Preferential utilization of newly synthesized
cholesterol for brain growth in neonatal lambs
STEPHEN D. TURLEY,1 DENNIS K. BURNS,2 AND JOHN M. DIETSCHY1
Departments of 1Internal Medicine and 2Pathology, University of Texas
Southwestern Medical Center at Dallas, Dallas, Texas 75235-8887
central nervous system; hepatic; cholesteryl ester; lowdensity lipoprotein; ovine
CHOLESTEROL is transported in the plasma in different
classes of lipoproteins that are taken up by specific
receptors located in the liver and extrahepatic organs.
Within the liver, a number of functionally distinct
lipoprotein receptors have been identified (8). One of
these, the low-density lipoprotein (LDL) receptor, is
responsible for mediating the clearance of most of the
LDL that is removed daily from the circulation (14),
whereas the SR-B1 receptor facilitates the binding of
high-density lipoproteins that transport cholesterol
from the peripheral organs to the liver (1). Hepatocytes
also express a receptor that may be partially responsible for clearing triacylglycerol-rich remnants containing apolipoprotein E (8). Multiple lipoprotein receptors,
including the LDL receptor, have also been found in the
central nervous system (CNS), but in most cases, their
function has not been clearly established (19, 26, 37).
There is now considerable interest in determining what
role these receptors might play in regulating cholesterol homeostasis in the CNS because there are several
major neurological disorders afflicting children and
adults in which a severe derangement of cholesterol
metabolism is manifest (3, 28).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
Currently, there are two main avenues of investigation into the control of brain cholesterol metabolism.
One of these is focusing on the question of how cholesterol turnover in the adult brain is regulated. Early
experiments in baboons and more recent studies in
humans have shown that the turnover of cholesterol in
the adult brain is very slow and usually amounts to
much less than 1% of the total tissue pool of sterol per
day (22, 38). There is now evidence that this turnover is
facilitated by the conversion of cholesterol to a hydroxylated derivative that passes the blood-brain barrier
more readily than cholesterol itself (22).
The other major area of study concerns the origin of
cholesterol in CNS tissue. The brain is the most cholesterol-rich organ in the body (7), and it acquires most of
its sterol during myelination, which takes place in the
early stages of development (29). Ultimately, this cholesterol must have been either synthesized locally and/or
delivered there through the uptake of specific lipoproteins. The presence of mRNA for the LDL receptor in
the brains of neonatal rats and rabbits implies that
LDL from the plasma might be a source of cholesterol
for brain growth in the developing animal (24, 31).
There is no in vitro system in which such a contribution
can be quantitated, and methods for doing so in vivo
cannot be applied to neonates of species such as the rat
because of their prohibitively small body size. However,
such measurements are technically feasible in lambs
that weigh ,4 kg at birth and have, by 2–3 wk of age,
plasma LDL cholesterol concentrations that approach
those reported for breast-fed human infants (5, 9).
Furthermore, in this model, the mass of CNS tissue is
sufficiently great to permit detailed metabolic measurements in multiple regions of the brain. The present
studies thus describe the application of techniques for
the measurement in vivo of the fraction of cholesterol
contained in newly formed CNS tissue in suckling
lambs that is derived from de novo synthesis and from
the uptake of LDL from the plasma. The data support
other lines of evidence that newly synthesized cholesterol is preferentially utilized for the growth and differentiation of the CNS in the neonate.
MATERIALS AND METHODS
Animals. Timed-pregnant crossbred Rambouillet ewes (Ovis
aries) carrying either single or twin fetuses were obtained
from the Department of Veterinary Science, University of
Texas M. D. Anderson Cancer Center (Bastrop, TX), and
maintained as described (5). After natural birth, the lambs
had unrestricted access to their dams until taken for study. A
total of 18 early neonatal lambs (14 males, 4 females) at ages
ranging from 0.2 to 3 days and 11 late neonatal animals (3
males, 8 females) aged 16–18 days were used for different
types of expe riments. All procedures were approved by the
0193-1849/98 $5.00 Copyright r 1998 the American Physiological Society
E1099
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
Turley, Stephen D., Dennis K. Burns, and John M.
Dietschy. Preferential utilization of newly synthesized cholesterol for brain growth in neonatal lambs. Am. J. Physiol.
274 (Endocrinol. Metab. 37): E1099–E1105, 1998.—These
studies used the suckling lamb as a model to determine the
sources of cholesterol that are utilized for development of the
central nervous system in the neonate. Lambs were studied
at 1.3 and 16.4 days after birth. Over this 15-day interval, 14
g of new brain tissue were formed. About 9–10 mg of
cholesterol were utilized daily for this new tissue growth. To
determine the source of this cholesterol, the absolute rates of
low-density lipoprotein clearance and cholesterol synthesis
were measured in vivo in nine separate regions of the central
nervous system. Low-density lipoprotein clearance throughout the brain was very low and at most could have contributed
only 0.3–0.4 mg cholesterol daily for new brain growth. In
contrast, the brain synthesized 7–8 mg of cholesterol/day.
There were pronounced regional differences in the concentration of cholesterol throughout the brain, and these correlated
closely with the rate of sterol synthesis (r 5 0.95) in these
same regions. We conclude that the principal source of sterol
for brain growth in suckling lambs is de novo synthesis.
E1100
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
per gram of tissue (nmol · h21 · g21 ). These data and the organ
weights were used to calculate whole animal sterol synthesis
(µmol · h21 · kg body wt21 ). The rate of incorporation of [3H]water into sterol by certain organs such as the brain and also by
the whole body was converted to an equivalent milligram
quantity of cholesterol, assuming 0.69 3H atoms were incorporated into the sterol molecule per carbon atom entering the
biosynthetic pathway as acetyl-CoA (13).
Dissection of brain into various regions. After death, the
cranial vault of the lamb was opened superiorly, and the
rostral spinal cord was exposed via a posterior vertebral
laminectomy. Cranial nerves and vascular connections were
severed, the brain and spinal cord were removed, and the
brains were immediately weighed. The cerebral hemispheres,
brain stem, and cerebellum were sectioned in the fresh state
in the coronal, horizontal, and sagittal planes, respectively, at
,5-mm intervals. The following regions were obtained for
biochemical and radioisotopic analysis: superior frontal neocortex, superior parietal neocortex, white matter from corpus
callosum and centrum ovale, corpus striatum, thalamus
posterior to corpus striatum, midbrain at the level of the
superior colliculus, midportion of the basis pontis, midportion
of the medulla oblongata, cerebellar vermis, and cervical
spinal cord. The proportion of total brain mass distributed
among the nine regions that were studied was determined in
a separate lot of organs taken from four late fetal animals.
The weight of tissue in each region was recorded and expressed as a percentage of the total brain weight, which
averaged 59.2 6 2.7 g. The relative proportions of each region
were as follows: frontal cortex, 20.5 6 1.5%; parietal cortex,
18.3 6 1.8%; cerebellar folia, 9.2 6 0.2%; corpus striatum,
2.5 6 0.2%; thalamus, 4.0 6 0.4%; midbrain, 4.1 6 0.3%;
cerebral white matter, 8.7 6 0.3%; brain stem, 3.7 6 0.2%;
and remaining brain parenchyma including temporal and
occipital neocortex (these were combined and designated
‘‘other regions’’), 29.0 6 1.5%.
Analytical procedures. Plasma and tissue total cholesterol
concentrations and the level of esterified and unesterified
cholesterol in liver and brain were measured as described
elsewhere (16, 35). Although noncholesterol sterols such as
desmosterol are found in brain tissue in fetal sheep (S. D.
Turley and J. M. Dietschy, unpublished findings) and in the
neonates of other species such as the rat (17), in the lambs
studied here, cholesterol was the only sterol detected throughout all regions of the brain.
Analysis of data. The data are presented either as values
for individual animals or as means 6 SE of values obtained
from the number of lambs specified. The data for male and
female lambs were combined because no gender-related differences in any of the metabolic parameters studied were
apparent. Differences between mean values were tested for
statistical significance using the two-tailed unpaired Student’s t-test.
RESULTS
The average age of all the early and late neonatal
lambs studied here was 1.3 and 16.4 days, respectively.
Over this 15-day interval, body weights more than
doubled as did plasma LDL cholesterol levels, and
brain mass increased by 14 g (Table 1). Because the
mean concentration of cholesterol in brain tissue was
similar at both stages of development, the increment in
brain cholesterol content, which averaged ,9–10 mg/
day, was due entirely to the growth of new CNS tissue.
The first experiment was designed to determine the
extent to which the cholesterol utilized for this CNS
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
Institutional Animal Care and Research Advisory Committee.
Measurement of the rate of LDL clearance by the liver and
brain in vivo. LDLs were isolated by density gradient ultracentrifugation in the density range of 1.020–1.055 g/ml from the
plasma of donor adult sheep that had been fed a cholesterolfree commercial diet and grass. Gradient gel electrophoresis
revealed apolipoprotein B-100 to be the only protein in this
fraction (11). The LDL was labeled either with 125I-labeled
tyramine cellobiose or directly with 131I as described (34). The
labeled lipoproteins were used within 48 h and were filtered
through 0.45-µm Millex filters (Millipore, Bedford, MA) immediately before use. Five early neonatal lambs (3 males, 2
females) ranging in age from about 1 to 3 days and five late
neonatal animals (2 males, 3 females) aged 16–18 days were
sedated with xylazine (1 mg/kg body wt, given im; Gemini SA,
Rugby Laboratories, Rockville Center, NY), fitted with two
jugular vein catheters, and given a bolus injection of 125Ityramine cellobiose-LDL. This was followed by a continuous
infusion of the same labeled LDL preparation over the next 5
h, during which time the lambs were kept sedated under
warm covers. The plasma level of labeled LDL was measured
at several points throughout this period. Ten minutes before
the end of the infusion, a bolus injection of 131I-labeled LDL
was administered as a marker for tissue blood contamination.
The lambs were then given an overdose of pentobarbital
sodium (50 mg/kg body wt; Nembutal, Abbott Laboratories,
Chicago, IL) and were exsanguinated from the abdominal
aorta. The liver and brain were removed, rinsed, and weighed.
The 125I and 131I contents of aliquots of plasma and of tissue
from each organ, including multiple regions of the brain
(dissection is detailed in Dissection of brain into various
regions), were then measured (34). Plasma was fractionated
to obtain the LDL cholesterol level in each animal. The rate of
LDL clearance by each organ was expressed as microliters of
plasma cleared of LDL content per hour per gram of tissue
(µl · h21 · g21 ). Multiplication of this value by the whole organ
weight and by the plasma LDL cholesterol concentration
(µg/µl) yielded the amount of LDL cholesterol cleared by the
liver or brain (µg/h for each organ).
Measurement of the rate of sterol synthesis by the liver and
extrahepatic organs in vivo. A total of 13 early neonatal lambs
(11 males, 2 females) aged from 0.2 to 2.5 days were sedated,
fitted with a jugular vein catheter, and placed under a
well-ventilated fume hood. After the intravenous administration of a bolus of [3H]water (750 mCi/kg body wt; ICN
Biomedicals, Irvine, CA), the lambs were kept sedated under
warm covers for the next 60 min, after which time they were
given an intravenous overdose of pentobarbital sodium and
exsanguinated. The liver, whole small intestine, adrenal
glands, kidneys, spleen, and whole brain, as well as a section
of spinal cord, were removed, rinsed, and weighed. Aliquots of
tissue were taken for the measurement of cholesterol concentration and [3H]sterol content (32). The entire remaining
carcass was also digested in alcoholic KOH, and its total
[3H]sterol content was similarly determined. In the case of
the brain, the tissue aliquots were taken from multiple
regions as described in Dissection of brain into various
regions. Before the brain was removed, cerebrospinal fluid
(CSF) was collected directly from the cisterna magna. Aliquots of CSF and plasma were taken for the measurement of
water and 3H content so that the specific activity (SA) of water
in both fluids could be determined. The SA of water in the
plasma and CSF averaged 15.0 6 1.0 and 14.6 6 1.0
counts · min21 · nmol21, respectively. The rate of sterol synthesis in each organ was expressed as the nanomoles of [3H]water incorporated into digitonin-precipitable sterols per hour
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
Table 1. Plasma LDL cholesterol levels, brain weight,
and brain cholesterol content in neonatal lambs
at 2 stages of development
Parameter
Early
Neonatal
Late
Neonatal
Age, days
Gender, male/female
Body wt, kg
Brain wt, g
Plasma LDL cholesterol concn, mg/dl
Brain mean cholesterol concn, mg/g
Brain cholesterol content, mg/brain
1.3 6 0.2
14/4
4.2 6 0.2
59.0 6 1.4
29.4 6 3.2
11.8 6 0.2
698 6 22
16.4 6 0.2
3/8
8.9 6 0.3*
73.3 6 1.8*
62.7 6 4.1*
11.5 6 0.4
841 6 30*
interval, it can be calculated from these clearance
values that, at most, only 0.3–0.4 mg of the 9–10 mg of
cholesterol used daily in the formation of new brain
tissue could have been derived from LDL.
The finding that essentially none of the cholesterol
utilized for brain growth in neonatal lambs was derived
from LDL prompted us to make a detailed study of
sterol biosynthetic activity and its relationship to tissue cholesterol content in the brain of these animals.
These studies, which were carried out only in young
neonatal lambs in the age range of 0.2–2.5 days,
involved the measurement in vivo of the rate of incorporation of [3H]water into sterols by all the organs,
including nine different regions of the brain and a
section of the spinal cord. The rate of sterol synthesis in
each organ, expressed per gram wet tissue, is shown in
Fig. 2A. Although the adrenal gland manifested an
exceptionally high rate of synthesis, all the other
tissue growth was derived from plasma LDLs. This was
done by measuring in vivo the clearance of radiolabeled
homologous LDL by various regions of the brain in
groups of early and late neonatal lambs. For comparative purposes, the clearance of LDL by the liver was
measured in these same animals. The data in Fig. 1
show that in the older lambs, hepatic LDL clearance
(27 6 4 µl · h21 · g21 ) was far lower than in the early
neonatal group (85 6 11 µl · h21 · g21 ). A distinctly
different result was seen in the brain where no region
manifested LDL clearance rates of more than ,0.5 µl ·
h21 · g21 in either group of animals. The values given in
Fig. 1 for brain represent the average clearance rates
for all nine regions. With the assumption of an average
plasma LDL cholesterol concentration of 46 mg/dl and
an average whole brain mass of 66 g over this 15-day
Fig. 1. Rate of low-density lipoprotein (LDL) clearance by liver and
brain in neonatal lambs at 2 stages of development. Rates of LDL
clearance by liver and brain in 5 animals from each age group were
measured in vivo as described in MATERIALS AND METHODS. Age of
younger lambs (3 males, 2 females) ranged from about 1 to 3 days,
and older group (2 males, 3 females) was aged 16–18 days. These
values represent microliters of plasma cleared of its LDL content per
hour per gram of tissue. Values for brain are averages of clearance
rates found in 9 separate regions. Each point represents data from a
single animal.
Fig. 2. Rate of sterol synthesis in liver and extrahepatic organs in
early neonatal lambs. Lambs were given an intravenous bolus
injection of [3H]water and killed 1 h later. [3H]sterol content of liver
and several extrahepatic organs including brain and whole remaining carcass was then determined as described in MATERIALS AND
METHODS. These contents were taken as a direct measure of rate of
sterol synthesis in each tissue (A), which was expressed as nmol of
[3H]water incorporated into sterols per hour per gram of tissue. This
rate was used in turn to calculate whole organ sterol synthesis (B).
Whole animal sterol synthesis was determined as sum of synthesis in
all organs normalized per kilogram body wt (whole animal sterol
synthesis 5 52.4 6 5.9 µmol · h21 · kg body wt21 ). Values represent
means 6 SE of data from 13 lambs (11 males, 2 females) ranging in
age from 0.2 to 2.5 days.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
Values represent means 6 SE of data from combined number of
males and females specified. After natural birth, lambs had unrestricted access to their dams until they were used in various
metabolic studies. Age range of early and late neonatal groups was
0.2–3 days and 16–18 days, respectively. * P , 0.05 vs. corresponding
value for early neonatal group.
E1101
E1102
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
Fig. 3. Rate of sterol synthesis and cholesterol concentration (conc) in different regions
of brain in early neonatal lambs. Throughout
brain, cholesterol was detected only in unesterified form. These data were derived from
study described in Fig. 2. Values represent
means 6 SE of data from 13 animals.
ally, the lambs that had suckled longer had higher total
cholesterol concentrations in their livers, with the
additional cholesterol being present only in the esterified fraction. The relationship between the rate of
synthesis and the respective concentrations of esterified and unesterified cholesterol in the livers of all 13
lambs is shown in Fig. 4. In those animals that had
suckled the most, the concentration of esterified cholesterol increased severalfold, and this was accompanied
by a dramatic compensatory reduction in the rate of
hepatic cholesterol synthesis (Fig. 4A). There was no
correlation between synthesis and unesterified cholesterol levels, which generally remained at 2.5–3.0 mg/g
irrespective of how long the lambs had suckled (Fig.
4B). Thus, unlike the brain, in which cellular cholesterol content was determined by the rate at which
cholesterol was synthesized locally, in the liver, the rate
of synthesis changed in accordance with cellular cholesteryl ester content, which in turn varied as a function of the amount of chylomicron cholesterol that was
delivered to the liver from the small intestine.
DISCUSSION
In the adult human, about one-third of all the
cholesterol in the body is present in the CNS, where it
is contained primarily within myelin, the formation of
which is completed in the early stages of development
(7, 21). Although developing CNS tissue manifests a
very high rate of cholesterol synthesis (29), the question of how much brain cholesterol ordinarily originates
from plasma lipoproteins remains the subject of investigation. The expression of mRNA for the LDL receptor in
the brains of neonatal rabbits and rats and the finding
that transcytosis of LDL across the blood-brain barrier
occurs under in vitro conditions suggest that cholesterol from LDL might be utilized for CNS tissue growth
in the developing animal (10, 24, 31). The present
studies were designed to determine the magnitude of
this contribution by applying a primed constant infusion technique to directly measure, in vivo, the rate of
LDL clearance by all the major regions of the brain in
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
organs also exhibited significant levels of biosynthetic
activity. The rate for the brain, which represents the
average of the values for all the regions studied, was
higher than that for all other organs except the adrenal
and small intestine. These synthesis data were also
expressed on a whole organ basis (Fig. 2B). Although
the bulk of the synthetic activity was manifest in the
tissues of the residual carcass, the brain accounted for
,4.5% of whole animal sterol synthesis. With the use of
the value for the rate of [3H]water incorporation into
sterol (Fig. 2B) together with a previously determined
value for the number of 3H atoms that are incorporated
into the sterol molecule per carbon atom entering the
biosynthetic pathway as acetyl-CoA (13), it can be
calculated that these lambs (average body wt 4.2 kg)
synthesized a total of ,165 mg of cholesterol/day.
Within the whole brain itself, ,7–8 mg of cholesterol
were synthesized daily. This largely accounts for the
daily increment of 9–10 mg in whole brain cholesterol
content (Table 1).
The importance of de novo synthesis as a source of
cholesterol for brain development was evident not only
from the similarity in the values for the amount of
cholesterol that the brain synthesized and laid down in
new tissue each day but also from the remarkably close
correlation (r 5 0.95) that existed between tissue
cholesterol concentration and the rate of cholesterol
synthesis throughout all the major regions of the brain
and spinal cord (Fig. 3). The values for both synthesis
and concentration varied about threefold throughout
the CNS, with those regions richest in white matter
manifesting the highest levels of sterol synthetic activity and the greatest cholesterol content.
The relationship between tissue cholesterol content
and synthesis in the brain was different than in the
other organs, particularly the liver. Unlike brain tissue,
which contained only unesterified cholesterol, in the
liver, both unesterified cholesterol and esterified cholesterol were present. Among the 13 early neonatal animals used in these sterol synthesis experiments, some
were studied as soon as 0.2 days after birth, whereas
others were ,2.5 days old at the time of study. Gener-
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
E1103
Fig. 4. Rate of sterol synthesis in liver as a
function of hepatic esterified and unesterified
cholesterol concentrations in early neonatal
lambs. These data were derived from study
described in Fig. 2. Each point represents data
from a single animal.
levels do not increase, whereas in other organs, particularly the liver and spleen, the levels are markedly
elevated (23). Although the function of the LDL receptor in brain remains to be elucidated, another type of
lipoprotein receptor, gp330/megalin, which is a member
of the LDL-receptor gene family, is essential for normal
CNS development. In mice lacking this receptor, forebrain development is defective and the animals die
perinatally from respiratory problems (37). Thus there
is at least one kind of lipoprotein receptor that plays a
crucial role in transporting essential nutrients needed
during early development.
The data from the sterol synthesis studies in lambs
aged 0.2–2.5 days established that the bulk, if not all,
of the 9–10 mg of cholesterol used daily for new brain
growth was synthesized locally. In all organs, including
the brain, the absolute rates of cholesterol synthesis
were calculated from several experimentally determined values, including one for the average SA of water
in the plasma over the 60-min period after administration of the [3H]water bolus. In the calculations for the
brain, we assumed that the average SA of CSF water
over the 60 min after injection of the [3H]water was the
same as that for plasma water. Although the SA of
water in the plasma and CSF was the same at the end
of the 60-min circulation time, it is possible that the
equilibration of the [3H]water with CSF water was
slower than that with plasma water. Therefore, to the
extent that this might have been the case, the average
SA of intracellular water in brain tissue would have
been overestimated. In turn, the absolute rates of
cholesterol synthesis would have been slightly underestimated. Although any such error would likely have
been small, it should nevertheless be recognized that
the value of 7–8 mg of cholesterol synthesized by the
whole brain per day is possibly an underestimate of the
true amount of sterol that was generated. Be that as it
may, the near-perfect positive correlation between the
rate of cholesterol synthesis and the tissue concentra-
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
the suckling lamb. By studying lambs just after birth
and again at ,2 wk postpartum, we were able to
directly compare the daily increment in CNS tissue
cholesterol content with the amount of LDL cholesterol
cleared from the plasma by the whole brain. In neither
the 1- nor the 16-day-old lambs did any region of the
CNS manifest LDL clearance rates above 0.5 µl·h21 ·g21.
Values ,1 µl · h21 · g21 are within the error of measurement of this method. Even if these low clearances
represented true net LDL cholesterol uptake, it can be
calculated that, at most, only 3–4% of the 9–10 mg of
cholesterol utilized daily for brain growth during this
phase of the lamb’s development could have come from
LDL. Such a minimal contribution is also apparent
from the finding that there were no regional differences
in LDL clearance, yet there was a threefold difference
in cholesterol concentration throughout different parts
of the brain. Two points should be emphasized here.
First, the trivial rates of LDL clearance by the brain
were found in the same animals in which the liver
actively transported LDL at rates that were regulated
in accordance with the amount of cholesterol absorbed
from the gastrointestinal tract (14). Second, the lack of
significant LDL transport by the brain in these lambs
cannot be attributed to suckling because similar results
were found previously for fetal sheep, even at gestational ages preceding closure of the blood-brain barrier
(33).
These clearance data provide direct evidence in vivo
that the supply of cholesterol for new CNS tissue
growth in the neonate is not dependent on the receptormediated transport of LDL out of the plasma. This
conclusion is fully consistent with the previous finding
that brain growth and development are normal in mice
and rabbits genetically lacking the LDL receptor (12,
25). It also fits well with the observation that in young
mice with Niemann-Pick (type C) disease, in which
there is a defect in the mechanism for redistributing
LDL cholesterol from the lysosomes, brain cholesterol
E1104
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
important steps in better understanding the causes of
neurological disease in humans.
We thank Mark Herndon, Jeffrey Graven, Trista Prasifka, and
Brian Darnell for excellent technical assistance and Merikay Presley
for expert preparation of the manuscript.
These studies were supported by National Heart, Lung, and Blood
Institute Grant HL-09610 and a grant from the Moss Heart Fund.
Address for reprint requests: S. D. Turley, Dept. of Internal
Medicine, Univ. of Texas Southwestern Medical Center, 5323 Harry
Hines Blvd., Dallas, TX 75235-8887.
Received 7 January 1998; accepted in final form 5 March 1998.
REFERENCES
1. Acton, S., A. Rigotti, K. T. Landschulz, S. Xu, H. H. Hobbs,
and M. Krieger. Identification of scavenger receptor SR-BI as a
high density lipoprotein receptor. Science 271: 518–520, 1996.
2. Alling, C., and L. Svennerholm. Concentration and fatty acid
composition of cholesteryl esters of normal human brain. J.
Neurochem. 16: 751–759, 1969.
3. Blanchette-Mackie, E. J., N. K. Dwyer, L. M. Amende, H. S.
Kruth, J. D. Butler, J. Sokol, M. E. Comly, M. T. Vanier, J. T.
August, R. O. Brady, and P. G. Pentchev. Type-C NiemannPick disease: low density lipoprotein uptake is associated with
premature cholesterol accumulation in the Golgi complex and
excessive cholesterol storage in lysosomes. Proc. Natl. Acad. Sci.
USA 85: 8022–8026, 1988.
4. Burris, M. J., and C. A. Baugus. Milk consumption and growth
of suckling lambs. J. Anim. Sci. 14: 186–191, 1955.
5. Cavender, C. P., S. D. Turley, and J. M. Dietschy. Sterol
metabolism in fetal, newborn, and suckled lambs and their
response to cholesterol after weaning. Am. J. Physiol. 269
(Endocrinol. Metab. 32): E331–E340, 1995.
6. Clarenburg, R., I. L. Chaikoff, and M. D. Morris. Incorporation of injected cholesterol into the myelinating brain of the
17-day-old rabbit. J. Neurochem. 10: 135–143, 1963.
7. Cook, R. P. Cholesterol. Chemistry, Biochemistry, and Pathology. New York: Academic, 1958, p. 1–542.
8. Cooper, A. D. Hepatic uptake of chylomicron remnants. J. Lipid
Res. 38: 2173–2192, 1997.
9. Cruz, M. L., F. Mimouni, W. Wong, D. L. Hachey, P. Klein,
and R. C. Tsang. Effects of infant nutrition on cholesterol
synthesis rates in infancy (Abstract). Clin. Res. 39: 689A, 1991.
10. Dehouck, B., L. Fenart, M.-P. Dehouck, A. Pierce, G. Torpier, and R. Cecchelli. A new function for the LDL receptor:
transcytosis of LDL across the blood-brain barrier. J. Cell Biol.
138: 877–889, 1997.
11. De Silva, H. V., J. Más-Oliva, J. M. Taylor, and R. W. Mahley.
Identification of apolipoprotein B-100 low density lipoproteins,
apolipoprotein B-48 remnants, and apolipoprotein E-rich high
density lipoproteins in the mouse. J. Lipid Res. 35: 1297–1310,
1994.
12. Dietschy, J. M., T. Kita, K. E. Suckling, J. L. Goldstein, and
M. S. Brown. Cholesterol synthesis in vivo and in vitro in the
WHHL rabbit, an animal with defective low density lipoprotein
receptors. J. Lipid Res. 24: 469–480, 1983.
13. Dietschy, J. M., and D. K. Spady. Measurement of rates of
cholesterol synthesis using tritiated water. J. Lipid Res. 25:
1469–1476, 1984.
14. Dietschy, J. M., S. D. Turley, and D. K. Spady. Role of liver in
the maintenance of cholesterol and low density lipoprotein
homeostasis in different animal species, including humans. J.
Lipid Res. 34: 1637–1659, 1993.
15. Edmond, J., R. A. Korsak, J. W. Morrow, G. Torok-Both, and
D. H. Catlin. Dietary cholesterol and the origin of cholesterol in
the brain of developing rats. J. Nutr. 121: 1323–1330, 1991.
16. Hamilton, J. G., and K. Comai. Rapid separation of neutral
lipids, free fatty acids and polar lipids using prepacked silica
Sep-Pak columns. Lipids 23: 1146–1149, 1988.
17. Holstein, T. J., W. A. Fish, and W. M. Stokes. Pathway of
cholesterol biosynthesis in the brain of the neonatal rat. J. Lipid
Res. 7: 634–638, 1966.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
tion of cholesterol across all regions of the brain provides further compelling evidence that the requirement
for additional sterol by the rapidly growing CNS of the
neonate is met fully through local synthesis. This
conclusion is in agreement with the results of studies in
the rat and rabbit (6, 15, 18) and with data from
children with the Smith-Lemli-Opitz syndrome, in
which an inherent block in the terminal enzymatic step
in the sterol biosynthetic pathway causes profound
changes in brain cholesterol metabolism that result in
major neurological dysfunction (28).
In addition to providing a measure of the quantity of
cholesterol generated daily by the brain, the data from
the sterol synthesis experiments also showed how
different the regulation of cholesterol levels in CNS
tissue is from that in the other major organs, particularly the liver. From other published data, it can be
calculated that the dietary cholesterol intake of the
lambs in the present studies was ,180 mg · day21 · kg
body wt21 (4, 5), which exceeded by more than fourfold
their rate of whole body cholesterol synthesis (39.2
mg · day21 · kg body wt21 ). This resulted in an expansion
of the cholesteryl ester pool in the liver and a striking
compensatory downregulation of local cholesterol synthesis. This response has been well documented in
several species (36). The lack of any correlation between unesterified cholesterol content and the rate of
cholesterol synthesis in the liver was in marked contrast to the relationship between these two parameters
in the brain. Although developing CNS tissue has been
widely reported to contain minute amounts of esterified
cholesterol (2, 6, 20, 27), none was detected in any
region of the brain in these young neonatal lambs.
Irrespective of the reason for this difference, it is clear
from the present data that tissue cholesterol content in
the CNS is dictated by the rate of local cholesterol
synthesis. In contrast, in the liver and all the major
extrahepatic organs, the rate of sterol synthesis is
adjusted in accordance with the intracellular cholesteryl ester content, which fluctuates depending
mainly on the amount of cholesterol that is transported
into the cells through the clearance of lipoproteins from
the plasma (14, 36).
One additional feature of cholesterol metabolism in
the CNS is that, although its rate of cholesterol synthesis falls precipitously after myelination is complete, the
brain continues to synthesize sterol at a very low basal
rate in adulthood (30). Recent studies involving the
measurement of cholesterol turnover in the brain of
adult humans demonstrated not only that the minute
daily fractional loss of sterol was facilitated through its
conversion to 24S-hydroxycholesterol but that the quantity of cholesterol lost from the mature brain is within
the range of the theoretical amount of sterol that it
synthesizes, based on data from a primate model (22).
The determination of what regulates the conversion of
cholesterol to this particular hydroxylated derivative
and the elucidation of the roles that each of the
lipoprotein receptors play in regulating cholesterol
homeostasis in the developing and mature brain will be
ORIGIN OF BRAIN CHOLESTEROL IN SHEEP
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
the Smith-Lemli-Opitz syndrome. J. Lipid Res. 37: 1169–1180,
1996.
Snipes, G. J., and U. Suter. Cholesterol and myelin. In:
Subcellular Biochemistry edited by R. Bittman. New York, NY:
Plenum, 1997, vol. 28, chapt. 7, p. 173–204.
Spady, D. K., and J. M. Dietschy. Sterol synthesis in vivo in 18
tissues of the squirrel monkey, guinea pig, rabbit, hamster and
rat. J. Lipid Res. 24: 303–315, 1983.
Swanson, L. W., D. M. Simmons, S. L. Hofmann, J. L.
Goldstein, and M. S. Brown. Localization of mRNA for low
density lipoprotein receptor and a cholesterol synthetic enzyme
in rabbit nervous system by in situ hybridization. Proc. Natl.
Acad. Sci. USA 85: 9821–9825, 1988.
Turley, S. D., J. M. Andersen, and J. M. Dietschy. Rates of
sterol synthesis and uptake in the major organs of the rat in vivo.
J. Lipid Res. 22: 551–569, 1981.
Turley, S. D., D. K. Burns, C. R. Rosenfeld, and J. M.
Dietschy. Brain does not utilize low density lipoproteincholesterol during fetal and neonatal development in the sheep.
J. Lipid Res. 37: 1953–1961, 1996.
Turley, S. D., and J. M. Dietschy. Mechanisms of LDLcholesterol lowering action of psyllium hydrophillic mucilloid in
the hamster. Biochim. Biophys. Acta 1255: 177–184, 1995.
Turley, S. D., M. W. Herndon, and J. M. Dietschy. Reevaluation and application of the dual-isotope plasma ratio method for
the measurement of intestinal cholesterol absorption in the
hamster. J. Lipid Res. 35: 328–339, 1994.
Turley, S. D., D. K. Spady, and J. M. Dietschy. Identification
of a metabolic difference accounting for the hyper- and hyporesponder phenotypes of cynomolgus monkey. J. Lipid Res. 38:
1598–1611, 1997.
Willnow, T. E., J. Hilpert, S. A. Armstrong, A. Rohlmann,
R. E. Hammer, D. K. Burks, and J. Herz. Defective forebrain
development in mice lacking gp330/megalin. Proc. Natl. Acad.
Sci. USA 93: 8460–8464, 1996.
Wilson, J. D. The measurement of the exchangeable pools of
cholesterol in the baboon. J. Clin. Invest. 49: 655–665, 1970.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.5 on June 16, 2017
18. Jurevics, H., and P. Morell. Cholesterol for synthesis of myelin
is made locally, not imported into brain. J. Neurochem. 64:
895–901, 1995.
19. Kim, D.-H., H. Iijima, K. Goto, J. Sakai, H. Ishii, H.-J. Kim,
H. Suzuki, H. Kondo, S. Saeki, and T. Yamamoto. Human
apolipoprotein E receptor 2. J. Biol. Chem. 271: 8373–8380,
1996.
20. Kinney, H. C., J. Karthigasan, N. I. Borenshteyn, J. D. Flax,
and D. A. Kirschner. Myelination in the developing human
brain: biochemical correlates. Neurochem. Res. 19: 983–996,
1994.
21. Lentner, C. Geigy Scientific Tables. West Caldwell, NJ: CibaGeigy, 1981, p. 223.
22. Lütjohann, D., O. Bruer, G. Ahlborg, I. Nennesmo, Å. Sidén,
U. Diczfalusy, and I. Björkhem. Cholesterol homeostasis in
human brain: evidence for an age-dependent flux of 24Shydroxycholesterol from the brain into the circulation. Proc.
Natl. Acad. Sci. USA 93: 9799–9804, 1996.
23. Miyawaki, S., S. Mitsuoka, T. Sakiyama, and T. Kitagawa.
Sphingomyelinosis, a new mutation in the mouse. J. Hered. 73:
257–263, 1982.
24. Ness, G. C. Developmental regulation of the expression of genes
encoding proteins involved in cholesterol homeostasis. Am. J.
Med. Genet. 50: 355–357, 1994.
25. Osono, Y., L. A. Woollett, J. Herz, and J. M. Dietschy. Role of
the low density lipoprotein receptor in the flux of cholesterol
through the plasma and across the tissues of the mouse. J. Clin.
Invest. 95: 1124–1132, 1995.
26. Pitas, R. E. Cerebrospinal fluid lipoproteins, lipoprotein receptors, and neurite outgrowth. Nutr. Metab. Cardiovasc. Dis. 7:
202–209, 1997.
27. Ramsey, R. B., and A. N. Davison. Steryl esters and their
relationship to normal and diseased human central nervous
system. J. Lipid Res. 15: 249–255, 1974.
28. Salen, G., S. Shefer, A. K. Batta, G. S. Tint, G. Xu, A. Honda,
M. Irons, and E. R. Elias. Abnormal cholesterol biosynthesis in
E1105