Published December 11, 2014
Use of Oral Tolerance Tests to Investigate
Disaccharide Digestion in Neonatal Foals'r2
L. Rice*, E. A. Ottt, D. K. Beede*, C. J. Wilcox*,
E. L. Johnson+, S. Liebt, and P. Bonuna
Departments of *Biochemistry and Molecular Biology,
+Animal Science, *Dairy Science, and §Food Science and Human Nutrition,
University of Florida, Gainesville 3261 1
ABSTRACT: Oral tolerance tests were performed
on 13 neonatal foals to determine their ability to
digest disaccharides on d 1, 3 and 5 postpartum.
Foals were assigned randomly to treatments consisting of 20% (wt/voll solutions of either maltose,
lactose, or sucrose, dosed at 1 g/kg of BW, or
glucose, dosed at .5 g/kg of BW. After a
2-h fast, a n initial blood sample was collected via
jugular catheter. Foals were administered the
appropriate solution orally, and blood was collected every 15 min for 1 h and then every 30 min
for 3 h. Plasma glucose increased after dosing with
lactose or glucose but not with sucrose. Plasma
glucose concentrations increased slightly on d 3
and 5 in foals dosed with maltose. These findings
suggest that although lactose is well digested by
neonatal foals, maltose is digested only slightly,
and sucrose is not digested by d 5. Results of this
experiment indicate that maltose and sucrose
would not be suitable for inclusion in artificial
diets for foals < 1 wk old. Oral tolerance tests
could be useful for determining the ability of
premature or sick foals with lactose intolerance to
digest alternate carbohydrate sources.
Key Words: Foals, Oral Tolerance Test, Disaccharides
J. Anim. Sci. 1992. 70:1175-1181
Introduction
Milk may not be well digested by foals (Tzipori
et al., 1984) or human infants (Mayne et al., 1986)
suffering from premature birth or enteric diseases
that can reduce the activity of brush border
disaccharidases, especially lactase. Therefore, artificial diets designed for lactose-intolerant foals
must contain a n alternative carbohydrate source.
Formulations in which the energy requirement is
met with monosaccharides, such as glucose, have
high osmolarity. Although clinical observations
suggest that diets with high osmolarity can result
in diarrhea, malabsorption, and possibly necrotizing enterocolitis when fed to compromised human
neonates (Lebenthal et al., 19831, this has not been
documented in controlled experiments with foals.
'Florida Agric. Exp. Sta. Journal article no. R01304.
2Authors gratekdly acknowledge the staff of the Horse Res.
Center for their care of the animals and Me1 Tooker for his
technical assistance.
Received January 28, 1991.
Accepted November 15, 1991.
Disaccharides such as maltose and sucrose can
be digested by newborn human infants (Mobassaleh et al., 1985). Work with healthy foals indicates that maltase and sucrase activities are low
at birth and increase slowly during the first few
months postpartum (Roberts, 1974, 1975). However,
the pattern of development of disaccharidase
activity of foals < 1 wk old is not known.
It has been shown that estimates of disaccharide digestion obtained using oral tolerance tests
agree well with results obtained using homogenized intestinal tissue samples from foals (Roberts,
19751, calves (Dollar and Porter, 19571, and infants
(Mayne et al., 1986). Therefore, the present experiment was designed using this noninvasive method
to determine the ability of neonatal foals to digest
certain disaccharides that might be included in
artificial diets.
Experimental Procedure
AnimaZs. Thirteen foals (1 1 Quarter Horses and 2
Thoroughbreds) and their dams were maintained
on bahiagrass (Paspalurn notaturn) pasture and a
1175
RICE ET AL.
1176
Table 1. ANOVA model for changes in plasma glucose
Source
Treatment
Sex
Treatment x sex
Foal (treatment x sex)
Day
Treatment x day
Sex x day
Treatment x sex x day
Foal (treatment x s e d x day
Time
Treatment x time
Sex x time
Foal (treatment x sex) x time
Day x time
Treatment x day x time
Sex x day x time
Treatment x sex x day x time
Remainder
supplemental, balanced concentrate mix to meet
the mares’ requirements (NRC, 1989). Foals nursed
within 2 h after birth and were allowed to nurse
normally for the first 6 h postpartum. Foals were
weighed at 6 to 8 h postpartum and on d 5. At
birth, foals were blocked by sex and assigned at
random to four treatment groups. Foals remained
with their dams during the trial periods.
Blood Sampling and Treatment Administration. On
d 1, at 6 h postpartum, each foal was aseptically
fitted with a teflon jugular vein catheter (QuickCath, Travenol Laboratories, Deerfield, ILI to
allow blood sampling. To minimize stress to the
foals, a small amount of local anesthetic was
injected at the catheter site. Plastic wings of the
catheters were fixed to the skin with a n adhesive.
This was removed easily with acetone at the
conclusion of each trial period. The catheters were
filled with sterile, heparinized, normal saline between samplings. With the catheters in place, the
foals were allowed to return to the mares and
resume normal activity. When the foals completed
a nursing bout, they were muzzled. After a
2-h fast, blood was collected via the catheter (time
= 0). At this time, the muzzles were removed to
permit oral administration of treatment solutions.
Treatments were 20% (wt/vol) solutions of either
n = 41, lactose (L,n = 31,or sucrose (S,
maltose (M,
n = 3)dosed at 1 g/kg of BW or glucose (G,
n = 3)
dosed a t .5 g/kg of B W . Treatment dosages were
similar to those used by Roberts (1975).
The solutions were offered to the foals in a
large-animal nursing bottle. Foals that did not
voluntarily consume test solutions were dosed via
dosing syringe or nasogastric tube. After treatment administration, the foals were remuzzled for
the duration of the trial period. Blood was collected every 15 min for 1 h and then every 30 min
for 3 h. The procedure was repeated on d 3 and 5
Test term
df
3
1
3
5
2
6
2
6
50
10
30
10
50
20
10
20
60
101
Foal (treatment
Foal (treatment
Foal (treatment
Remainder
Foal (treatment
Foal (treatment
Foal (treatment
Foal (treatment
Remainder
Foal (treatment
Foal (treatment
Remainder
Remainder
Remainder
Remainder
Remainder
Remainder
x sex)
x sex)
x sex)
x sex) x
x sex) x
x sex) x
x sex) x
day
day
day
day
x sex) x time
x sex) x time
postpartum. Results of a preliminary trial indicated that plasma glucose concentrations reached
a low baseline in newborn foals 1 h after a nursing
bout. Therefore, a 2-h fast followed by a
4-h collection period was sufficient to monitor
changes in blood glucose concentration resulting
from administration of test solutions.
Analyses. Blood was collected in tubes containing EDTA and stored on ice until several tubes
were collected. The tubes were centrifuged at 3,500
x g for 20 min, and the plasma was removed and
frozen. Plasma glucose concentration was determined using the Trinder peroxidase method
(Sigma Chemical, St. Louis, MO), with absorbance
measured at 505 nm (DU-20 Spectrophotometer,
Beckman Instruments, Palo Alto, CAI.
Statistical Analyses. Statistical analyses were
performed using the GLM procedure of SAS
(Freund and Littell, 1981) with time as a class
variable (Snedecor and Cochran, 1909). The complete model is given in Table 1. The appropriate
error term is shown for each independent variable.
Equations for the time curves were obtained
using reduced models with time as a continuous
variable. The reduced, pooled model is presented
in Table 2. The models for comparison of day,
treatment, and day x treatment interaction curves
were similar to the pooled model except that day x
timen, treatment x timen, and treatment x day x
timen, respectively, were used in place of timen.
Tests for homogeneity of the time curves W tests)
were calculated based on the techniques described
by Snedecor and Cochran (1969). Orthogonal contrasts were used to determine differences among
the curves. The contrasts performed were L vs G, L
+ G vs M, and L + G + M vs S. Rejection of the
null hypothesis indicated that the curves differed,
that is, they were not parallel.
IN
DISACCHARIDE DIGESTION
NEONATAL FOALS
1177
Table 2. Pooled regression model for plasma glucose time curves (reduced model)
df
Source
Treatment
Sex
Treatment x sex
Foal (treatment x sex)
Test term
3
Foal (treatment x sex)
1
Foal (treatment x
Foal (treatment x
Remainder
Foal (treatment x
Foal (treatment x
Foal (treatment x
Remainder
Remainder
Remainder
Remainder
Remainder
Remainder
Remainder
3
5
2
0
Day
Treatment x day
Treatment x sex x day
Foal (treatment x sex) x day
Time
Time2
0
10
1
1
1
~ i m e ~
m e 4
1
1
1
Time6
Remainder
sex)
sex)
sex) x day
sex) x day
sex) x day
383
Results and Discussion
Foals ingested the treatment solutions by bottle
readily on d 1 but were often less cooperative
during subsequent treatment periods and were
dosed by syringe or nasogastric tube. The method
of treatment administration did not affect the
plasma glucose concentration.
Fasting Blood Glucose. The plasma samples collected from all foals at time = 0 were used to
determine fasting glucose concentrations. At time
= 0 no treatment had been given; therefore, data
from all foals were pooled. No breed or breed
interaction effects were found (P > ,051. Rogers et
al. (19831, studying foals from birth to 48 h of age,
reported higher blood glucose concentrations in
fillies than in colts, although results of statistical
analyses were not reported. However, we found no
differences due to sex in the 13 foals studied (P >
.891; therefore, data from both sexes were pooled.
Blood glucose concentration from fasted foals
was 4.14 mM on d 1. This was lower (P e .00011
than d 3 (6.08 mM) and d 5 (6.31 mM) glucose
concentrations, SEM = .27, which were not different from each other.
Plasma glucose concentrations in the present
trial were lower than those reported in the
literature for nursing foals up to 12 h old, which
ranged from 6.4 f .8 to 8.0 f 1.6 mM (Rose et al.,
1979; Bauer et al., 19841. Bauer et al. (19841
determined serum glucose to be 9.2 f 1.6 mM on d
1, 9.3 f 2.0 mM on d 3, and 9.0 f 1.0 mM on d 7.
Table 3. Polynomial equations for changes in plasma glucose over time
Regression coefficientss
Treatment
and day
Lactose
1
3
5
Maltose
1
3
5
Glucose
1
3
5
Sucrose
1
3
5
Intercept
5.22
7.01
6.85
1
.o 1
.10
.10
2nd
3rd
(1o - ~ )
(1o - ~ )
4th
dh
5th
r10-l~)
2.27
-3.17
-3.04
-3.98
1.03
1.81
2.53
.30
-.25
-.71
-.50
-.OB
7.50
10.20
2.20
3.51
6.19
0.03
1.40
-1.94
-3.00
-2.63
1.81
2.89
1.94
-.70
-1.38
-.05
.08
.12
.31
8.18
.23
-2.54
3.02
5.70
0.28
.17
.23
.13
-2.04
-11.40
-5.10
-1.38
19.81
0.92
3.10
-15.53
-4.41
-1.50
5.02
1.30
25.10
-70.23
-10.22
3.09
5.48
0.75
-. 12
3.41
.80
-1.30
4.23
2.51
2.58
-1.11
-.70
-1.09
.34
7.41
10.28
-3.95
.001
-.001
.04
-
%e
equation is represented by Y =
plasma glucose 0 and x
minutes, 0
-2.56
1.72
&-, + PIX + &$ +
< x .s 240.
$39
+
.OB
B4x4
+
$52
+
&jx6; y
-
1178
RICE ET AL.
Rose et al. (1979)reported foal plasma glucose
concentrations of 8.0 f .6mM a t 12 to 36 h and 8.4
f .4 mM at 1 to 4 wk postpartum. Many factors,
such as time after nursing, can affect blood
glucose concentration. These workers did not note
time from the last nursing bout to sample collection. Even though foals nurse frequently, this may
affect results.
Fasting blood glucose concentrations for d 3
(6.08mM) and d 5 (6.31mM) were greater than the
resting concentrations for adult horses (4.77 f .61
mM) reported by Roberts (1975). Diet, hormones,
and development of the intestinal tract may
influence these values.
Glucose Absorption Time Curves. Considering
time to be a continuous variable, equations for
treatment, day, and treatment x day curves were
calculated. Sequential addition of higher-order
terms showed time to the fifth power to be
significant (P e .05). However, fifth-order equations led to negative estimates near the right-hand
portion of some curves, so equations of the sixth
order then were used. Although sequential addition of higher-order terms revealed that the linear
term was sufficient to describe plasma glucose
E
U
al
o
VI
0
VI
lo-ol
GLUCOSE
time curves resulting from sucrose administration,
sixth-order equations were used in all instances so
that tests of heterogeneity of regression could be
made.
The equations for the treatment x day plasma
glucose time curves are given in Table 3. Only
treatment x day curves are presented because
they are needed to evaluate changes in response
to treatments over time on each day. Changes in
plasma glucose that occurred after dosing with
oral glucose are represented by the G time curves.
Similarly, L, MI and S curves represent changes in
plasma glucose in foals dosed with these sugars.
Comparison of Day x Treatment Curves. The
plasma G curves for d 1, 3, and 5 for each
treatment are given in Figures 1 through 3,
respectively. On cl 1, the M curve was different
from the L curve CP e .0015). The M curve
increased slightly, but did not have a peak during
the 4-h collection period. The L curve, which
peaked between 60 and 90 min after dosing, was
not parallel ( P e .0028) to the G curve, which
peaked between 30 and 45 min. This may be due to
the time required for lactose to be hydrolyzed or to
hepatic conversion of galactose to glucose
I
8.0
107
LACTOSE
6-01
2.0j
nE
"
3
0.a o
30
60
90
120
150
iao
210
0.0
240
0
30
Time after dosing (min)
n
E
90
120
150
180
210
240
Time after dosing (min)
O-O1
1
MALTOSE
I
60
::r:-y
SUCROSE
8.0 -
8.0-
W
W
6.0 -
0.0
0
Time after dosing (min)
30
60
90
120
150
180
210
240
Time a f t e r dosing (rnin)
Figure 1. Changes in plasma glucose (least squares point estimates) over time in 1-d-old foals fed solutions of
SE of Y at x.
either glucose (SE
. E ) , lactose (SE = .18], maltose (SE
.12), or sucrose (SE = .17j; SE
-
-
-
1179
DISACCHARIDE DIGESTION IN NEONATAL FOALS
(Roberts, 19751. The S curve decreased slightly
after time = 0 and remained almost flat during the
trial period. Data from d 1 suggest that glucose
and lactose administered orally in solution were
absorbed readily but that sucrose and maltose
were not readily absorbed.
Lactose and G curves on d 3 were similar to d-1
curves, except that the peaks occurred earlier in
the postadministration period, at 30 to 45 min and
at approximately 15 min, respectively. Age and
colostrum ingestion may influence digestion and
absorption efficiency. It has been shown in pigs,
puppies, and infants that colostrum intake has a
profound effect on the digestive tract, resulting in
increased length, weight, and absorptive surface
area and stimulated production of enzymes (Widdowson, 1984, 1985). The M curve rose slightly
during h 1 but had no peak. The M curve was not
parallel to the L curve (P c .0027) but did not differ
( P > .05)from the G curve. This may be due to
fluctuations seen in the right-hand portion of the G
curve, although a characteristic glucose-treatment
peak occurred during the first 30 min.
The d-5 L curve had a large peak between 30
and 60 min after treatment, and the G curve had a
small peak at 15 to 30 min, but the curves did not
differ (P > .051. The M curve was not different (P >
.05) from either the L or G curves. The d-5 S curve,
similar to the d-1 and d-3 curves, was almost flat
but was not different from the pooled G, L, and M
curves (P > .05).
Because it has been shown in calves (Dollar and
Porter, 1957) and foals (Roberts, 1975) that results
of oral tolerance tests correlate well with results of
small intestine tissue analysis for determining
relative enzyme activities, some conclusions can
be made concerning disaccharidases in foals.
Administration of L or G resulted in plasma
glucose peaks on all sampling days of the trial.
Roberts (19751 reported a n increase in plasma
glucose after young horses c 3 yr of age were
administered oral solutions of either lactose, glucose, or galactose. This indicated absorption and
metabolism of both products of lactose digestion.
In the present experiment, sucrose was not
readily absorbed. Age and prior exposure of the
gastrointestinal tract to sucrose on d 1 and 3 did
not result in increased plasma glucose concentrations on d 5. This agrees with the data of Roberts
(19741, who reported very low sucrase activity in
10.07
-- 1
LACTOSE
GLUCOSE
2
E 8.04
1
1
U
0 . 0 1 ,
I
0
30
,
I
I
,
60
'
1
'
I
120
90
I
'
'
I
180
150
"
210
240
0
30
Time a f t e r dosing ( m i n )
MALTOSE
180
210
240
SUCROSE
'O-O
1
8.0
0
$ 1
m
0
6.0
v
w
4.0
t
150
3
10.0,
-:1
120
90
60
Time after dosing (rnin)
4.0
1
J
zj 2.0
a
0.01
0
,
I
30
,
I
60
,
I
90
,
1
120
,
I
150
,
I
180
'
I
210
'
1
0.0
240
o
.
I
30
,
I
60
.
I
90
,
I
120
,
I
150
,
I
iao
,
,
210
'
I
240
Time a f t e r dosing (min)
T i m e a f t e r dosing ( m i n )
Figure 2. Changes in plasma glucose (least squares point estimates] over time in 3-d-old foals fed solutions of
either glucose (SE = .26], lactose (SE
.12), maltose (SE
.08),or sucrose (SE
.08);SE
SE Y at x.
-
-
-
-
1180
RICE ET AL.
10.0 7
10.0 1
Lactose
a,
6.01
u
7
-
L
I
2.0
a
0.0
,
I
I
,
30
0
.
60
1
'
I
'
120
90
I
'
I
150
'
I
180
'
0.0
*
210
0
240
30
Time after dosing (min)
60
90
120
150
180
210
240
Time after dosing (rnin)
O.O
1
SUCROSE
4.0-
F
-
2.0-
a
0.04
0
'
I
30
'
'
I
60
1
90
'
0
120
.
I
150
'
I
180
'
I
210
.
I
240
2.0L------
0.0 0
30
60
90
120
150
180
210
240
Time after dosing (rnin)
Time after dosing (rnin)
Figure 3. Changes in plasma glucose (least squares point estimates) over time in 5-d-old foals fed solutions of
either glucose (SE
.21If lactose (SE
.13)f maltose (SE
.13)f or sucrose (SE
.ll);SE
SE Y at x.
-
-
the small intestine of newborn foals. Dollar and
Porter (19571 could not detect sucrase activity in
neonatal calves, even with sucrose feeding. Sucrase activity does not appear in calves younger
than 44 d old (Huber et al., 19611. Although pigs
utilize glucose and lactose equally well at birth
(Dollar et al., 1957; Aherne et al., 19691, sucrase
activity does not develop until after the 1st wk
(Huber et al., 1961).
Maltose administration resulted in a slight rise
in plasma glucose concentration on d 1, and a
slight peak was observed on d 3 and 5. These data
suggest that maltase activity was very low at birth
and increased only slightly by d 5. This supports
work by Roberts (19741, who detected low maltase
activity in near-term equine fetuses and young
foals. Similarly, pigs (Cunningham and Brisson,
1957) and calves (Huber et al., 19611 are not able to
digest maltose during the 1st wk after birth.
Implications
Lactose, the normal substrate presented to the
intestinal mucosa in neonates, was well digested
and absorbed in foals used in this study. However,
-
-
-
artificial formulas designed for compromised foals
that may be lactose-intolerant require a n alternate
carbohydrate source. The use of monosaccharides
such as glucose may increase the osmolarity of the
diet beyond the amount tolerated by compromised
neonates. The lack of substantial peaks in plasma
glucose after ingestion of maltose and sucrose
suggests that maltase and sucrase activities were
not developed sufficiently to support the use of
maltose or sucrose in enteral formulas for foals e
1 wk old. Although healthy foals were used in this
trial, oral tolerance tests may be useful in determining appropriate carbohydrate sources for inclusion in artificial diets for lactose-intolerant
foals.
Literature Cited
Aherne, F.,V. W. Hays, R. C. Ewan, and V. C. Speer. 1969.
Absorption and utilization of sugars by the baby pigs. J.
A-.
Sci. 29:444.
Bauer, J. E., J. W. Harvey, R. L. Asquith, K. McNulty, and J.
Kivipelto. 1984. Clinical chemistry reference values of foals
during the first year of life. Equine Vet. J. 16:361.
Cunningham, H. M., and G. J. Brisson. 1957. The utilization of
DISACCHARIDE DIGESTION IN NEONATAL FOALS
maltose by newborn pigs. J. Anim. Sci. 16:574.
Dollar, A. M., K. G. Mitchell, and J.W.G. Porter. 1957. The
utilization of carbohydrates in the young pig. Proc. Nutr.
SOC.16:Xii.
Dollar, A. M., and J.W.G. Porter. 1957. The utilization of carbohydrates by the young calf. Nature 179:1299.
Freund, R.J.,and R.C. Littell. 1981.SAS for Linear Models. SAS
Inst. Inc., Cary, NC.
Huber, J. T., N. L. Jacobson, R. S.Allen, and P. A. Hartman.
1961.Digestive enzyme activities in the young calf. J. Dairy
Sci. 44:1495.
Lebenthal, E., P. C. Lee, and L. A. Heitlinger. 1983. Impact of
development of the gastrointestinal tract on infant feeding.
J. Pediatr. 102:l.
Mayne, A. J., G. A. Brown, D. Sule, and A. S. McNeish. 1986.
Postnatal development of disaccharide activities in jejunal
fluid of preterm neonates. Gut 27:1357.
Mobassaleh, M., R.K. Montgomery, J. A. Biller, and R.J. Grand.
1985.Development of carbohydrate absorption in the fetus
and neonate. Pediatr. 75:160.
1181
NRC. 1989.Nutrient Requirements of Horses (5th Ed.). National
Academy Press, Washington, DC.
Roberts, M. C. 1974. Amylase activity in the small intestine of
the horse. Res. Vet. Sci. 17400.
Roberts, M. C. 1975. Carbohydrate digestion and absorption
studies in the horse. Res. Vet. Sci. 1834.
Rogers, P. A., G. C. Fahey, and W. W. Albert. 1983. Blood
fructose levels in the newborn foal. Equine Vet. J. 15:60.
Rose, R.J., W. Backhouse, and W. Chan. 1979.Plasma biochemistry changes in Thoroughbred foals during the first four
weeks of life. J. Reprod. Fertil. (Suppl.) 27:601.
Snedecor, G., and W. Cochran. 1969. Statistical Methods (6th
Ed.). Iowa State University Press, Ames.
Tzipori, S. J. Hayes, L. Sims, and M. Withers. 1984. Streptococcw durarm An unexpected enteropathogen of foals. J. Infect. Diseases 150:589.
Widdowson, E. M. 1984. Milk and the newborn animal. Proc.
NU&. SOC. 43:87.
Widdowson, E. M. 1985. Development of the digestive system
Comparative animal studies. Am. J. Clin. Nutr. 41:384.