: the importance of n-3 fatty acids in the retina and brain. Nutr. Rev. 1992;50:21-29[Medline]
14. Cook H. W. Fatty acid desaturation and chain elongation in eucaryotes. Vance D. E. Vance J. eds. Biochemistry of Lipids, Lipoproteins and Membranes 1991:141-169 Elsevier Amsterdam, The Netherlands.
15.
Delion S., Chalon S., Herault J., Guilloteau D., Besnard J.-C., Durand G. Chronic dietary -linolenic acid deficiency alters dopaminergic and serotoninergic neurotransmission in rats. J. Nutr. 1994;124:2466-2476
16. Dobbing J. eds. Developing Brain and Behavior: The Role of Lipids in Infant Formula 1997 Academic Press London, England.
17. Everitt B. J., Robbins T. W. Central cholinergic systems and cognition. Annu. Rev. Psychol. 1997;48:649-684[Medline]
18.
Folch J., Lees M., Sloane-Stanley G. H. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 1957;226:497-509
19. Galli C., Trzeciak J. B., Paoletti R. Effects of dietary fatty acids on the fatty acid composition of brain ethanolamine phosphoglyceride: reciprocal replacement of n-6 and n-3 polyunsaturated fatty acids. Biochim. Biophys. Acta 1971;248:449-454
20. Greenwood C. E., Winocur G. Cognitive impairment in rats fed high fat diets: a specific effect of saturated fatty acid intake. Behav. Neurosci. 1996;110:451-459[Medline]
21. Innis S. Essential fatty acids in growth and development. Prog. Lipid Res. 1991;30:39-103[Medline]
22. Makrides M., Neumann M. A., Gibson R. A. Is dietary docosahexaenoic acid essential for term infants?. Lipids 1996;31:115-119[Medline]
23. Mohrhauer H., Holman R. T. Alteration of the fatty acid composition of brain lipids by varying levels of dietary essential fatty acids. J. Neurochem. 1963;10:523-530[Medline]
24. Morrison W. R., Smith L. M. Preparation of fatty acid methyl esters and dimethylacetals from lipids with boron trifluoride-methanol. J. Lipid Res. 1964;5:600-608[Abstract]
25. Reeves P. G., Nielsen F. H., Fahey G. C., Jr AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 1993;123:1939-1951
26. Reisbick S., Neuringer M. Omega-3 fatty acid deficiency and behavior: a critical review and directions for future research. Yehuda S. Mostofsky D. I. eds. Handbook of Essential Fatty Acid Biology: Biochemistry, Physiology and Behavioral Neurobiology 1997:397-426 Humana Press Totowa, NJ.
27. Reisbick S., Neuringer M., Gohl E., Wald R. Visual attention in infant monkeys: effects of dietary fats and age. Dev. Psychol. 1997;33:387-395[Medline]
28.
Saste M. D., Carver J. D., Stockard J. E., Benford V. J., Chen L. T., Phelps C. P. Maternal diet fatty acid composition affects neurodevelopment in rat pups. J. Nutr. 1998;128:740-743
29. Sastry P. S. Lipids of nervous tissue: composition and metabolism. Prog. Lipid Res. 1985;24:69-176[Medline]
30. Stewart C. A., Morris R. G. The water maze. Saghal A. eds. Behavioural Neuroscience: A Practical Approach 1993;Vol. 1:107-122 Oxford University Press New York, NY.
31. Strupp B. J., Levitsky D. A. Enduring cognitive effects of early malnutrition: a theoretical appraisal. J. Nutr. 1995;125:2221S-2232S
32. Tinoco J., Babcock R., Hincenbergs I., Medwadowski B., Miljanich P. Linolenic acid deficiency: changes in fatty acid patterns in female and male rats raised on a linolenic acid-deficient diet for two generations. Lipids 1978;13:6-17[Medline]
33. Wainwright P. E. Essential fatty acids and behaviour: is there a role for the eicosanoids?. Yehuda S. Mostofsky D. I. eds. Handbook of Essential Fatty Acid Biology: Biochemistry, Physiology and Behavioral Neurobiology 1997:299-341 Humana Press Totowa, NJ.
34. Wainwright P. E., Huang Y. S., Bulman-Fleming B., Lévesque S., McCutcheon D. The effects of dietary fatty acid composition combined with environmental enrichment on brain and behavior in mice. Behav. Brain Res. 1994;60:125-136[Medline]
35. Wainwright P. E., Huang Y. S., Coscina D. V., Lévesque S., McCutcheon D. Brain and behavioral effects of dietary n-3 deficiency in mice: a three generation study. Dev. Psychobiol. 1994;27:467-487[Medline]
36. Wainwright, P. E., Mutsaers, L. M., Jalali, E. & Bell, R. (1996) A comparison of the effects on behavioral development in mice of a diet with a low (n-3): (n-6) ratio with those of malnutrition. AOCS Conference on PUFA in Infant Nutrition: Consensus and Controversies, Barcelona, Spain, November 1996.
37. Wainwright P. E., Ward G. R. Early nutrition and behavior: a conceptual framework for critical analysis of research. Dobbing J. eds. Developing Brain and Behavior: The Role of Lipids in Infant Formula 1997:387-425 Academic Press London, England.
38. Wainwright P. E., Xing H.-C., Girard T., Parker L., Ward G. R. Effects of dietary (n-3) fatty acid deficiency on Morris water-maze performance and amphetamine-induced conditioned place preference in rats. Nutr. Neurosci. 1998;1:281-293
39.
Wainwright P. E., Xing H.-C., Mutsaers L., McCutcheon D., Kyle D. Arachidonic acid offsets the effects on mouse brain and behavior of a diet with a low (n-6):(n-3) ratio and very high levels of docosahexaenoic acid. J. Nutr. 1997;127:184-193
40.
Ward G. R., Huang Y.-S., Bobik E., Xing H.-C., Mutsaers L., Auestad N., Montalto M., Wainwright P. E. Long-chain polyunsaturated fatty acid levels in formulae influence deposition of docosahexaenoic acid and arachidonic acid in brain and red blood cells of artificially reared neonatal rat pups. J. Nutr. 1998;128:2473-2487
41. Ward G. R., Wainwright P. E. The contribution of animal models to understanding the role of fats in infant nutrition. Huang Y.-S. Sinclair A. J. eds. Recent Advances in Roles of Lipids in Infant Nutrition 1998:39-62 AOCS Press Champaign, IL.
42. Weisinger H. S., Vingrys A. J., Sinclair A. J. The effect of docosahexaenoic acid on the electroretinogram of the guinea pig. Lipids 1996;31:65-70[Medline]
43. Werkman S. H., Carlson S. E. A randomized trial of visual attention of preterm infants fed docosahexaenoic acid until nine months. Lipids 1996;31:91-97[Medline]
44. Whishaw I. Q. A comparison of rats and mice in a swimming pool place task and matching to place task: some surprising differences. Physiol. Behav. 1995;58:687-693[Medline]
45. Willats P., Forsyth J. S., DiMondugno M. K., Varma S., Colvin M. Effect of long-chain polyunsaturated fatty acids in infant formula on problem solving at 10 months of age. Lancet 1998;352:688-691[Medline]
This article has been cited by other articles:
|
|
 |
|
  P. Kaur, K. Schulz, M. Aschner, and T. Syversen Role of Docosahexaenoic Acid in Modulating Methylmercury-Induced Neurotoxicity Toxicol. Sci., December 1, 2007; 100(2): 423 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Kan, E. Melamed, D. Offen, and P. Green Docosahexaenoic acid and arachidonic acid are fundamental supplements for the induction of neuronal differentiation J. Lipid Res., March 1, 2007; 48(3): 513 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C McCann and B. N Ames Is docosahexaenoic acid, an n-3 long-chain polyunsaturated fatty acid, required for development of normal brain function? An overview of evidence from cognitive and behavioral tests in humans and animals Am. J. Clinical Nutrition, August 1, 2005; 82(2): 281 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. LYNCH Long-Term Potentiation and Memory Physiol Rev, January 1, 2004; 84(1): 87 - 136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-Y. Lim and H. Suzuki Changes in Maze Behavior of Mice Occur after Sufficient Accumulation of Docosahexaenoic Acid in Brain J. Nutr., February 1, 2001; 131(2): 319 - 324. [Abstract] [Full Text]
|
|
|
|
|
|
I. P. M. Wauben, H.-C. Xing, and P. E. Wainwright Neonatal Dietary Zinc Deficiency in Artificially Reared Rat Pups Retards Behavioral Development and Interacts with Essential Fatty Acid Deficiency to Alter Liver and Brain Fatty Acid Composition J. Nutr., October 1, 1999; 129(10): 1773 - 1781. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||
