Modulation of Learning, Pain Thresholds, and Thermoregulation in the Rat by Preparations of Free Purified alpha-Linolenic and Linoleic Acids: Determination of the Optimal omega3-to-omega6 Ratio

Abstract

Ingested polyunsaturated fatty acids are postulated to lead to changes in central nervous system activity, presumably by altering the lipid composition of neuronal membranes. In support of this hypothesis, we and other investigators have previously demonstrated cognitive effects in rats fed oils that contain both alpha-linolenic acid (18:3omega3) and linoleic acid (18:2omega6), with the relative content of alpha-linolenic acid being seen as the critical variable. The present study in rats examined the effects of preparations containing different ratios of highly purified free alpha-linolenic acid to linoleic acid (about 25 mg/kg of body weight daily) on learning performance (Morris water tank), pain thresholds (heated plate), and thermoregulatory control of d-amphetamine-induced hypothermia during 4 weeks of treatment. Preparations with omega3-to-omega6 ratios ranging from 1:3.5 to 1:5 (specifically a ratio of 1:4) produced significant favorable effects on all of these variables. Although the specific mode of action remains to be elucidated, these results suggest that such preparations of free fatty acids should be evaluated in the treatment of memory disorders and pain conditions.

Full text

Proc.
Natl.
Acad.
Sci.
USA
Vol.
90,
pp.
10345-10349,
November
1993
Neurobiology
Modulation
of
learning,
pain
thresholds,
and
thermoregulation
in
the
rat
by
preparations
of
free
purified
a-linolenic
and
linoleic
acids:
Determination
of
the
optimal
w3-to-o6
ratio
SHLOMO YEHUDA*
AND
RALPH
L.
CARASSOt
*Psychophalmacology
Laboratory,
Department
of
Psychology,
Bar-Ilan
University,
Ramat-Gan,
52900,
Israel;
and
tNeurological
Unit,
Hillel
Yaffe
Hospital,
Hadera,
Israel
Communicated
by
Ralph
T.
Holman,
July
2,
1993
(received
for
review
September
2,
1992)
ABSTRACT
Ingested
polyunsaturated
fatty
acids
are
pos-
tulated
to
lead
to
changes
in
central
nervous
system
activity,
presumably
by
altering
the
lipid
composition
of
neuronal
mem-
branes.
In
support
of
this
hypothesis,
we
and
other
investigators
have
previously
demonstrated
cognitive
effects
in
rats
fed
oils
that
contain
both
a-linolenic
acid
(18:3w3)
and
linoleic
acid
(18:2o6),
with
the
relative
content
of
a-lnolenic
acid
being
seen
as
the
critical
variable.
The
present
study
in
rats
examined
the
effects
of
preparations
containing
different
ratios
of
highly
purified
free
a-linolenic
acid
to
linoleic
acid
(about
25
mg/kg
of
body
weight
daily)
on
larning
performance
(Morris
water
tank),
pain
thresholds
(heated
plate),
and
thermoregulatory
control
of
d-amphetaminduced
hypothermia
during
4
weeks
of
treatment.
Preparations
with
w3-to-w6
ratios
ranging
from
1:3.5
to
1:5
(ificaly
a
ratio
of
1:4)
produced
sinfcant
favorable
effects
on
all
of
these
variables.
Although
the
specific
mode
of
action
remains
to
be
elucidated,
these
results
suggest
that
such
preparations
of
free
fatty
acids
should
be
evaluated
in
the
treatment
of
memory
disorders
and
pain
conditions.
Previous
evidence
suggests
that
certain
biological
constitu-
ents,
when
administered
in
pure
form
or
ingested
in
food,
can
function
as
drugs.
They
may
induce
changes
in
the
chemical
composition
of
structures
in
the
brain
and
consequently
modify
brain
activity
in
experimental animals
(1,
2).
For
example,
it
has
been
hypothesized
that
the
mode
of
action
of
tryptophan,
tyrosine,
and
choline
involves
their
role
as
precursors
for
brain
neurotransmitters
(1).
The
ratio
between
the
level
of
tryptophan
and
large
neutral
amino
acids
(the
total
of
tyrosine,
phenylalanine,
leucine,
isoleucine,
and
valine)
in
the
plasma
is
reported
to
be
a
critical
determinant
of
brain
tryptophan
bioavailability
(1).
Changes
in
the
levels
of
these
amino
acids
in
the
central
nervous
system
are
postulated
to
induce
changes
in
the
functional
activity
of
brain
neurotransmitters
and
consequently
in
behavior.
However,
there
are
several
observations
regarding
other
types
of
food
components
that
cannot
be
satisfactorily
ac-
counted
for
by
such
an
explanation-e.g.,
the
cognitive
effects
of
soybean
oil
and
the
regional
decrease
in
the
level
of
cholesterol
after
learning
(3,
4).
We
have
also
previously
proposed
(2,
4-6)
that
diet-induced
changes
in
the
lipid
composition
of
neuronal
membranes
may
mediate
the
ob-
served
changes
in
learning
and
behavior.
It
should
be
em-
phasized
that
this
hypothesis,
the
neuronal
membrane
func-
tional
modification
hypothesis
(2),
does
not
contradict
but
supplements
the neurotransmitter
precursor
hypothesis
(1).
Several
researchers
have
examined
the
effects
of
various
oils
in
the
diet
on
brain
development,
brain
biochemistry,
and
behavior
(7-11).
Most
of
the
studies
showed
that
oils
con-
taining
either
a-linolenic
acid
(18:3w3)
(such
as
soybean
oil
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.
and
perilla
oil)
or
docosahexanoate
(22:6c3)
had
beneficial
effects
on
various
types
of
learning
(7-11).
These
studies
are
consistent
with
our
own
observations
that
rats
fed
a
soybean
source
lipid
diet
exhibited
a
significantly
improved
capacity
in
an
environmentally
cued
testing
paradigm
(4-6).
In
addi-
tion,
they
exhibited
a
higher
pain
threshold
and
were
pro-
tected
from
d-amphetamine-induced
hypothermia
when
ex-
posed
to
an
ambient
temperature
of
4°C.
In
contrast,
rats
fed
a
lard
or
sunflower
source
diet
did
not
differ
from
rats
fed
a
control
(Chow)
diet.
None
of
the
diets
induced
changes
in
the
level
of
motor
activity
(4-6).
The
initial
hypothesis
that
attempted
to
explain
these
results
focused
on
the
amount
of
polyunsaturated
fatty
acids
(PUFAs)
in
soybean
oil.
However,
sunflower
oil,
which
contains
a
higher
level
of
PUFAs
than
soybean
oil,
failed
to
produce
the
positive
effects
of
soybean
oil
(5, 6).
Since
soybean
oil
contains
a
considerably
higher
level
of
a-linolenic
acid
(8-9%)
than
sunflower
oil
(about
0.4%),
we
postulated
that
the
relative
quantity
(ratio)
of
a-linolenic
acid
to
linoleic
acid
(18:2w6),
rather
than
the
absolute
quantities
of
the
fatty
acids,
was
the
critical
factor
for
brain
bioavailability
and
the
central
nervous
system-mediated
effects.
Whereas
the
importance
of
linoleic
acid
for
normal
health
as
well
as
for
brain
development
and
the
maintenance
of
normal
brain
function
had
already
been
demonstrated
(12),
the
biological
effects
of
a-linolenic
acid
[which
is
also
tradi-
tionally
classified
as
an
essential
fatty
acid
(EFA)]
are
only
recently
becoming
clarified.
The
selective
enrichment
of
elongated
w3
fatty
acids
in
the
retina
and
the
brain
and
their
relative
resistance
to
depletion
has
puzzled
many
investiga-
tors.
Experiments
with
14C-labeled
fatty
acids
have
shown
a
preferential
brain
uptake
of
a-linolenic
over
linoleic
acid
(13).
Early
studies
also
suggested
that
a-linolenic
acid
may
have
a
biochemical
function
distinct
from
that
of
linoleic
acid,
because
decreases
in
5'-mononucleotidase
activity
in
the
brain,
caused
by
lipid
deprivation,
could
be
normalized
only
by
linolenic
acid
supplementation
(14).
Although
there
were
until
now
only
a
few
clinical
reports
of
a-linolenic
acid
deficiency
(15-17),
experiments
in
monkeys
and
rats
have
shown
visual
and
learning
impairment
after
consumption
of
diets
deficient
in
cO3
fatty
acids
(18-20).
These
studies
prompted a
recent
surge
of
interest
in
the
role
of
a-linolenic
acid
and
its
derivatives
in
brain
development,
brain
and
retinal
function,
and
maintenance
of
normal
well-being
(21-
33).
The
aims
of
this
study
were
to
test
the
hypothesis
that
the
ratio
of
a-linolenic
to
linoleic
acid
is
a
key
factor
in
mediating
the
beneficial
effects
of
PUFAs
and
to
identify
the
optimal
ratio
of
these
free
fatty
acids.
To
avoid
changes
in
the
percentage
of
fatty
acid
in
commercially
available
oils
and
to
exclude
the
confounding
effects
of
other
fatty
acids
or
lipids,
the
test
materials
were
prepared
from
highly
purified
free
Abbreviations:
PUFA,
polyunsaturated
fatty
acid;
EFA,
essential
fatty
acid.
10345

10346
Neurobiology:
Yehuda
and
Carasso
Table
1.
Nutritional
factors
Group
Food
intake,
kcal
Weight
gain,
g
A
2565
±
39
237
±
4.7
B
2575
±
80 230
±
7.0
C
2545
±
75
235
±
2.8
D
2534
±
68
237
±
4.6
E
2543
±
72
239
±
6.1
F
2562
±
57
235
±
3.3
G
2586
+
48
238
±
3.9
H
2533
±
61
234
±
5.5
The
values
given
are
the
means
±
SEM
from
nine
rats
per
group.
The
unsaturated
fatty
acid
treatments
had
no
effect
on
the
amount
of
food
intake
(kcal)
or
on
the
rate
of
body
weight
gain.
The
data
were
calculated
at
the
end
of
treatment.
Group
A,
0.9%o
NaCl;
groups
B-H,
a-linolenic
acid-to-linoleic
acid
ratios
of
1:3,
1:3.5,
1:4,
1:4.5,
1:5,
1:5.5,
and
1:6,
respectively.
linoleic
and
linolenic
acids.
We
tested
the
effects
of
mixtures
with
various
ratios
on
learning,
motor
activity,
pain
thresh-
old,
and
thermoregulation
in
adult
male
rats.
MATERIALS
AND
METHODS
Test
Material.
a-Linolenic
(0.92
g/cm3)
and
linoleic
(0.90
g/cm3)
free
fatty
acids,
both
=99%
pure
(as
evaluated
by
capillary
gas
chromatography),
were
purchased
from
Sigma
(L2367
and
L1376).
The
test
substances
were
stored
at
4°C
in
the
dark.
A
stock
solution
(1
ml)
containing
the
two
fatty
acids
(0.25
ml),
mineral
oil
(0.73
ml),
and
a-tocopherol
(0.02
ml)
was
prepared
every
3
days.
Seven
different
stock
solu-
tions
with
different
ratios
of
the
two
fatty
acids
(see
Exper-
imental
Design
below)
were
used
in
the
experiment.
Animals.
Male
Long
Evans
hooded
rats
[1
month
of
age
and
-100
g
(range,
90-110
g)]
were
purchased
from
local
breed-
ers.
They
were
housed
individually
in
hanging
stainless
steel,
wire-mesh
cages
in
a
well-ventilated
room
that
was
air-
conditioned
by
means
of
a
system
designed
to
maintain
the
room
temperature
at
an
average
of
22°C
and
a
relative
humid-
ity
of
about
45%.
The
room
was
illuminated
by
a
fluorescent
light
that
simulates
the
spectrum
of
the
sun
(Vita-Lite;
Dura-
Test,
Clifton,
NJ)
to
permit
an
artificial
24-hr
cycle
of
12
hr
of
light
(from
6
a.m.
to
6
p.m.)
daily.
Tap
water
and
Israeli
Chow
diet
were
available
ad
libitum.
Experimental
Design.
Seventy-two
rats
(12
each
month
for
6
months)
were
randomized
to
one
of
eight
treatment
groups
(each
with
9
rats):
group
A,
saline
(0.9o
NaCl);
groups
B-H,
a-linolenic
acid-to-linoleic
acid
ratios
of
1:3
(B),
1:3.5
(C),
1:4
(D),
1:4.5
(E),
1:5
(F),
1:5.5
(G),
and
1:6
(H).
An
equal
volume
(1
ml)
of
placebo
(0.9%o
NaCl)
or
test
material
(2.25
mg
per
rat
with
an
initial
average
weight
of
100
g,
thus
about
25
mg/kg)
was
injected
intraperitoneally
daily
on
days
1-28
in
a
double-
blind
fashion.
The
bolus
consisted
of
2.25
mg
of
free
fatty
acids
(0.01
ml
of
the
stock
solution;
see
Test
Material
above)
mixed
with
mineral
oil
to
make
1
ml.
Observations.
Testing
was
carried
out
immediately
before
the
start
of
administration
and
was
repeated
at
the
end
of
weeks
1,
2,
3,
and
4.
The
animals
were
weighed
at
the
start
and
then
again
at
the
end
of
week
4.
Baseline
measurements
in
the
learning
apparatus
were
obtained
in
eight
trials
daily
on
3
consecutive
days
(days
-3,
-2,
and
-1)
immediately
prior
to
the
start
of
daily
injections
for
4
weeks.
At
the
end
of
each
week
of
treatment,
testing
was
repeated
8
times
daily
on
3
consecutive
days.
Week
1
tests
correspond
to
study
days
7,
8,
and
9;
week
2
tests
correspond
to
study
days
14,
15,
and
16;
week
3
tests
correspond
to
study
days
21,
22,
and
23;
and
week
4
tests
correspond
to
study
days
28,
29,
and
30.
All
tests
were
administered
between
10
a.m.
and
2
p.m.
using
the
same
equipment,
test
instruments,
and
personnel.
All
testing
was
performed
by
an
experimenter
who
was
unaware
of
the
diets
fed
to
the
individual
subjects.
Daily
food
intake
was
mea-
sured
and
converted
into
kcal.
The
order
of
the
additional
testing
was
as
follows:
on
the
first
day
of
the
3-day
testing
periods,
motor
activity
was
measured
(day
28
only),
whereas
pain
threshold
was
measured
on
the
second
day
(days
-2,
8,
15,
22,
and
29),
and
thermoregulation
as
well
as
retention
of
old
learning
was
tested
on
the
third
day
(days
-1,
9,
16,
23,
and
30).
The
Lrning
Apparatus.
The
Morris
water
tank
(see
ref.
34
for
complete
review
of
the
learning
model),
a
circular
tank
(110
cm
in
diameter),
was
filled
with
water
(to
the
level
of
40
cm),
which
was
made
opaque
by
the
addition
of
powdered
milk,
so
that
rats
swimming
in
the
tank
were
unable
to
see
an
escape
platform
(7.5
cm
in
diameter)
submerged
2
cm
below
water
level.
Each
animal
was
released
facing
the
wall
in
one
of
four
predetermined
starting
points
each
separated
by
90°
around
the
inner
perimeter.
While
the
rat
was
in
the
tank,
it
was
able
to
observe
the
contents
of
the
room.
Special
care
was
given
to
keep
things
in
the
room
in
the
same
location.
The
rat
could
navigate
in
the
tank
only
by
external
cues.
Each
rat
was
tested
8
times
per
day
in
the
tank.
The
order
of
the
starting
points
was
determined
by
random
selection.
To
prevent
possible
effects
of
a
magnetic
field,
each
rat
was
allowed
120
sec
to
find
the
platform,
with
an
interval
of
20
sec
between
trials.
The
maximum
duration
of
the
test
for
each
rat
Table
2.
Number
of
trials
to
reach
criterion
(10
sec)
Number
of
trials
Days
pretreatment
Days
after
start
of
treatment
Group
-3,
-2,
-1
7,
8,
9
14,
15,
16
21,
22,
23
28,
29,
30
P
A
19.6
t
3.3
19.0
t
3.7
20.3
±
2.5
18.5
±
2.9
19.1
±
2.7
NS
B
20.1
±
4.1
18.0
+
4.0
19.9
±
4.5
17.1
±
4.0
17.0
±
3.2
NS
C
17.1
+
3.3
12.5
+
2.1*
10.7
+
4.1*
5.6
±
2.5*
5.6
±
2.5*
0.01
D
18.5
±
2.0
9.3
±
2.6*
7.1
+
2.9*
6.1
±
2.5*
6.1
±
2.5*
0.001
E
19.1
+
2.3
14.2
±
3.7*
12.8
±
3.9*
9.0
±
3.4*
9.0
±
3.4*
0.01
F
19.5
±
3.5
16.1
±
2.6
11.2
±
1.1*
7.9
±
1.0*
7.9
±
1.0*
0.01
G
19.7
+
3.8
18.1
+
3.3
18.4
±
2.9
18.6
±
2.6
18.6
±
2.6
NS
H
21.0
±
4.0
20.0
+
3.0
19.6
±
3.1
19.1
+
3.0
19.1
±
3.0
NS
P
NS
0.01
0.01
0.01
0.01
Groups
(nine
rats
per
group)
are
as
identified
in
the
legend
to
Table
1.
Observations
were
made
before
start
of
treatment
and
at
the
end
of
weeks
1-4.
Values
represent
the
means
±
SEM
of
eight
tests
daily
on
3
consecutive
days.
Unsaturated
fatty
acid
treatment
with
co3-to-w6
ratios
of
1:3.5-1:5
(groups
C-F)
had
a
significant
effect
on
learning.
NS,
not
statistically
significant.
*The
P
value
in
the
last
column
indicates
the
P
value
relative
to
the
pretreatment
value
for
that
group.
The
P
value
in
the
bottom
row
indicates
the
P
value
relative
to
the
control
(saline;
group
A)
for
those
days.
Proc.
Natl.
Acad
Sci.
USA
90
(1993)

Proc.
Natl.
Acad.
Sci.
USA
90
(1993)
10347
Table
3.
Time
in
the
"wrong"
location
Time
in
the
wrong
location
Day
pretreatment
Day
after
start
of
treatment
Group
-1
9
16
23
30
P
A
22.9
±
3
24.3
±
4
19.0
±
3
22.3
±
4
25.1
±
4
NS
B
18.5
±
3
19.4
±
4
20.6
±
6
20.6
±
4
20.1
±
5
NS
C
20.3
±
4
30.9
±
2*
35.3
±
4*
39.2
±
4*
49.4
±
3*
0.001
D
19.5
±
3
24.1
±
3
29.3
±
4*
36.6
±
4*
39.1
±
4*
0.01
E
20.8
±
4
25.1
±
4
30.1
±
3*
33.1
±
4*
36.1
±
5*
0.01
F
19.4
±
3
22.1
±
3
29.1
±
5*
30.1
±
5*
32.2
±
5*
0.01
G
22.8
±
4
19.4
±
3
19.0
±
3
19.6
±
4
18.1
±
4
NS
H
19.1
±
5
18.7
±
5
19.9
±
4
21.1
±
3
19.6
±
5
NS
P
NS
0.01
0.001
0.001
0.001
Groups
(nine
rats
per
group)
are
as
identified
in
the
legend
to
Table
1.
The
means
±
SEM
of
the
first
two
trials
are
shown.
Unsaturated
fatty
acid
treatment
with
O3-to-w6
ratios
of
1:3.5-1:5
(groups
C-F)
had
a
significant
effect
on
retention
of
old
learning.
NS,
not
statistically
significant.
*The
P
value
in
the
last
column
indicates
the
P
value
relative
to
the
pretreatment
value
for
that
group.
The
P
value
in
the
bottom
row
indicates
the
P
value
relative
to
the
control
(saline;
group
A)
for
that
day.
was
16
min,
and
three
rats
were
tested
each
hour.
The
rats
were
tested
on
3
consecutive
days.
During
this
period,
the
platform
was
in
the
same
location
in
the
tank.
For
each
of
the
24
trials
(eight
trials
x
three
days),
the
latency
to
reach
the
platform
was
recorded.
A
cutoff
criterion,
defined
as
the
first
successful
trial
with
a
maximum
latency
of
10
sec
without
any
increase
in
latency
on
a
later
trial,
was
used
to
calculate
an
index
of
learning
ability
(rate
of
learning)
for
each
diet
group.
After
the
completion
of
the
eighth
trial
on
day
3,
the
platform
was
removed
and
placed
in
a
different
location
in
the
tank.
The
time
that
the
rats
spend
in
the
"old"
(wrong)
position
for
two
trials
was
used
to
calculate
the
resistance
to
extinction
(retention
of
old
learning).
To
our
knowledge,
this
is
the
first
time
that
this
method
has
been
used
in
a
Morris
water
tank.
The
Level
of
Motor
Activity.
This
endpoint
was
assessed
in
an
open
field
apparatus
by
recording
the
number
of
horizontal
movements
(infrared
photobeam
crossings)
and
rearing
movements
(determined
from
videotapes)
made
during
the
15-min
sessions.
The
apparatus
was
very
similar
to
the
one
previously
described
by
Coscina
and
Yehuda
(4).
Pain
Threshold.
A
plate,
60
x
60
cm,
was
held
at
58
±
0.2°C
by
a
thermostatic
bath
(Hakka,
Karlswke,
Germany).
The
21
C~~~~~~
.0
a)
15
~12-
9OD
E
z
3
z
0
animal
was
placed
on
the
plate.
The
latency
(to
the
nearest
0.1
sec)
to
lick
the
paw
was
recorded
(35).
Thermoregulation.
As
previously
described
(36),
the
basal
colonic
temperature
of
each
rat
was
measured
(Yellow
Springs
Instruments
telethermometer,
model
43TA),
after
which
the
rat
was
injected
intraperitoneally
with
d-amphet-
amine
(15.0
mg/kg)
and
placed
immediately
into
a
4°C
cold
room
for
1
hr.
The
temperature
was
recorded
again
after
60
min
in
the cold
room.
Statistics.
Group
comparisons
were
made
using
ANOVA
(one-way
repeated
measures)
with
subsequent
contrast
t
tests.
RESULTS
Effect
of
Food
Intake
and
Body
Weight.
As
shown
in
Table
1,
treatment
with
the
test
material
had
no
effect
on
food
intake
or
body
weight
in
any
of
the
groups.
Effect
on
Learning.
Tables
2
and
3
and
Fig.
1
present
the
results
of
the
learning
experiments.
At
the
end
of
weeks
1-4,
groups
C,
D,
and
E
(i.e.,
ratios
of
1:3.5,
1:4,
and
1:4.5)
showed
a
significant
reduction
in
the
number
of
trials
nec-
essary
to
reach
the
criterion
(10
sec)
as
compared
to
base-
line-i.e.,
before
start
of
treatment
(Table
2
and
Fig.
1).
This
was
also
observed
for
group
F
(ratio
1:5)
after
week
2.
-a--
-8-..
A
(NaCI)
B
(1:3)
C
(1
:3.5)
D
(1:4)
-*-
E
(1:
4.5)
F
(1:5)
G
(1:5.5)
-X-
H
(1:
6)
FIG.
1.
Improvement
of
learning
(Morris
water
tank
testing
during
4
weeks
of
treatment
at
25
mg
per
kg
per
day)
in
the
rat
by
a-linolenic/linoleic
acid
mixtures
with
w3-to-w6
ratios
of
1:3.5-1:5
(groups
C-F).
The
values
plotted
are
the
means;
error
bars
indicate
the
SEM.
0
1
2
3
4
Weeks
of
treatment
Neurobiology:
Yehuda
and
Carasso

10348
Neurobiology:
Yehuda
and
Carasso
ANOVA
shows
that
the
ratio
1:4
(group
D)
differs
statisti-
cally
at
repeated
measures
at
the
level
of
P
<
0.001,
indicating
that
this
ratio
may
be
most
effective
in
improving
perfor-
mance
in
a
cognitive
task.
At
the
end
of
weeks
1-4,
group
C
(ratio
1:3.5)
and
at
the
end
of
weeks
2-4,
groups
D,
E,
and
F
(ratios
of
1:4,
1:4.5,
and
1:5)
also
showed
significant
effects
on
the
retention
of
old
learning
as
determined
by
the
duration
of
time
spent
in
the
wrong
(old)
platform
location
(Table
3).
Effect
on
Motor
Activity.
At
the
end
of
weeks
1-4,
none
of
the
treatment
groups
showed
any
significant
effects
on
hor-
izontal
or
vertical
movement,
as
determined
by
counts
of
infrared
photo
beam
crossings
and
the
frequency
of
rearings
(data
not
shown).
Effect
on
Pain
Threshold.
At
the
end
of
weeks
1-4,
groups
D
and
E
(ratios
of
1:4
and
1:4.5)
and
at
the
end
of
weeks
2-4
and
3-4,
groups
C
and
F
(ratios
of
1:3.5
and
1:5),
respec-
tively,
also
showed
a
significant
effect
on
pain
threshold
(analgesia)
as
determined
by
the
latency
for
the
rat
to
lick
the
paw
after
being
placed
on
a
hot
plate
(Table
4).
Effect
on
Thermoregulation.
At
the
end
of
weeks
1-4,
groups
C-F
(ratios
of
1:3.5-1:5)
showed
significant
protec-
tion
from
d-amphetamine-induced
hypothermia,
as
deter-
mined
by
colonic
temperature
measurements
before
and
after
placement
in
a
cold
room
(4°C)
for
1
hr
(data
not
shown).
DISCUSSION
The
results
of
this
study
show
that
administration
of
a-linolenic
acid
and
linoleic
acid
preparations
with
ratios
of
the
fatty
acids
ranging
from
1:3.5
to
1.5
has
a
significant
effect
on
learning
(Table
2
and
Fig.
1),
which
cannot
be
explained
by
changes
either
in
food
intake
or
weight
gain
(Table
1)
or
in
motor
activity
(data
not
shown).
Furthermore,
as
all
animals
were
about
1
month
of
age,
there
was
no
difference
in
the
stage
of
develop-
ment
between
treatment
and
control
groups
that
could
impact
on
the
results.
Formulations
with
a
ratio
of
1:4
have
been
selected
for further
experimental
and
clinical
evaluation.
The
effect
of
these
specific
PUFA
preparations
on
pain
threshold
(Table
4)
and
thermoregulation
(data
not
shown)
after
1-4
weeks
of
administration
suggests
that
the
activity
is
mediated
via
central
sites,
since
d-amphetamine-induced
hypothermia
is
regulated
by
dopaminergic
neurons
in
the
brain
(36).
Linoleic
and
a-linolenic
acids
and
their
elongated
and
desaturated
derivatives
are
polyunsaturated
EFAs,
which
presently
are
recognized
to
have
several
important
biological
functions.
Approximately
20%
of
the
dry weight
of
the
brain
consists
of
EFAs,
which
are
incorporated
into
phospholipids
that
are
critically
important
for
the
structural
integrity
of
neuronal
membranes,
membrane
fluidity
(28),
and
mem-
brane-related
functions
such
as
receptor,
enzyme,
and
ion
channel
kinetics,
as
well
as
eicosanoid
functions
(37-39).
The
03
and
w6
fatty
acid
families
are
closely
interrelated
with
positive
and
negative
feedback
regulation
of
desaturat-
ing
conversion
enzymes
(40).
Although
alteration
of
the
fatty
acid
composition
of
brain
lipids
by
varying
levels
of
ingested
EFAs
was
demonstrated
almost
three
decades
ago
(41),
only
recently
has
research
focused
on
determination
of
the
optimal
3-to-w6
ratio
in
adult
diet
and
in
infant
or
parenteral
formulas.
Whereas
Neuringer
et
al.
(18)
considered
a
ratio
in
the
range
1:4-1:10
to
be
prudent,
several
recent
reports
(42-45)
support
our
observation
that
a
1:4
ratio
may
be
optimal.
A
recent
North
Atlantic
Treaty
Organization
con-
ference
on
essential
fatty
acids
recommended
a
ratio
of
1:4
(42).
Analysis
of
the
fatty
acid
composition
of
human
milk
from
Canadian
Eskimos
eating
a
traditional
diet
showed
a
w3-to-co6
ratio
of
1:4
(43).
Wainwright
et
al.
(44)
found
evidence
in
the
developing
mouse
that
maximal
incorporation
of
w3
fatty
acids
in
the
phosphatidylethanolamine
fraction
of
the
brain
membrane
occurred
with
a
ratio
of
1:4.
Clark
et
al.
(45)
found
evidence
that
maximal
w3
incorporation
into
erythrocyte
membranes
(a
postulated
marker
for
brain
mem-
branes)
in
human
infants
occurred
after
use
of
formulas
with
a
ratio
of
1:4.
It
is
possible
that
such
a
ratio
may
optimize
enzyme
kinetics
within
the
w3
and
w6
fatty
acid
families
to
allow
maximal
conversion
to
the
elongated
and
desaturated
species.
Thus,
it
is
likely
that
at
least
some
of
the
beneficial
effects
of
certain
PUFA
formulations
are
related
to
optimal
incorporation
of
w3
fatty
acids
into
the
brain
membranes,
without
causing
a
concomitant
inhibition
of
co6
conversions
(and
subsequent
depletion
of
w6
derivatives)
in
the
mem-
brane
due
to
negative
feedback
regulation.
This
concept
is
also
in
agreement
with
a
recent
report
that
expressed
caution
in
the
use
of
w3
supplements
in
infant
formulas
without
additional
w6
supplementation
(46).
A
fundamental
question
is
how
can
a
specific
ratio
of
EFA
in
administered
preparations
be
biologically
meaningful.
The
basic
diet
(Israeli
Chow)
according
to
specifications
by
the
manufacturer
and
our
own
biochemical
analysis
contains
about
0.15
mg
of
a-linolenic
acid
per
kg
of
diet
and
about
35
mg
of
linoleic
acid
per
kg
of
diet
(i.e.,
a
ratio
of
1:233;
total
fat
about
5.1%).
This
could
indicate
that
the
feeding
of
a
diet
low
in
a-linolenic
acid
may
have
caused
a
relative
EFA
deficiency
and
cognitive
dysfunction
at
baseline
in
all
groups
as
well
as
during
the
total
study
period
in
the
control
groups.
If
so,
the
observation
that
supplemental
a-linolenic
acid
and
linoleic
Table
4.
Pain
threshold
Latency
to
lick
the
paw
after
being
placed
on
a
hot
plate,
sec
Day
pretreatment
Day
after
start
of
treatment
Group
-2
8
15
22
29
P
A
7.9
±
0.9
7.8
±
0.8
8.0
±
0.6
7.9
±
0.9
8.1
±
0.9
NS
B
8.0
±
0.8
7.9
±
0.7
8.0
±
0.9
8.1
±
0.7 7.8
±
0.7
NS
C
7.8
±
0.6
11.9
±
0.7
13.9
±
0.7*
16.5
±
0.6*
20.1
±
1.1*
0.01
D
8.1
±
0.8
12.1
±
0.6*
14.5
±
0.6*
18.2
±
0.7*
21.1
±
0.9*
0.01
E
7.8
±
0.6
9.0
±
0.9*
9.0
±
0.8*
14.1
±
0.7*
17.4
±
0.7*
0.01
F
8.1
±
0.9
9.9
±
0.9
11.5
±
0.7
14.1
±
0.7*
16.3
±
0.7*
0.01
G
7.6
±
0.7
8.0
±
0.3
8.8
±
0.8
8.0
±
0.8
8.1
0.9
NS
H
8.0
±
0.9
8.0
±
0.4
8.5
±
0.5
8.3
±
0.7
8.3
0.7
NS
P
NS
0.05
0.01
0.01 0.01
Groups
(nine
rats
per
group)
are
as
identified
in
the
legend
to
Table
1.
Values
given
are
the
mean
+
SEM.
Unsaturated
fatty
acid
treatments
with
ratios
of
1:3.5-1:5
(groups
C-F)
caused
analgesia
in
rats
that
were
placed
on
a
heated
plate
(58°C).
NS,
not
statistically
significant.
*The
P
value
in
the
last
column
indicates
the
P
value
relative
to
the
pretreatment
value
for
that
group.
The
P
value
in
the
bottom
row
indicates
the
P
value
relative
to
the
control
(saline;
group
A)
for
that
day.
Proc.
Natl.
Acad
Sci.
USA
90
(1993)

Proc.
Natl.
Acad.
Sci.
USA
90
(1993)
10349
acid
in
a
specific
range
of
ratios
may
improve
cognitive
function
becomes
even
more
intriguing.
This
would
suggest
that
such
treatment
may
have
potential
therapeutic
value
in
Alzheimer
disease,
where
autopsy
studies
have
shown
re-
duced
content
of
certain
EFAs
in
affected
brain
regions
(47).
We
are
presently
unable
to
offer
a
definitive
explanation
for
the
effects
of
certain
PUFA
formulations
on
memory
function,
pain
thresholds,
and
thermoregulation.
Previous
data
provides
evidence
that
PUFAs
in
the
diet
can
decrease
plasma
levels
of
cholesterol
and
that
membrane
cholesterol
levels
(which
are
correlated
with
plasma
levels)
are
inversely
related
to
mem-
brane
fluidity
(37,
38).
Thus,
it
seems
reasonable
to
postulate
that
the
beneficial
effects
may
be
directly
related
to
effects
of
the
administrated
EFAs
on
the
composition
and
fluidity
of
neural
membranes
in
the
central
nervous
system.
Another
biologically
important
function
of
a-linolenic
acid
may
be
to
provide
acetate
for
the
de
novo
synthesis
of
palmitic
and
other
long-chain
fatty
acids,
which
are
essential
for
membrane
integrity
(27).
Furthermore,
since
elongated
and
desaturated
EFAs
are
enriched
in
the
brain
and
their
conversion
mecha-
nisms
have
been
reported
to
be
competent
in
rats
as
well
as
in
humans,
even
at
an
advanced
age
(48),
it
is
likely
that
some
of
the
beneficial
effects
of
the
administered
EFAs
are
mediated
by
such
longer
chain
derivatives.
We
have
previously
demonstrated
that
treatment
of
rats
with
soybean
oil
(49)
or
certain
peptides
(50)
also
provides
protection
from
d-amphetamine-induced
hypothermia
when
the
animals
are
placed
in
a
cold
room.
Our
theory
is
that
d-amphetamine-induced
hypothermia
is
mediated
by
the
do-
paminergic
system
in
the
striatum
(51).
It
is
possible
that
certain
formulations
of
PUFAs
affect
the
dopaminergic
sys-
tem,
most
likely
the
D2
receptors.
In
summary,
our
results
show
that
treatment
of
rats
for
2-4
weeks
with
preparations
of
a-linolenic
acid
in
combination
with
linoleic
acid
in
ratios
ranging
from
1:3.5
to
1:5
had
a
significant
effect
on
the
rate
of
learning,
retention
of
old
learning,
pain
thresholds
(analgesia),
and
prevention
of
the
d-amphetamine-induced
hypothermic
response
to
reduced
ambient
temperature.
A
clearer
understanding
of
the
mode
of
action
of
certain
formulations
of
PUFAs
will
assist
in
the
further
evaluation
in
animal
models
as
well
as
in
human
memory
disorders,
such
as
Alzheimer
disease,
and
other
degenerative
disorders,
where
free
radical
formation,
oxida-
tion,
deficiency
of
PUFAs
(47,
52,
53)
or
degeneration
of
brain
membrane
phospholipids
(54)
have
been
implicated.
We
thank
Karl
L.
Mettinger,
M.D.,
Ph.D.
(IVAX/Baker
Norton
Pharmaceuticals,
Miami)
for
his
helpful
comments
on
the
research
and
Professor
David
I.
Mostofsky,
University
of
Boston,
who
followed
this
research
from
its
early
stages
to
the
manuscript.
We
would
like
to
acknowledge
the
support
received
from
the
Ginsburg
Chair
and
the
William
Farber
Center
for
Alzheimer
Research.
1.
Wurtman,
R.
J.,
Hefti,
F.
&
Melamed,
E.
(1981)
Pharmacol.
Rev.
32,
315-335.
2.
Yehuda,
S.
(1987)
Int.
J.
Neurosci.
35,
21-36.
3.
Kessler,
R.
A.
&
Yehuda,
S.
(1985)
Int.
J.
Neurosci.
28,
73-82.
4.
Coscina,
D.
V.
&
Yehuda,
S.
(1986)
Life
Sci.
38,
1789-1794.
5.
Yehuda,
S.
&
Carasso,
R.
L.
(1987)
Int.
J.
Neurosci.
32,
919-925.
6.
Yehuda,
S.
(1989)
in
Health
Effects
of
Fish
and
Fish
Oils,
ed.
Chandra,
R.
K.
(ARTS,
St
John's,
NF,
Canada),
pp.
327-335.
7.
Lamptey,
M.
S.
&
Walker,
B.
L.
(1976)
J.
Nutr.
106,
80-93.
8.
Yamaoto,
N.,
Nomura,
M.
&
Okuyama,
H.
(1987)
J.
Lipid
Res.
28,
144-151.
9.
Mills,
D.
E.,
Ward,
R.
P.
&
Young,
C.
(1988)
Nutr.
Res.
8,273-286.
10.
Fujimoto,
K.,
Yao,
K.,
Miyazawa,
T.,
Hirono,
H.,
Nishikawa,
M.,
Kimura,
S.,
Maruyama,
K.
&
Nonaka,
M.
(1989)
in
Health
Effects
of
Fish
and
Fish
Oils,
ed.
Chandra,
R.
K.
(ARTS,
St
John's,
NF,
Canada),
pp.
275-284.
11.
Bourre,
J.-M.,
Francois,
M.,
Youyou,
A.,
Dumont,
O.,
Piciotti,
M.,
Pascal,
G.
&
Durand,
G.
(1989)
J.
Nutr.
119,
1880-1892.
12.
Dhopeshwarkar,
G. A.
(1983)
Nutrition
and
Brain
Development
(Plenum,
New
York),
pp.
59-69.
13.
Dhopeshwarkar,
G.
A.
&
Mead,
J.
F.
(1973)
in
Adv.
Lipid
Res.
11,
109-142.
14.
Bernsohn,
J.
&
Spitz,
F.
J.
(1974)
Biochem.
Biophys.
Res.
Com-
mun.
57,
293-298.
15.
Holman,
R.
T.,
Johnson,
S.
B.
&
Hatch,
T.
F.
(1982)
Am.
J.
Nutr.
35,
617-623.
16.
Bjerve,
K.
S.
(1989)
J.
Intern.
Med.
225,
Suppl.
1,
171-175.
17.
Uauy,
R.
D.,
Birch,
D.
G.,
Birch,
E.
E.,
Tyson,
J.
E.
&
Hoffman,
D.
R.
(1990)
Pediatr.
Res.
28,
485-492.
18.
Neuringer,
M.,
Anderson,
G.
J.
&
Connor,
W.
(1988)
Annu.
Rev.
Nutr.
8,
517-541.
19.
Bourre,
J.-M.,
Dumont,
O.,
Piciotti,
M.,
Clement,
M.,
Chaudiere,
J.,
Boneil,
M.,
Nalbone,
G.,
Lafont,
H.,
Pascal,
G.
&
Durand,
G.
(1991)
World
Rev.
Nutr.
Diet
66,
103-117.
20.
Connor,
W.
E.,
Neuringer,
M.
&
Reisbick,
S.
(1991)
World
Rev.
Nutr.
Diet
66,
118-132.
21.
Bazan,
N.
G.
(1990)
Nutr.
Brain
8,
1-24.
22.
Bazan,
N.
G.
&
de
Turco,
R.
(1980)
Adv.
Neurol.
28,
197-205.
23.
Bourre,
J.
M.,
Francois,
M.,
Youyou,
A.,
Dumont,
O.,
Piciotti,
M.,
Pascal,
G.
&
Durand,
G.
(1989)
J.
Nutr.
119,
1880-1892.
24.
Clandinin,
M.
T.,
Cheema,
S.,
Field,
C.
J.,
Grag,
M.
L.,
Venka-
traman,
J.
&
Clandinin,
T.
R.
(1991)
FASEB
J.
5,
2761-2769.
25.
Cook,
H.
W.
(1991)
Nutrition
7,
440-442.
26.
Crawford,
M.
A.,
Hassam,
A.
G.
&
Williams,
G.
(1976)
Lancet
1,
452-453.
27.
Cunnane,
S.
C.,
Chen,
Z.
Y.,
Yang,
J.,
Liede,
A.
C.,
Hamadeh,
M.
&
Crawford,
A.
C.
(1991)
Nutrition
7,
437-439.
28.
Holman,
R.
T.,
Johnson,
S.
B.
&
Ogburn,
P.
L.
(1991)
Proc.
Natl.
Acad.
Sci.
USA
88,
4835-4839.
29.
Scott,
B.
L.
&
Bazan,
N.
G.
(1989)
Proc.
Natl.
Acad.
Sci.
USA
86,
2903-2907.
30.
Simopoulos,
A.
P.
(1991)
Am.
J.
Clin.
Nutr.
54,
438-463.
31.
Spector,
A.
A.
(1989)
INSERM
Symp.
195,
11-20.
32.
Wainwright,
P.
E.
(1992)
Neurosci.
Biobehav.
Rev.
16,
193-205.
33.
Wiegand,
R.
D.,
Koutz,
C.
K.,
Stinson,
A.
M.
&
Anderson,
R.
E.
(1991)
J.
Neurochem.
57,
1690-1699.
34.
Brandeis,
R.,
Brandys,
Y.
&
Yehuda,
S.
(1989)
Int.
J.
Neurosci.
48,
29-69.
35.
Yehuda,
S.
&
Youdim,
M.
B.
H.
(1984)
Eur.
J.
Pharmacol.
104,
245-251.
36.
Yehuda,
S.
&
Wurtman,
R.
J.
(1972)
Nature
(London)
240,477-478.
37.
Houslay,
M.
D.
&
Stanley,
K.
K.
(1982)
Dynamics
of
Biological
Membranes
(Wiley,
New
York).
38.
Shinitzky,
M.
(1984)
Physiology
of
Membrane
Fluidity
(CRC,
Boca
Raton,
FL),
Vol.
1,
pp.
1-51.
39.
Greenwood,
G.
E.
&
Craig,
R.
E.
A.
(1987)
in
Basic
and
Clinical
Aspects
of
Nutrition
and
Brain
Development,
eds.
Rassin,
D.
K.,
Haber,
B.
&
Drujan,
B.
(Liss,
New
York),
pp.
159-216.
40.
Holman,
R.
T.
(1964)
Fed.
Proc.
Fed.
Am.
Soc.
Exp.
Biol.
23,
1062-1067.
41.
Mohrhauer,
H.
&
Holman,
R.
T.
(1963)J.
Neurochem.
10,
523-530.
42.
Galli,
C.
&
Simopoulos,
A.
P.,
eds.
(1988)
in
Dietary
o--3
and
w-6
Fatty
Acids:
Biological
Effects
and
Nutritional
Essentiality
(Ple-
num,
New
York),
pp.
403-404.
43.
Innis,
S.
M.
(1989)
in
Dietary
w-3
and
w-6
Fatty
Acids:
Biological
Effects
and
Nutritional
Essentiality,
eds.
Galli,
C.
&
Simopoulos,
A.
P.
(Plenum,
New
York),
pp.
135-146.
44.
Wainwright,
P.
E.,
Huang,
Y.
S.,
Bulman-Fleming,
B.,
Dalby,
D.,
Mills,
D.
E.,
Redden,
P.
&
McCutcheon,
D.
(1992)
Lipids
27,
98-103.
45.
Clark,
K.
J.,
Makrides,
M.,
Neumann,
M.
A.
&
Gibson,
R.
A.
(1992)
J.
Pediatr.
120,
S151-S158.
46.
Arbuckle,
L.
D.
&
Innis,
S.
M.
(1992)
Lipids
27,
89-93.
47.
Soderberg,
M.,
Edlund,
C.,
Kristensson,
K.
&
Dallner,
G.
(1991)
Lipids
26,
421-425.
48.
Rudman,
D.,
Mattson,
D.
E.
&
Feller,
A.
G.
(1989)
J.
Am.
Geriatr.
Soc.
37,
229-234.
49.
Yehuda,
S.,
Leprohon-Greenwood,
C.
E.,
Dixon,
L.
M.
&
Co-
scina,
D.
V.
(1986)
Pharmacol.
Biochem.
Behav.
24,
1775-1777.
50.
Yehuda,
S.
(1978)
Curr.
Dev.
Psychopharmacol.
5,
125-175.
51.
Yehuda,
S.
&
Kastin,
A.
J.
(1980)
Neurosci.
Biobehav.
Rev.
4,
459-471.
52.
Harman,
D.
(1984)
Age
7,
111-131.
53.
Holman,
R.
T.,
Johnson,
S.
B.
&
Kokmen,
E.
(1989)
Proc.
Natl.
Acad.
Sci.
USA
86,
4720-4724.
54.
Nitsch,
R.
M.,
Blusztajn,
J.
K.,
Pittas,
A.
G.,
Slack,
B.
E.,
Grow-
don,
J.
H.
&
Wurtman,
R.
J.
(1992)
Proc.
Natl.
Acad.
Sci.
USA
89,
1671-1675.
55.
Lawson,
L.
D.
&
Hughes,
B.
G.
(1988)
Biochem.
Biophys.
Res.
Commun.
152,
328-335.
56.
Horrocks,
L.
A.
(1988)
in
Phospholipids
in
Nervous
Tissues,
ed.
Eichberg,
J.
(Wiley,
New
York),
pp.
173-179.
Neurobiology:
Yehuda
and
Carasso