Oral and Intraperitoneal Administration of A/

Acetylneuraminic Acid: Effect on Rat Cerebral

and Cerebellar /V-Acetylneuraminic Acid1

SUSAN E. CARLSON AND STEPHEN G. HOUSE

Department of Pediatrics/Division of Newborn Medicine,

University of Mississippi Medical Center, 2500 N. State St.,
Jackson, MS 39216

ABSTRACT Rat pups were administered N-acetylneuraminic acid (NANA) by

Downloaded from jn.nutrition.org by on May 8, 2009

i.p. injection or via a feeding catheter for eight consecutive days beginning on d 14
of life All pups were given the identical dose: 1.0 mg on d 1 and 2 and 1.2 mg on the
remaining days, or approximately 20 mg/kg body weight per day. A control group was
injected i.p. with glucose instead of NANA. On the morning of d 25, pups were de
capitated, and the heads were frozen immediately in liquid nitrogen. The brains were
later dissected and analyzed for both cerebral and cerebellar ganglioside and glyco-
protein NANA. Administration of NANA by both oral and i.p. routes resulted in sig
nificantly more cerebral and cerebellar ganglioside and glycoprotein NANA than did
glucose injection (with the exception of cerebellar glycoprotein NANA after NANA
intubation). There were no significant differences in NANA concentration in these
brain fractions for the two routes of NANA administration. J. Nutr. 116: 881-886,
1986.
INDEXING KEY WORDS gangliosides •glycoproteins •N-acetylneura-
minic acid •cerebrum •cerebellum

N-Acetylneuraminic acid (NANA) found radiolabeled NANA (4). The effect of re

in milk oligosaccharides accounts for, on the
average, 0.1 % (wt/vol) of the diet of ex-
clusively breast-fed infants in the first 2 wk
of life (1), and while the amount declines
exponentially throughout the first 2 mo of
lactation (1), it continues to be a significant
component of human milk throughout at
least 7 mo of lactation. The potential nutri-
t ¡onalsignificance of this material to the
human infant appears not to have been
studied. Morgan and Winick (2) reported
that NANA administered i.p. during a peri-
od of maximal brain ganglioside accumula
tion (3)
v'_. increased
.. the amount
- i ti of NANA i.
recovered in cerebral and cerebellar gangli-
osides and glycoproteins. Subsequently, they
j j .f , by the Biomédical Research Support Grant Program, Division of Research
demonstrated that the SynaptOSOmeS
the major Site Of accumulation
J Pediatrics, University of South Florida College of Medicine, 12901N. 30th
material following i.p. administration
peated oral administration during periods
of maximal brain ganglioside accumulation
has not been studied. Administration of a
single oral dose of radiolabeled NANA (5-7)
to young rats, however, indicates that only
small amounts of the label appear in organs
and tissues, while the bulk of the material is
apparently absorbed from the intestine and
excreted in the urine within a matter of
hours. Furthermore, Nöhleand Schauer (6)
suggest that the N-acetylmannosamine por
tion of NANA, and not NANA itself, is the
©1986 American Institute of Nutrition. Received for publication:
w Maywss.Accepted
forpublication:
23December
ms.
'This
project
wassupported
byBRSG cram2SOT
RROWQawarded
Were Resources,National Institutesof Health, to the Universityof South Florida
Of labeled £°!leSeof Medicine. The work was performed in the Department of
of the st.,Tampa,FL 33612.

881
882 CARLSON AND HOUSE
source of most label recovered in tissues.
Although these reports suggest that NANA
occurring in food is not available in signifi
cant quantities for synthesis of glycolipids
and/or glycoproteins, the degree of accumu
lation of dietary NANA by nursing mammals
has not been studied. In this study, free
NANA was administered to young rats by
feeding catheter for 8 d in an amount simi
lar to that injected by Morgan and Winick
(2) to determine whether repeated admin
istration of dietary NANA could increase
the NANA concentration of cerebral and
cerebellar gangliosides and glycoproteins.
Another group of young rats was adminis
tered the same amount of NANA by the i.p.
route.
MATERIALS AND METHODS
Experimental design. Pregnant (dated,
sperm-positive) Sprague-Dawley rats were
purchased from Holtzman, Madison, WI,
and fed a nonpurified diet (Rodent Lab
Chow 5001, Ralston Purina Co., St. Louis,
MO) until d 15 of gestation. On d 15 the
dams were assigned to one of three groups
fed high fat (40 % of energy) diets, varying in
degree of fat polyunsaturation and linoleic-
linolenic acid ratio (8), or to a fourth group
that continued to receive the nonpurified
diet as a control for growth. Litters were
culled to eight young per dam on the day of
delivery, and pups in experimental groups
were weaned to the maternal diet at 17 d of
age. Those whose dams were fed the non-
purified diet were weaned at a more typical
21 d of age. They also received the maternal
diet after weaning.
The pups within each litter were subdi
vided by chance into three treatment groups
identified by indelible color codes applied to
the tail. Daily from 14 to 21 d of age these
groups received 1) i.p. injection of glucose in
normal saline (1 mg/d on d 14 and 15, 1.2
mg/d on d 16 through 21), 2) an i.p. injection
of the same quantity of NANA (crystalline,
synthetic; Sigma Chemical Co., St. Louis,
MO) in saline or 3) intragastrically, via a
feeding catheter, the same quantity of NANA
in water. All solutions of glucose and NANA
were prepared fresh daily. It was planned
that another control group would receive
glucose by feeding catheter, but about one-
third of the animals ordered did not deliver
pups.
At 25 d of age, pups were anesthetized
with a combination of ketamine • HC1 (Keta-
set, Rristol Laboratories, Syracuse, NY) and
acepromazine maléate(Med-Tech, Inc., El-
wood, KS) and exsanguinated by heart punc
ture They were decapitated, and the heads
were frozen immediately in liquid nitrogen
and stored at -75°C until analysis.
Analytical procedures. On the day of
analysis, the frozen rat heads were partially
thawed on ice, and the brains were dissected
free and weighed. The cerebrum and cere
bellum were then separated, weighed and
sonicated in distilled water at less than 5 °C.
Aliquota of the resulting homogenates were
lyophilized, and gangliosides were extracted
Downloaded from jn.nutrition.org by on May 8, 2009
from the dry tissue with constant shaking
according to the procedure of Roukema and
Heijlman (9). Glycoproteins remaining after
ganglioside extraction were twice reextracted
to remove contaminating glycolipid NANA.
Both ganglioside and glycoprotein fractions
were hydrolyzed in 0.1 N H2SO4 at 80 °C.
Gangliosides were hydrolyzed for l h and
glycoproteins for 2 h before analysis for
NANA according to the procedure of Warren
(10). The ganglioside hydrolysate was ex
tracted with peroxide-free ether prior to
analysis. Spectrophotometric readings were
made at 532, 549 and 562 nm, and correc
tions were made for interference at 532 nm
by the equations of dichromatographic read
ings (562 vs. 532 and 549 vs. 532 nm) of
Warren (10).
Statistical analysis. The design of the ex
periment was such that with four dietary
treatments and three NANA treatments,
each diet-NANA group contained 10-12 rat
pups. Analysis of the data indicated that the
degree of polyunsaturation of the diet and
the ratio of linoleic to linolenic acid did not
affect the concentrations of cerebral and
cerebellar NANA, and pups from all dietary
groups were grouped by NANA treatment
alone. Data for each NANA treatment are
presented as means ±SEM. Differences be
tween groups were determined by using the
Student two-tailed i-test. P-values were
multiplied by 2 for Bonferroni's adjustment
for two-treatment comparisons (11).Although
numbers of animals within each group dif
fered, results presented are from the analysis
 EXOGENOUS AND BRAIN N-ACETYLNEURAMINIC ACID
of all available samples. Dissection of the
complete brain, cerebrum and cerebellum
was impossible for some brains that frac
tured in liquid nitrogen. In some cases, the
amount of cerebral and/or cerebellar material
available was still adequate for determination
of NANA in micromoles/gram.
RESULTS
There were no significant differences
in the body, cerebral, cerebellar or whole-
brain weights in the three treatment groups
(table 1).
NANA concentrations of cerebral and cere
bellar gangliosides were higher in NANA-
treated rats than in rats injected with glu
cose (table 2). NANA-injection and intuba
tion also increased cerebral and cerebellar
glycoprotein (table 2). All means of treated
animals were different from control means
at a significance level of P < 0.001, except
those for cerebellar glycoprotein NANA: the
concentration of cerebellar glycoprotein
NANA in NANA-injected pups differed sig
nificantly (P < 0.02), while that of NANA-
intubated pups did not.
Gangliosides and glycoproteins accounted
for 80 and 20 %, respectively, of the exog
enous NANA incorporated into the brain.
NANA incorporation into cerebral gangli
osides in treated compared to control pups
was quantitatively the most important, ac
counting on average for an additional 0.30
/¿molNANA/brain. Cerebellar gangliosides,
cerebral glycoproteins and cerebellar glyco
proteins accounted for increments with
treatment of 0.07, 0.07 and 0.02 /imol each,
respectively. Thus, on the average, cerebral
and cerebellar NANA was higher in treated
Brain and body weights of experimental groups '
Group
Glucose-injectedN-Acetylneuraminic 1.6(38)82.4
injectedN-Acetylneuraminic
acid 1.7(42)82.4
acid intubated82.2
'Values are means ±SEM(n). 'There were no significant differences due to treatment.
883
compared to control pups 0.46 /tmol (140.8
/ig). This amounted to 1.53% of the NANA
administered during an 8-d period.
While 80% of the additional NANA in
treated compared to control groups occurred
in the cerebrum, the cerebellum showed the
greatest percentage increase. As a result of
treatment cerebellar ganglioside and glyco
protein NANA increased by 13-19%, whereas
cerebral ganglioside and glycoprotein NANA
increased by 7-11 %.
DISCUSSION
Depending on the stage of lactation, in
fants fed mother's milk receive (on average)
between 135 and 1475 mg oligosaccharide
NANA/L of milk while those fed currently
Downloaded from jn.nutrition.org by on May 8, 2009
available infant formulas receive between 0
and 72 mg of oligosaccharide NANA/L (1).
Furthermore, the NANA associated with
cow milk is known to be chemically differ
ent from that in human milk (12). Whereas
mammals have the capacity to synthesize
NANA from simple sugars and the metabolic
intermediate, phosphoenolpyruvate, there is
evidence that exogenous NANA may serve as
a substrate for sialydation of glycolipids and
glycoproteins. Uptake of label from exog
enous NANA into gangliosides and glyco
proteins has been demonstrated in cell cul
tures (13-15), tissue slices (16) and in whole
animals after i.v. (16) and i.p. (2) adminis
tration.
NANA administered via i.p. injection has
been shown to result in increases in NANA
in brain gangliosides and glycoproteins (2).
This is the first report that indicates that
orally administered NANA has the same ef
fect in developing rats. In previous studies,
TABLE 1
Weight of
Body Cerebrum Cerebellum Brain
± 0.01(34)1.01
± 0.004(35)0.204
± 0.02(37)1.47
±
± 0.01(38)1.01
± 0.004(38)0.201
± 0.02(37)1.47
±
1.7(47)1.02
± 0.01(43)0.200
± 0.004(37)1.44
± 0.02(40)
±
884 CARLSON AND HOUSE

TABLE 2
Cerebral and cerebellar ganglioside and glycoprotein N-acefy/neuromimc acid (NANA) concentrations1
Where
measured Treatment Ganglioside NANA

¡imol/gCerebrumCerebellumGlucose-injected
±0.05 (36) ±0.02 (34)
NANA-injected 3.36 ±0.06 (40) 1.05 ±0.02 (34)
NANA-intubatedGlucose-injected3.36
(43)1.86
±0.06 (40)0.62
1.03 ±0.02
±0.05 (34) ±0.04 (28)
NANA-injected 2.21 ±0.05 (37) 0.73 ±0.03 (26)
NANA-intubated3.03 2.19 ±0.04 (38)0.96 0.70 ±0.03 (29)
Values are means ±SEM(n). 2Means differing from glucose-injected controls by 0.07 /tmol/g are statis
tically different (P < 0.001) except for glycoprotein NANA in cerebellar tissue. The level of significance for NANA
injection in cerebellar glycoprotein was P < 0.02 while NANA intubation produced only a nonsignificant trend
(P < 0.10). P-values reflect Bonferroni's adjustment for two i-test comparisons (11).

bellar gangliosides and glycoproteins (25).

Downloaded from jn.nutrition.org by on May 8, 2009

the results of administering single doses of
labeled NANA (5-7) suggest that most orally
administered NANA is rapidly excreted in
the urine However, the use of weanling rats
and mice (6, 7), as opposed to animals in
which neural NANA-conjugates were rapidly
accumulating, could have led to an under
estimation of the potential importance of
dietary NANA to the nursing mammal. It is
clear that more precise experiments with
labeled NANA compounds administered
orally during periods of rapid central ner
vous system development are still necessary.
Data of both Witt and co-workers (5) and
Nöhle and Schauer (17) indicate that the
tissue uptake of labeled NANA may depend
on the form in which it is administered.
On account of their asymmetrical location
in the outer half of the membrane bilayer,
gangliosides are thought to play an impor
tant role as membrane receptors and cell
surface markers (18). NANA is an integral
part of each molecule of ganglioside and is
postulated to be the actual receptor for
neurotransmitters (19). Both essential fatty
acid deficiency and protein deficiency dur
ing development (20-22) have been shown
to decrease the amount of brain gangliosides
as measured by NANA concentration. These
reductions in brain ganglioside/NANA are
associated with impaired learning behavior
in rats (22). Antisera to brain gangliosides
alters behavior in rats (23). Also long-term
active avoidance conditioning and environ
mental stimulation have increased, respec
tively, brain ganglioside synthesis (24) and
NANA concentrations of cerebral and cere
The NANA administered in this study,
approximately 20 mg/kg body weight for
eight consecutive days, was responsible for
7-19% of that recovered in cerebral and
cerebellar gangliosides and glycoproteins.
Although this indicates that orally adminis
tered NANA can provide a significant por
tion of total-brain NANA, it may actually
underestimate the usual contribution of di
etary NANA to that of glycoproteins and
glycolipids for a number of reasons. 1) Rat
milk contains an unknown quantity of
NANA-lactose (26). All rats in this study and
in that of Morgan and Winick (2) were
suckled for between 17 and 21 d after birth,
which would dilute the contribution of ex
perimentally administered NANA to some
degree 2) NANA administered in these ex
periments [milligrams/(kilogram • day)] was
quantitatively small compared to that usually
received by infants fed human milk (1). The
average intake of oligosaccharide NANA by
infants fed only human milk would be 170
mg/kg per day in the first 2 wk of lactation,
falling to 20 mg/kg per day only after 10 wk
or more of lactation. 3) Administration of
NANA in this study occurred during the
second major period of brain ganglioside
NANA increase in rats (3). Orally adminis
tered NANA might be expected to be more
significant when administered during the
most rapid period of brain ganglioside ac
cumulation in the rat (0-10 d). 4) Exogenous
NANA might be more efficiently utilized as
a substrate for sialydation of membrane
gangliosides and glycoproteins if the same
 EXOGENOUS AND BRAIN JV-ACETYLNEURAMINIC ACID
dose were administered in several aliquots
throughout the day rather than at a single
time point.
Since most brain NANA accumulates be
tween wk 25 of gestation and term, dietary
NANA such as obtained from human milk
might be of even greater significance to the
very premature infant. NANA concentrations
in preterm human milk are equivalent to
those of term milk, at least in the early
weeks of lactation (1).
Although NANA administered in this ex
periment was free and that in milk is in the
form of oligosaccharides and glycoproteins,
it seems reasonable to suppose that milk
oligosaccharides and glycoproteins are di
gested to their sugar and amino acid compo
nents in the newborn gut, and that NANA
can become available as a substrate for si-
alydation of gangliosides and glycoproteins.
Nevertheless, the disposition of dietary oligo
saccharides and glycoproteins by the gut of
the newborn infant requires further study.
Neither our study nor that of Morgan and
Winick (2) give any clues as to whether the
increase of NANA in brain gangliosides and
glycoproteins is caused by increased numbers
of molecules of these compounds or merely
to increased sialydation of existing molecules.
Although measurement of NANA concen
tration is the standard method for quantitat-
ing ganglioside concentration, gangliosides
themselves are classified on the basis of
degree of sialydation, and it is well known
that the composition of mono-, di-, and tri-
saccharides in brain changes during develop
ment (27).
LITERATURE CITED
1. Carlson, S. E. (1985) N-acetylneuraminic acid
concentrations in human milk oligosaccharides
and glycoproteins during lactation. Am. ]. Clin.
Nutr. 41, 720-726.
2. Morgan, B. L. G. & Winick, M. (1980) Effects
of administration of ÕV-acetylneuraminic acid
(NANA) on brain NANA content and behavior. J.
Nutr. 110, 416-424.
3. Vanier, M. T., Holm, M., Ohman, R. & Svenner-
holm, L. (1971) Developmental profile of gangli
osides in human and rat brain. J. Neurochem. 18,
581-592.
4. Morgan, B. L. G. & Winick, M. (1981) The
subcellular localization of administered JV-acetyl-
neuraminic acid in the brains of well fed and
undernourished rats. Br. J. Nutr. 46, 213-238.
885
5. Witt, W., von Nicolai, H. & Zilliken, F. (1979)
Uptake and distribution of orally applied AP-acetyl
[14C]neuraminosyllactose and N-acetyl['*C]neura-
minic acid in the organs of newborn rats. Nutr.
Metab. 23, 51-61.
6. Nöhle,U. & Schauer, R. (1981) Uptake, metab
olism and excretion of orally and intravenously ad
ministered, 14C- and 3H-labeled N-acetylneura-
minic acid mixture in the mouse and rat. Hoppe-
Seyler's Z. Physiol. Chem. 362, 1495-1506.
7. Nöhle,U., Beau, J.-M. & Schauer, R. (1982) Up
take, metabolism and excretion of orally and intra
venously administered, double-labeled N-glycoloyl-
neuraminic acid and single labeled 2-deoxy-2,3-
dehydro-N-acetylneuraminic acid in mouse and
rat. Eur. J. Biochem. 126, 543-548.
8. Carlson, S. E., Carver, J. D. & House, S. G. (1986)
High fat diets varying in ratios of polyunsaturated
to saturated fatty acid and linoleic to linolenic
acid: a comparison of rat neural and red cell mem
brane phospholipids. J. Nutr. 116, 718-725.
Downloaded from jn.nutrition.org by on May 8, 2009
9. Roukema, P. A. & Heijlman, J. (1970) The re
gional distribution of sialoglycoproteins, ganglio
sides and sialidase in bovine brain. J. Neurochem.
17, 773-780.
10. Warren, L. (1959) The thiobarbituric acid as
say of sialic acids. J. Biol. Chem. 234, 1971-1975.
11. Miller, R. P. (1966) Simultaneous Statistical
Inference, McGraw-Hill, New York.
12. Zilliken, F., Braun, G. A. & Gyorgy, P. (1956)
"Gynaminic acid" and other naturally occurring
forms of JV-acetylneuraminic acid. Arch. Biochem.
Biophys. 63, 394-402.
13. Hirschberg, C. B., Goodman, S. R. & Green, C.
(1976) Sialic acid uptake by fibroblasts. Biochemis
try 25, 3591-3599.
14. Hirschberg, C. B. & Goodman, S. R. (1976) Free
sialic acid uptake by fibroblasts and subsequent
incorporation into glycoproteins and glycolipids.
J. Cell Biol. 70, 1492.
15. Hirschberg, C. B. & Yeh, M. (1977) Sialic acid
uptake by BHK cells and subsequent incorporation
into glycoproteins and glycolipids. J. Supramol.
Struct. 6, 571-577.
16. Carey, D. J. & Hirschberg, C. B. (1979) Metab
olism of N-acetylneuraminic acid in mammals.
Isolation and characterization of CMP-N-acetyl-
neuraminic acid. Biochemistry 18, 2086-2092.
17. Nöhle,U. & Schauer, R. (1984) Metabolism of
sialic acids from exogenously administered sialyl-
lactose and mucin in mouse and rat. Hoppe-Seyler's
Z. Physiol. Chem. 365, 1457-1467.
18. Berrà , B., Lindi, C., Omodeo-Salè, F., Beltrame,
D. & Cantone, A. (1981) Effect of maternal
diet on ganglioside distribution in fetal rat brain.
J. Nutr. Ill, 1980-1984.
19. Schengrund, C. L. & Nelson, J. T. (1975) Influ
ence of cation concentration on the sialidase ac
tivity of neuronal synaptic membranes. Biochem.
Biophys. Res. Commun. 63, 217-223.
20. Karlsson, I. & Svennerholm, L. (1978) Biochemi
cal development of rat forebrain in severe protein
and essential fatty acid deficiencies. ]. Neurochem.
32, 657-662.
21. Merat, A. & Dickerson, J. W. T. (1974) The
effect of severity and timing of malnutrition on
886 CARLSON AND HOUSE
brain gangliosides in the rat. Biol. Neonate 25,
158-170.
22. Morgan, B. L. G., Oppenheimer, J. & Winick, M.
(1981) Effects of essential fatty acid deficiency
during late gestation on brain N-acetylneuraminic
acid metabolism and behavior in the progeny. Br.
J. Nutr. 46, 223-230.
23. Irwin, L. N. & Sampson, F. E. (1971) Content
and turnover of gangliosides in rat brain following
behavioral stimulation. J. Neurochem. 18, 203-211.
24. Savaki, H. E. & Lewis, G. M. (1977) Changes
in rat brain gangliosides following active avoidance
conditioning. Pharmacol. Biochem. Behav. 7, 7-12.
25. Morgan, B. L. G. & Winick, M. (1980) Effects
of environmental stimulation on brain N-acetyl-
neuraminic acid content and behavior. J. Nutr.
110, 425-432.
26. Trucco, R. E. & Caputto, R. (1954) Neuramin-
lactose, a new compound isolated from the mam
mary gland of rats. J. Biol. Chem. 206, 901-909.
27. Tettamanti, G. (1971) Brain gangliosides in
development. Adv. Exp. Med. Biol. 13, 75-89.
Downloaded from jn.nutrition.org by on May 8, 2009