Psychoneuroendocrinology (2006) 31, 395–406
www.elsevier.com/locate/psyneuen
Effects of maternal immobilization stress on birth
weight and glucose homeostasis in the offspring
Anil P D’mello*, Ying Liu1
University of the Sciences in Philadelphia, 600 South 43rd Street, Philadelphia, PA 19104 USA
Received 14 June 2005; received in revised form 30 August 2005; accepted 21 October 2005
KEYWORDS
Prenatal stress;
Corticosterone;
Glucose tolerance;
Birth weight;
Fructose feeding;
Triglycerides
Abstract
Recent epidemiological studies have shown strong associations
between low birth weight and the incidence of diabetes in the adult offspring. It
has been hypothesized that exposure to maternal glucocorticoids programs cellular
changes in the fetus which increases the susceptibility of the offspring to diabetes.
Stressors produces large increases in maternal glucocorticoids. The present study
determined the effects of immobilization stress during weeks one, two or three of
pregnancy on offspring birth weight, glucose homeostasis, and the ability of the
offspring to cope with metabolic stress. Immobilization stress produced large
increases in maternal levels of ACTH and corticosterone, but did not affect birth
weight of the pups. Chronic administration of high fructose diet, a metabolic
stressor, to 60 days old control and prenatally stressed offspring produced large
increases in plasma levels of triglyceride and insulin. However, there were no
differences between the groups either in peak levels, or in the rates of increase and
decrease (upon discontinuation of the diet) of plasma triglyceride and insulin
concentrations. Basal levels of glucose and insulin, and areas under the glucose and
insulin plasma concentration—time curves after an i.p. glucose dose were similar
between 120 days old control and prenatally stressed offspring. These results suggest
that in young adult rats prenatal immobilization stress did not affect glucose
homeostasis or the ability of these rats to cope with chronic metabolic stress.
Q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
Non-insulin dependant diabetes (type 2) is a chronic
disease that exerts an enormous economic and
social cost in modern society (Seidell, 2000). It
* Corresponding author. Tel.: C1 215 596 8941; fax: C1 215
895 1161.
E-mail address: [email protected] (A.P. D’mello).
1
Present address: Giaxo Smith Kline, King of Prussia; PA 19406,
USA.
accounts for 90% of all cases of diabetes in the
United States (Skyler and Oddo, 2002). Research
examining the mechanism(s) of the etiology of
diabetes has focused on genetic issues and factors
like diet, lack of exercise, and obesity. Recent
epidemiological studies have shown strong associations between low birth weight and diabetes in
adulthood (Eriksson et al., 2003; Barker, 2003;
Ozanne et al., 2004; Hales and Ozanne, 2003). The
associations are independent of factors like smoking, alcohol consumption, obesity and social class,
0306-4530/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.psyneuen.2005.10.003
396
although these factors certainly add to the effects
of early life (Phipps et al., 1993). The authors of the
studies have made the provocative suggestion that
the intra-uterine environment is a major determinant of diabetes. They have hypothesized that
increased exposure of the fetus to glucocorticoids
programs long term cellular changes, which produces low birth weight and predisposes the
organism to diabetes in adult life.
A variety of synthetic glucocorticoids are administered to pregnant women in preterm labor to
promote lung maturation in the fetus and reduce
mortality in preterm infants. Some of these clinical
studies have demonstrated that in utero exposure to
glucocorticoids reduces birth weight in infants (Gur
et al., 2004; Murphy et al., 2002; Bloom et al., 2001).
Interestingly, recent longitudinal studies have
demonstrated that prenatal exposure to glucocorticoids is also associated with elevated blood pressure
and increased behavioral disturbances in adolescence (Doyle et al., 2000; Trautman et al., 1995).
However, to the best of our knowledge, clinical
studies have not systematically examined the
effects of prenatal glucocorticoid exposure on
glucose and insulin homeostasis in adolescence or
adulthood. In contrast, a number of animals studies
have shown that prenatal exposure to glucocorticoids decreases birth weight (Reininisch et al., 1978;
Ain et al., 2005; Jobe et al., 1998), and is associated
with insulin resistance, hyperglycemia (Lindsay
et al., 1996; Nyirenda et al., 1998), and hypertension
(Benediktsson et al., 1993; Gatford et al., 2000) in
the adult offspring. These changes occur long after
the last exposure to glucocorticoids, and the
mechanisms mediating these effects are not clearly
understood. Unfortunately, most of these studies
have used synthetic glucocorticoids like dexamethasone, betamethasone, and prednisone. The tissue
distribution, receptor affinities, metabolism routes,
and pharmacokinetics of these steroids are markedly different from the physiologically relevant
glucocorticoid (Reul et al., 1987; Diederich et al.,
1997). Consequently, their administration results in
non-physiological exposure of the fetus to glucocorticoids. Also, many of these studies have used
techniques, and doses that produce glucocorticoid
levels that are much higher than those encountered
physiologically.
One of the largest physiological increases in
glucocorticoids is produced by stressors. In humans,
low birth weight has been associated with elevated
placental levels of corticotropin releasing hormone
(a secretagogue of cortisol) and higher fetal cord
blood cortisol levels suggesting a role for endogenous glucocorticoids in producing fetal growth
retardation (Goland et al.,1993; Wadhwa et al.,
A.P. D’mello, Y. Liu
2004). However, to the best of our knowledge,
clinical studies have not examined the association
between maternal stress and glucose and insulin
homeostasis in the offspring. Numerous animals
studies have examined the effects of maternal
stress on different development end points in the
offspring. A variety of prenatal stress paradigms
decrease birth weight of rodents and non-human
primates (Pollard, 1984; Cabrera et al., 1999; Drago
et al., 1999; Schneider et al., 1999; Lesage et al.,
2004). However, and much to our surprise, there is
limited information on the role of maternal stress
and resultant glucocorticoid production in the
etiology of altered glucose metabolism and diabetes in the offspring. We are interested in
determining, in rats, if fetal exposure to glucocorticoids via prenatal stress will produce irreversible
alterations in glucose metabolism in the exposed
animals. The primary hypothesis guiding these
studies is that prenatal stress will increase the
cumulative exposure of the fetus to maternal
corticosterone (B), the primary glucocorticoid in
rats. This exposure will ‘program’ irreversible
cellular changes in the fetus, and lead to reduced
birth weight, and long term alterations in glucose
metabolism in the adult offspring.
The specific aim of these studies was to
determine the effects of immobilization stress
administered during the first, second, and third
week of pregnancy on birth weight and glucose
homeostasis in the offspring. Glucose homeostasis
was evaluated by measuring basal fasting plasma
glucose, and insulin levels and by the conduct of i.p.
glucose tolerance tests in the adult offspring. Also,
we evaluated the ability of prenatally stressed rats
to cope with chronic metabolic stress administered
during adulthood. Adult rats prenatally stressed
during weeks one, two, or three of gestation were
exposed to the chronic metabolic stress of a high
fructose diet. In normal rats, a high fructose diet
causes hyperglycemia, hyperinsulinemia, and
hypertriglyceridemia, and has been used as a
model for Syndrome X (Reed et al., 1994; Tobey
et al., 1982; Bhanot et al., 1994). In control and
prenatally stressed rats, we measured plasma
triglyceride, insulin, and glucose levels before,
during, and after discontinuation of a 3 week
regimen of high fructose diet.
2. Methods
2.1. Materials
D-(C) glucose was purchased from Sigma Co.
(St. Louis, MO), and bovine aprotinin was obtained
Prenatal stress and glucose homeostasis
from Bayer(Kankakee, IL). The fructose diet
containing 60% fructose, 21% protein, and 5.2%
fat was purchased from Harlan Teklad (Madison,
WI). The diet derived 66% of its calories from
carbohydrate (fructose), 20% from protein, and
13% from fat.
2.2. Animal experiments
All animal experiments were approved by the
University of the Sciences in Philadelphia Institutional Animal Care and Use Committee and were
conducted in accordance with the National Institutes of Health Guide for the Care and Use of
Laboratory Animals (NIH Publications no. 80–23,
revised 1978).
2.3. Experiments with pregnant rats
All animals were allowed to acclimatize to the
vivarium for 2 weeks prior to the start of the
experiments. Virgin female Sprague—Dawley rats
(225–250 g, Charles-River, MA) were mated by
housing one male rat with two female rats. Day
one of pregnancy was assigned upon observation of
sperm in the daily morning vaginal smears. Rats
were then singly housed throughout the pregnancy.
Pregnant rats were divided into four groups of 5–7
rats per group. Group 1 was subject to immobilization stress on days 1–7 of pregnancy while groups 2
and 3 received immobilization stress on days 8–14
and 15–21 of pregnancy, respectively. Group 4
consisted of control rats that were left undisturbed
throughout their pregnancy. Stress was administered in a room adjacent to the room in which the
rats were housed. Immobilization stress was administered in three, 45 min sessions in the morning,
afternoon, and evening. Stress was administered by
enclosing rats in flexible plastic cones (Braintree
Scientific Inc., Braintree, MA). The wide end of the
cone was securely taped such that it conformed to
the individual body size of the animals. 0.25 mL tail
tip blood samples were collected from mothers in
groups one, two, or three before (basal) and
immediately after the 45 min morning stress session
(stress) on the first and last day of the seven day
immobilization stress regimen. Basal samples taken
before the administration of stressor were obtained
within two minutes of moving the animal from its
home cage. Blood samples were collected in cold
EDTA and aprotinin lined polypropylene tubes,
centrifuged at 4 8C, and plasma harvested and
stored at K70 8C pending analyses for corticosterone and ACTH.
397
2.4. Experiments with offspring
Upon birth offspring from litters in all four groups
were weighed. Offspring were weaned on day 21
post-birth and housed in isosexual groups according
to prenatal treatment. Only male offspring from
each litter were used in all subsequent
experiments.
2.5. Fructose feeding experiment
On day 60, 1–2 male offspring were randomly chosen
from each litter in each of the four groups. A 0.3 mL
tail tip blood sample was taken from all animals
(basal). Rats were then transferred onto a fructose
diet for the next 21 days. Tail tip blood samples
(0.3 mL) were taken from rats in all groups on day 7,
14, and day 21 of the fructose diet treatment To
assess the rate of recovery from the effects of
chronic ingestion of fructose, the fructose diet was
discontinued on day 21, and rats in all groups were
transferred to a diet of regular lab chow. Three days
later, a final 0.3 mL tail tip blood sample was
obtained from all rats (day 24). On the days of blood
sampling, food was withdrawn at 8:00 am and blood
collected at 1:00 pm after a 5 h fast. Blood was
collected in cold EDTA and aprotinin lined polypropylene tubes, centrifuged at 4 8C, plasma
collected, and stored at K70 8C pending analyses
of triglyceride, insulin, and glucose levels.
To determine if alterations in biochemical
parameters were only due to the fructose diet,
and to account for possible (but unlikely) changes in
biochemical parameters due to the passage of time
(24 days), male offspring from litters in each of the
four groups (4–6 rats per group) were housed
alongside the fructose fed animals, and fed a lab
chow diet for 24 days. Tail tip blood samples
(0.3 mL) were obtained from these animals, processed and stored as described above, and used for
the measurement of triglyceride, insulin, and
glucose levels.
2.6. Glucose tolerance test
On day 120, two randomly chosen male offspring
from each litter in all four groups were administered an i.p. glucose tolerance test. Animals were
fasted overnight, and the following morning glucose
tolerance tests were started and completed
between 8:30 AM–11:30 AM. Animals were administered a 2 gm/kg i.p. glucose dose, and 0.3 mL tail
tip blood samples were taken in EDTA and aprotinin
lined polypropylene tubes just before the dose and
at 15, 30, and 60 min after the dose. Samples were
398
A.P. D’mello, Y. Liu
centrifuged and plasma harvested and stored at
K70 8C pending analyses for glucose and insulin.
2.7. Biochemical methods
Plasma corticosterone concentrations were determined using a RIA kit supplied by ICN Biomedicals,
Inc (Costa Mesa, CA). The inter-day coefficient of
variation of the assay was 7.1%. ACTH plasma
concentrations were determined using a RIA kit
supplied by DiaSorin, Inc (Stillwater, MN). The
inter-day coefficient of variation of the assay was
9.4% at a concentration of 20 pg/mL and 7.2% at a
concentration of 500 pg/mL.
Plasma triglyceride concentrations were
measured using the GPO—Trinder reagents kit supplied by Sigma Diagnostics (St. Louis, MO). The interday coefficient of variation of the assay was 3.5%.
Plasma insulin concentrations were measured by
RIA using a kit supplied by LINCO Research (St.
Charles, Missouri). The kit uses an antibody made
specifically against rat insulin. The inter-day
coefficient of variation of the assay was 8.9% at a
concentration of 4.5 ng/mL and 9.2% at a concentration of 0.65 ng/mL.
Plasma glucose concentrations were measured
by the glucose oxidase method using the glucose
(Trinder) kit from Sigma Diagnostics (St. Louis, MO).
The inter-day coefficient of variation of the assay
was 4.1%
2.8. Data analyses
All data are expressed as meanGSEM In order to
compute mean birth weight of a group, all pups in a
litter from the group were weighed and the average
pup weight of the litter computed. This was
repeated for all litters in the group, and the mean
of these values is reported as the mean birth weight
of the group. Similarly, whenever two rats from a
Table 1
litter were used, the values of biochemical
parameters obtained from the two rats were
averaged and reported as a single point rather
than treat each offspring as a separate sample.
Consequently, in experiments with the offspring the
sample size in each of the four groups never
exceeded the number of treated dams in each
group. Such a conservative method of data analyses
is recommended for the analyses of offspring data
from multiparous species as it minimizes litter
effects (Holson and Pearce, 1992; Zorilla, 1997).
In the fructose experiment, biochemical parameters were repeatedly measured from the same
set of rats over the 24 day course of the
experiment. Therefore, plasma triglyceride, insulin, and glucose data were analyzed with a repeated
measure 2-way ANOVA followed, where appropriate, by a Newman-Keuls post—hoc multiple comparison test.
In the glucose tolerance test experiment, the
total area under the plasma concentration—time
curves for glucose and insulin were determined
using the linear trapezoidal rule. The basal areas
were then subtracted, and the net area under the
plasma concentration—time curves reported. Basal
glucose and insulin levels (prior to the i.p. glucose
injection), and net area under the plasma concentration—time curve data between groups were
compared with a one way ANOVA followed, where
appropriate, by a Newman-Keuls post—hoc multiple
comparisons test.
All statistical tests were conducted at a 0.05
level of significance.
3. Results
As shown in Table 1, immobilization stress during
weeks 1, 2, or 3 of pregnancy had no effect on the
gestation length, litter size, or birth weight of the
Effect of immobilization stress during weeks 1, 2, or 3 of pregnancy on reproductive performance
Parameter (units)
Gestation length (days)
Percent maternal weight gain (%)c
Litter size
Birth weight (gms)d
Treatmenta
Control
22.6G0.2
64.5G2
14.1G1
6.4G0.2
b
Week 1
Week 2
Week 3
22.3G0.2
56.2G2*
13.7G0.6
6.3G0.2
22.2G0.2
60.6G3
14.8G0.7
5.9G0.2
22.3G0.2
54.8G2*
14.0G1
6.0G0.2
a
Immobilization stress administered during weeks 1, 2, or 3 of pregnancy or animals left undisturbed throughout the pregnancy
(control).
b
MeanGSEM nZ6–7.
c
Percent weight gain relative to the weight on day 1 of pregnancy. Day 1 weight in all groups ranged from 231–278 g.
d
All pups in a litter were weighed and the average pup weight of a litter computed and reported as a single point. This was done for
all litters in a group and the mean of these points is reported as the group mean in the table.
* Significantly different from control, p!0.05, one way ANOVA with Newman-Keuls post hoc test.
Prenatal stress and glucose homeostasis
Table 2
399
Basal and immobilization stress induced ACTH levels in pregnant rats
Pregnancy week
during which stress
was administered
Plasma ACTH concentration (pg/mL)
Day 1
Day 7
Basal
Stress
Basal
Stress
Week 1
Week 2
Week 3
46G4a
45G5
43G4
352G78CCC
375G52CCC
436G52CCC
96G5**
112G15***
112G10***
301G50CCC
246G44CC
196G16CCC
P!0.01, CCCp!0.001 significantly different from basal concentration on the same day; 1-way ANOVA, Newman-Keuls post-hoc
test. **p!0.01, ***p!0.001 significantly different from day 1 basal concentration; 1-way ANOVA, Newman-Keuls post-hoc test.
Pregnant rats were subject to 7 days of an immobilization stress regimen during weeks 1, 2, or 3 of pregnancy. Plasma ACTH levels
were measured before (basal) and immediately after (stress) the morning stress session on day 1 and 7 of the stress regimen.
a
MeanGSEM; nZ6–7.
CC
pups. However, stress during weeks 1 and 3 of
pregnancy reduced the magnitude of pregnancy
induced increase in maternal body weight.
Table 2 shows that immobilization stress administered during weeks 1, 2, or 3 of pregnancy
produced large increases in the plasma concentrations of ACTH. In all three groups, the increase in
stress induced plasma ACTH levels were observed
both on the first and last day of the stress regimen.
This suggest that there was no acclimatization to
the chronic (7 days) stress regimen. Basal ACTH
levels were increased over the course of the 7 days
immobilization stress regimen probably reflecting a
chronic stress induced increase in basal hypothalamus—pituitary activity.
Table 3
Basal and immobilization stress induced corticosterone levels in pregnant rats
Pregnancy week
during which stress
was administered
Week 1
Week 2
Week 3
Plasma B levels showed a pattern of changes
similar to that of its secretagogue, ACTH. Stress
produced marked increases in plasma B levels both
on the first and last day of the chronic stress
regimen (Table 3). However, in contrast to the
ACTH data, there were no increases in basal B levels
over the course of the seven day immobilization
stress regimen, except in animals stressed during
week three of pregnancy. In this group, basal B
levels on day seven were significantly higher than
the corresponding B levels on day one.
Table 4 shows the body weight profile of
control male offspring, and male offspring prenatally stressed during weeks one, two, or three
of gestation. No statistically significant
Plasma corticosterone concentration (ng/mL)
Day 1
Day 7
Basal
44G15
31G12
17G2
Stress
a
670G42
634G26CCC
616G24CCC
CCC
Basal
Stress
12G5 ***
18G6
76G8 ***
624G68CCC
606G43CCC
566G65CCC
CCC
p!0.001, significantly different from basal concentration on the same day; 1-way ANOVA, Newman-Keuls post-hoc test. *** p!
0.001, significantly different from day 1 basal concentration; 1-way ANOVA, Newman-Keuls post-hoc test. Pregnant rats were
subject to seven days of an immobilization stress regimen during weeks 1, 2, or 3 of pregnancy. Plasma corticosterone levels were
measured before (basal) and immediately after (stress) the morning stress session on day 1 and 7 of the stress regimen.
a
MeanGSEM; nZ6–7.
Table 4 Body weight profile of male control rats and male rats prenatally stressed during weeks 1,2, or 3 of
gestation
Treatment
Control
Week 1
Week 2
Week 3
a
b
Body Weight (gms)
Birtha
b
6.4G0.2
6.3G0.2
5.9G0.2
6.0G0.2
Includes both male and female offspring.
MeanGSEM; nZ5–7.
21 days
60 days
120 days
38.9G2.0
40.4G1.6
37.5G1.8
39.9G1.5
365G17
351G7
351G4
360G7
560G12
540G24
547G15
567G19
400
A.P. D’mello, Y. Liu
Figure 1 Plasma triglyceride (A), insulin (B), and glucose (C) concentrations in control and prenatally stressed rats
before, during, and after discontinuation of a 21 days high fructose diet regimen. Rats prenatally stressed during weeks 1
(WK 1), 2 (WK 2) or 3 (WK 3) of gestation or left undisturbed throughout gestation (C) were administered a 60% fructose
diet for 21 days. Blood was sampled from rats in all groups before starting the fructose diet (BASAL), on days seven (DAY
7), 14 (DAY 14), and 21 (DAY 21) after starting the diet, and three days after discontinuation of the diet (DAY 24). See the
Methods section for more details on the experimental protocol. Bars represent meanGSEM of 5–7 rats per group.
differences were observed between the groups in
either birth weight or in body weight at any of the
three ages examined. As shown in Fig. 1(A),
administration of a 60% fructose diet caused
marked increases in plasma triglyceride concentrations in control and all three groups of
prenatally stressed offspring. These results show
that we were able to reproduce literature reports
which demonstrate high fructose diet induced
increases in plasma triglyceride concentrations.
The increase persists throughout the 21 day period
of fructose diet administration (2-way repeated
Prenatal stress and glucose homeostasis
measures ANOVA, significant effect of duration,
P!0.001). Three days after discontinuation of the
diet, triglyceride levels returned to their basal
values (day 24Zbasal, Newman—Keuls post-hoc
test). However, there were no statistically significant differences in plasma triglyceride concentrations between the four groups at any of the
time periods. Similarly, there were no significant
differences between the groups either in the rates
of increase of triglyceride concentrations or in
their rates of recovery subsequent to stopping the
fructose diet on day 21 (2-way repeated measures
ANOVA, main effect of prenatal stress, PZ0.18;
prenatal stress!duration interaction, PZ0.425).
As previously reported in the literature, our
results (Fig. 1(B)) demonstrate that the 60%
fructose diet increased plasma insulin concentrations by day 7 which persisted throughout the
21 day diet regimen (2-way repeated measures
ANOVA, significant duration effect, P!0.001).
Three days after discontinuation of the diet, insulin
levels decreased but remained higher than the basal
levels (day 24Obasal, Newman—Keuls post-hoc
test). As previously reported for triglyceride levels,
there were no differences in plasma insulin levels
between the four groups in the rats, or in the rates
of increase or decrease of insulin levels (2-way
repeated measures ANOVA, main effect of prenatal
stress, PZ0.628, prenatal stress!duration interaction, PZ0.067).
Fig. 1(C) shows the effect of chronic fructose
diet administration on plasma glucose levels.
Surprisingly, a 2-way repeated measure ANOVA
revealed a significant duration effect (P!0.001).
More detailed 1—way repeated measure ANOVA
revealed that statistical significance was achieved
due to small (7%) decrease in glucose levels
between discrete periods in the WK 2 and WK 3
groups. Results from the literature suggest that
chronic administration of a high fructose diet does
not affect plasma glucose levels. In the present
study, the relevance of the small changes in
401
glucose levels in two of the four groups is unclear.
There were no statistically significant differences
between the four groups in glucose levels (2-way
repeated measure ANOVA, main effect of prenatal
stress, PZ0.681, prenatal stress!duration interaction, PZ0.671).
Additional animals from the four groups were
housed alongside the fructose fed animals, and fed
lab chow for the entire 24 day duration of the
experiment. These rats served as negative controls
for the above mentioned experiment and plasma
triglyceride, insulin, and glucose levels were
measured in these animals at the same sampling
times. Analyses of the triglyceride and insulin data
demonstrated no duration effect (2-way repeated
measure ANOVA), which suggest that changes in
these biochemical parameters observed in the
fructose fed animals were due to the diet and not
the passage of time (data not shown). Surprisingly,
plasma glucose levels in these negative controls
showed a small but significant decrease over the
24 day period (data not shown). The very small
magnitude of these changes is of questionable
relevance.
Analyses of body weights during the fructose
diet experiment demonstrated no statistical
differences in body weights between the control
and prenatally stressed rats either prior to,
during, or three days after discontinuation of a
high fructose diet (Table 5). Also, comparison of
body weights during the five measurement periods
between control offspring fed a high fructose diet,
and control offspring fed lab chow (negative
controls) during the same period revealed no
significant differences in body weight (data not
shown). This confirms literature reports that the
high fructose diet mediated increases in plasma
triglyceride and insulin concentrations are not
accompanied by disproportionate increases in
body weight.
Fig. 2 shows plasma glucose and insulin levels in
all four groups during an i.p. glucose tolerance test
Table 5 Body weight profile of control and prenatally stressed rats before, during, and after a 21 day regimen of
high fructose diet
Treatment
Control
Week 1
Week 2
Week 3
Body weight (gms)
Basal
Day 7
a
365G18
351G7
351G4
360G7
380G19
367G9
366G4
372G7
Day 14
Day 21
Day 24
411G11
399G10
394G4
404G7
437C21
422G9
418G4
432G9
460G23
444G8
440G3
453G10
Adult control rats or rats stressed during weeks 1,2,or 3 of gestation were administered a high fructose diet. Body weight was
measured before (basal), during administration of the diet (days 7, 14, and 21), and 3 days after stopping the fructose diet (day 24).
a
MeanGSEM; nZ5–7.
402
A.P. D’mello, Y. Liu
during weeks 1, 2, or 3 of gestation did not affect
basal glucose or insulin plasma concentrations.
Administration of a 2 gm/kg i.p. glucose dose
caused increases in glucose and insulin levels. In
the control group, the net area under the glucose
and insulin plasma concentration time curves were
10,071 (mg/dL)!min and 35.9 (ng/mL)!min,
respectively. Prenatal stress during weeks one,
two, or three of gestation did not affect the net
area under the plasma glucose and insulin concentration—time curves.
4. Discussion
Figure 2 Plasma glucose (A) and insulin (B) levels
during a glucose tolerance test in control rats and rats
prenatally stressed during week 1, week 2, or week 3 of
gestation. Concentrations at each time point represent
the meanGSEM of 6–7 values per group. Each value is the
average of two pups in a litter.
while Table 6 summarizes the results of the i.p.
glucose tolerance test. Basal plasma glucose and
insulin levels in the control group were 103 mg/dL
and 0.39 ng/mL, respectively. Prenatal stress
Our studies were designed to examine if maternal
stress during discrete periods of gestation and the
consequent exposure of the developing fetuses to
B, would affect glucose and insulin homeostasis in
the adult offspring. In our study, basal B levels on
day one in all three groups were similar (Table 3).
These basal levels were obtained between the first
(in group week 1) and 15 (in group week 3) day of
pregnancy. The results support reports which
demonstrate that basal B levels during pregnancy
are not significantly affected until day 14 of
pregnancy and begin to rise about day 18 of
pregnancy (Waddell and Atkinson, 1994; Atkinson
and Waddell, 1995). In the group that received
stress during the last week of pregnancy (week 3),
basal B levels on day seven were higher than the
corresponding levels on day one. This is probably
not a stress related phenomenon, and reflects the
dam’s preparation for impending parturition (Waddell and Atkinson, 1994). The immobilization stress
paradigm used in our studies caused massive
increases in plasma levels of B and in the levels of
Table 6 Basal plasma and insulin levels and net area under the glucose and insulin plasma concentration—time
curves during an i.p. glucose tolerance test
Parameter (units)
Basal plasma glucose (mg/dL)
Net glucose AUC
(mg/dL)!min
Basal plasma insulin
(ng/mL)
Net insulin AUC (ng/
mL)!min
Treatment
Control
Week 1
Week 2
Week 3
103G3a
105G4
10071G688
9876G490
7814G1432
10037G1274
0.39G0.1
0.67G0.2
0.58G0.1
0.49G0.1
35.9G10
35.3G7
30.6G7
29.9G7
107G4
104G4
One hundred and twenty day old male control rats or rats subject to prenatal immobilization stress during weeks 1, 2 or 3 of
gestation were administered an i.p. glucose tolerance test as described in the Methods section. Blood samples were taken before
(basal) and at varying times after the glucose dose, and net area under the plasma glucose and insulin concentration—time curves
(AUC) were computed as described in the Methods section.
a
MeanGSEM; nZ6–7.
Prenatal stress and glucose homeostasis
its secretagogue ACTH. Despite the chronic seven
day stress regimen, stress induced increases in B
and ACTH levels were similar on day one and seven
of this regimen suggesting no biochemical acclimatization to the stressor. Some studies have shown
that a chronic stress regimen induces activation of
the basal activity of the HPA axis and increases in
basal levels of B (Vernikos et al., 1982). However, in
the present study, such increases were observed in
the basal levels of ACTH but not in the levels of B
probably due to the episodic nature of the chronic
stress regimen.
Maternal stress during weeks 1, 2, or 3 of
pregnancy did not adversely affect reproductive
performance as assessed by gestation length and
litter size. However, immobilization stress during
weeks one and three reduced maternal weight gain
during pregnancy; further evidence of the efficacy
of the stress regimen employed in these studies.
Prenatal immobilization stress did not affect birth
weight of the offspring. A review of the literature
shows inconsistency in the effects of prenatal stress
on birth weight with reports of both decreases
(Pollard, 1984; Cabrera et al., 1999; Drago et al.,
1999; Patin et al., 2002; Lesage et al., 2004) and no
change (Von Hoersten et al., 1995; Guo et al., 1993;
Holson et al., 1995; Gerardin et al., 2005). In most
of these studies, the weight of each individual pup
in all litters of a group was treated as an
independent observation in the statistical analyses.
Such an inflation of sample size can produce
spurious statistical significance and might account
for the reported inconsistency in the effects of
prenatal stress on birth weight. In our studies, the
average pup weight of each litter in a group was
computed and the values used to compute the
group mean. Consequently the sample size for birth
weight analyses in each group was equal to the
number of dams in the group. It must be noted that
if we had incorrectly treated individual pups in a
litter as independent samples we would have a
much larger sample size, and would have obtained
statistical significance for the effects of prenatal
stress on birth weight.
Our next study determined if prenatal stress
affected the ability of the adult offspring to cope
with metabolic stress. A high fructose diet is a well
established paradigm of a metabolic stressor.
Chronic administration of a high fructose diet is
known to increase triglyceride and insulin plasma
concentrations along with concomitant loss of
glycemic control. Therefore, it has often been
used as an animal model of the metabolic syndrome. The exact molecular mechanisms mediating
the effects of a high fructose diet are presently
unclear. Results of recent studies suggest that a
403
high fructose diet alters protein levels involved in
stress pathways, and acts by inducing hepatic stress
(Kelley et al., 2004).
Consistent with literature reports, our studies
demonstrate that a high fructose diet caused
marked increases in plasma levels of triglyceride
and insulin within 1 week of the start of diet
administration (Reed et al., 1994; Tobey et al.,
1982; Bhanot et al., 1994). Continued diet administration did not increase the magnitude of changes
in these parameters. Three days after discontinuation of the high fructose diet, triglyceride levels
decreased to their basal values and insulin levels
markedly decreased but continued to be slightly
higher than basal values. These results corroborate
literature reports, and demonstrate rapid recovery
of biochemical parameters from the effects of
chronic high fructose diet (Dai and McNeill, 1995).
Furthermore, as reported in the literature,
measurement of body weight during the experiment
demonstrated that the fructose diet mediated
increases in plasma insulin and triglyceride concentrations were not accompanied by a disproportionate increase in body weight (Reed et al., 1994;
Bhanot et al., 1994; Dai and McNeill, 1995).
Prenatal stress during weeks, one, two, or three
of gestation did not affect the rate of increase,
peak levels, or the rate of recovery of plasma
triglyceride and insulin concentrations. These
results suggest that prenatal stress did not affect
the ability of 60 day old offspring to cope with the
hepatic stress induced by administration of a
chronic high fructose diet.
Results from our final studies demonstrated that
prenatal stress did not affect basal glucose or
insulin levels in 120 day old rats and did not affect
glucose tolerance as measure by glucose and insulin
area under the plasma concentration—time curves
after an i.p. glucose load. In contrast to these
results, subcutaneous administration of dexamethasone, a synthetic glucocorticoid, to dams
during the third week of gestation, decreased
offspring birth weight, and produced glucose
intolerance in the 6 month old offspring (Nyirenda
et al., 1998). The fetus is normally protected from
exposure to the much higher levels of maternal
glucocorticoid by the placental enzyme 11
b-hydroxysteroid dehydrogenase, type 2 (11bHSD2).This enzyme converts physiologically active
glucocorticoid (corticosterone in rats) to the
inactive 11-keto products (11-dehydrocorticosterone in rats) (Monder and White, 1993; Heller et al.,
1988; Burton and Waddell, 1994). Dexamethasone is
a poor substrate for 11b-HSD2 (Siebe et al., 1993).
Consequently, its administration to pregnant dams
probably results in high, non-physiological exposure
404
of the fetus to glucocorticoids, and could account
for its ability to decrease birth weight and alter
glucose tolerance in the offspring. Few studies have
systematically explored fetal exposure to B after
maternal stress. One of the few studies that
measured maternal and fetal B levels after immobilization stress showed surprisingly small increases
in fetal B plasma levels despite huge increases in
maternal B—a testimony to the protective efficacy
of placental 11 b-HSD 2 (Ward and Weisz, 1984).
Maternal stress probably results in a much more
modest exposure of the fetus to maternal B which
might account for its inability to affect glucose
tolerance in the offspring.
Prenatal restraint stress in an illuminated
environment has been demonstrated to cause
modest (18%) increases in basal glucose levels in
five months old male rats (Vallee et al., 1996).
However, the number of pups examined in this
study was greater than the number of treated dams
and the results might be partially confounded by
litter effects. Interestingly, a recent study using an
experimental design similar to our study subjected
pregnant rats to restraint stress in an illuminated
environment during the last week of gestation. The
authors demonstrated prenatal stress mediated
reduction in fetal weight at term (Lesage et al.,
2004). Additionally, the authors showed that 24
months old prenatally stressed rats had elevated
basal glucose levels and higher glucose levels in
response to a oral glucose load as compared to
offspring of control rats. However, consistent with
our results, these authors did not observe differences between the control and prenatally stressed
groups either in basal insulin levels or in insulin
levels in response to the glucose load (Lesage et al.,
2004). The originators of the ‘restraint stress in an
illuminated environment’ paradigm created the
illuminated environment with the use of high
wattage bulbs with the accompanying generation
of significant amounts of heat (Ward and Weisz,
1984). Such a restraint paradigm is different from
the immobilization paradigm used in our studies
which is conducted at ambient temperature. The
possible use of such a paradigm by the aforementioned investigators could account for the differences in findings between the studies. Also, it is
conceivable that the relatively small and selective
effects of prenatal stress on glucose levels are
manifested only in aged animals. Due to experimental constraints, we evaluated glucose tolerance in four months old animals, in contrast to the
24 months old animals used by the earlier investigators. This could also account for our inability to
observe prenatal stress induced alterations in
glucose tolerance. Age specific effects of prenatal
A.P. D’mello, Y. Liu
insults on glucose tolerance have been previously
reported. A review of the literature showed that
maternal low protein diets, a widely used paradigm
of prenatal insult, which is hypothesized to act by
increasing fetal exposure to glucocorticoids actually improved glucose tolerance in younger animals
(Langley SC et al., 1994; Shepherd et al., 1997), and
deterioration of glucose tolerance was only
observed in aged offspring (Hales et al., 1996;
Petry et al., 2001).
Finally, a review of the literature demonstrates
that many studies which examine the effects of
maternal treatments on development in the offspring of multiparous species like rats are confounded by litter effects. The investigators treat a
relatively small number of dams, then proceed to
use multiple offspring from a litter in their
experiments, and treat each offspring in a litter as
an independent observation. This dramatically
inflates the sample size of the study, which in
turn inflates the alpha level (normally set at 0.05),
and leads to spurious statistical significance (Holson
and Pearce, 1992; Zorilla, 1997). In all our
experiments, whenever more than one pup in a
litter was used to measure a biological parameter,
the values obtained were averaged and reported as
a single point. Consequently in the analyses of
offspring data, the sample size of the group never
exceeded the number of treated dams in that
group. Such a conservative method of data reporting and statistical analyses is more accurate and is
highly recommended in the evaluation of maternal
treatments on the offspring of multiparous species
(Holson and Pearce, 1992; Zorilla, 1997). In our
studies, it minimized the risk of obtaining spurious
statistical significance, and consequently improved
the reliability of our conclusions.
Acknowledgements
The work was supported by National Institutes of
Health Grant HD 36786 (APD). We thank Rajeev
Shah and Pamela Mayfield for help in the conduct of
the fructose feeding experiments, and analyses of
the plasma triglyceride samples.
References
Ain, R., Canham, L.N., Soares, M.J., 2005. Dexamethasoneinduced intrauterine growth restriction impacts the placental
prolactin family, insulin-like growth factor-II and the Akt
signaling pathway. J. Endocrinol. 185, 253–263.
Prenatal stress and glucose homeostasis
Atkinson, H.C., Waddell, B.J., 1995. The hypothalamicÿpituitaryÿ
adrenal axis in rat pregnancy and lactation: circadian variation
and interrelationship of plasma adrenocorticotropin and
corticosterone. Endocrinology 136, 512–520.
Barker, D.J., 2003. The developmental origins of adult disease.
Eur. J. Epidemiol. 18, 733–736.
Benediktsson, R., Lindsay, R.S., Noble, J., Seckl, J.R., Edwards,
C.R.W., 1993. Glucocorticoid exposure in utero: new model
for adult hypertension. Lancet 341, 339–341.
Bhanot, S., McNeill, J.H., Breyer-Ash, M., 1994. Vanadyl sulfate
prevents fructose induced hyperinsulinemia and hypertension
in rats. Hypertension 23, 308–312.
Bloom, S.L., Sheffield, J.S., McIntire, D.D., Leveno, K.J., 2001.
Antenatal dexamethasone and decreased birth weight.
Obstet. Gynecol. 97, 485–490.
Burton, P.J., Waddell, B.J., 1994. 11b-hydroxysteroid dehydrogenase in the rate placenta: development changes and the
effects of altered glucocorticoid exposure. J. Endocrinol.
143, 505–513.
Cabrera, R.J., Rodriguez-Echandia, E.L., Jatuff, A.S., Foscolo,
M., 1999. Effects of prenatal exposure to a mild chronic
variable stress on body weight, preweaning mortality and rat
behavior. Braz. J. Med. Biol. Res. 32, 1229–1237.
Dai, S., McNeill, J.H., 1995. Fructose induced hypertension in
rats is concentration and duration dependent. J. Pharmacol.
Toxicol. Methods 33, 101–107.
Diederich, S., Hanke, B., Oelkers, W., Bahr, V., 1997. Metabolism
of dexamethasone in the human kidney: nicotinamide
adenine dinucleotide-dependent 11beta-reduction. J. Clin.
Endocrinol. Metab. 82, 1598–1602.
Doyle, L.W., Ford, G.W., Davis, N.M., Callanan, C., 2000.
Antenatal corticosteroid therapy and blood pressure at 14
years of age in preterm children. Clin. Sci. 98, 137–142.
Drago, F., Di Leo, F., Giardina, L., 1999. Prenatal stress induces
body weight deficit and behavioural alterations in rats: the
effect of diazepam. Eur. Neuropsychopharmacol. 9, 239–245.
Erickson, J.G., Fjorsen, T.J., Osmond, C., Barker, D.J., 2003.
Pathways of infant and childhood growth that lead to type 2
diabetes. Diabetes Care 26, 3006–3010.
Gatford, K.L., Wintour, E.M., De Blasio, M.J., Owens, J.A.,
Dodic, M., 2000. Differential timing for programming of
glucose homoeostasis, sensitivity to insulin and blood
pressure by in utero exposure to dexamethasone in sheep.
Clin. Sci. 98, 553–560.
Gerardin, D.C., Pereira, O.C., Kempinas, W.G., Florio, J.C.,
Moreira, E.G., Bernardi, M.M., 2005. Sexual behavior,
neuroendocrine, and neurochemical aspects in male rats
exposed prenatally to stress. Physiol. Behav. 84, 97–104.
Goland, R.S., Jozak, S., Warren, W.B., Conwell, I.M., Stark, R.I.,
Tropper, P.J., 1993. Elevated levels of umbilical cord plasma
corticotropin-releasing hormone in growth-retarded fetuses.
J. Clin. Endocrinol. Metab. 77, 1174–1179.
Guo, A., Nappi, R.E., Criscuolo, M., Ficarra, G., Amram, A.,
Trentini, G.P., Petraglia, F., Genazzani, A.R., 1993. Effect
of chronic intermittent stress on rat pregnancy and
postnatal development. Eur. J. Obstet. Gynecol. Reprod.
Biol. 51, 41–45.
Gur, C., Diav-Citrin, O., Shechtman, S., Arnon, J., Ornoy, A.,
2004. Pregnancy outcome after first trimester exposure to
corticosteroids: a prospective controlled study. Reprod.
Toxicol. 18, 93–101.
Hales, C.N., Ozanne, S.E., 2003. For debate: fetal and early
postnatal growth restriction leads to diabetes, the
metabolic syndrome and renal failure. Diabetologia 46,
1013–1019.
405
Hales, C.N., Desai, M., Ozanne, S.E., Crowther, N.J., 1996.
Fishing in the stream of diabetes: from measuring insulin to
the control of fetal organogenesis. Biochem. Soc. Trans. 24,
341–350.
Heller, C.L., Weisenberg, L.S., Orti, E., DeNicola, A.F., 1988.
Steps in the glucocorticoid action in normal and diabetic rat
placenta. J. Steroid Biochem. 31, 119–123.
Holson, R.R., Pearce, B., 1992. Principals and pitfalls in the
analysis of prenatal treatment effects in multiparous species.
Neurotoxicol. Teratol. 14, 221–228.
Holson, R.R., Ghough, B., Sullivan, P., Badger, T., Sheehan,
D.M., 1995. Prenatal dexamethasone or stress but not ACTH
or corticosterone alter several behavior in male rats.
Neurotoxicol. Teratol. 17, 393–401.
Jobe, A.H., Wada, N., Berry, L.M., Ikegami, M., Ervin, M.G.,
1998. Single and repetitive maternal glucocorticoid
exposures reduce fetal growth in sheep. Am. J. Obstet.
Gynecol. 178, 880–885.
Kelley, G.L., Allan, G., Azhar, S., 2004. High dietary fructose
induces a hepatic stress response resulting in cholesterol and
lipid dysregulation. Endrocrinology 145, 548–555.
Langley, S.C., Browne, R.F., Jackson, A.A., 1994. Altered
glucose tolerance in rats exposed to maternal low protein
diets in utero. Comp. Biochem. Physiol. 109A, 223–229.
Lesage, J., Del-Favero, F., Leeonhardt, M., Louvart, H., Maccari,
S., Vieau, D., Darnaudery, M., 2004. Prenatal stress induces
intrauterine growth restriction and programmes glucose
intolerance and feeding behavior disturbances in the aged
rat. J. Endocrinol. 181, 291–296.
Lindsay, R.S., Lindsay, R.M., Waddell, B.J., Seckl, J.R., 1996.
Prenatal glucocorticoid exposure leads to offspring hyperglycaemia in the rat: studies with the 11 beta-hydroxysteroid
dehydrogenase inhibitor carbenoxolone. Diabetologia 39,
1299–1305.
Monder, C., White, P., 1993. 11b-hydroxysteroid dehydrogenase.
Vitam. Horm. 47, 187–271.
Murphy, D.J., Caukwell, S., Joels, L.A., Wardle, P., 2002. Cohort
study of the neonatal outcome of twin pregnancies that were
treated with prophylactic or rescue antenatal corticosteroids. Am. J. Obstet. Gynecol. 187, 483–488.
Nyirenda, M.J., Lindsay, R.S., Kenyon, C.J., Burchell, A., Seckl,
J.R., 1998. Glucocorticoid exposure in late gestation permanently programs rat hepatic phosphoenolpyruate carboxykinase
and glucocorticoid receptor expression and causes glucose
intolerance in adult offspring. J. Clin. Invest. 101, 2174–2181.
Ozanne, S.E., Fernandez-Twinn, D., Hales, C.N., 2004. Fetal
growth and adult diseases. Semin. Perinatol. 28, 81–87.
Patin, V., Lordi, B., Vincent, A., Thomas, J.L., Vandry, H.,
Caston, J., 2002. Effect of prenatal stress on maternal
behavior in the rat. Dev. Brain Res. 139, 1–8.
Petry, C.J., Dorling, M.W., Pawlak, D.B., Ozanne, S.E., Hales,
C.N., 2001. Diabetes in old male offspring of rat dams fed
a reduced protein diet. Int. J. Exp. Diabetes Res. 2, 139–
143.
Phipps, K., Barker, D.J.P., Hales, C.N., Fall, C.H.D., Osmond, C.,
Clark, P.M.S., 1993. Fetal growth and impaired glucose
tolerance in men and women. Diabetologia 36, 225–228.
Pollard, I., 1984. Effects of stress administered during pregnancy
on reproductive capacity and subsequent development of the
offspring of rats: prolonged effects on the litters of a second
pregnancy. J. Endocrinol. 100, 301–306.
Reed, M.J., Ho, H., Donnelly, R., Reaven, G.M., 1994. Saltsensitive and carbohydrate-sensitive rodent hypertension:
evidence of strain differences. Blood Press. 3, 197–201.
Reininisch, J.M., Simon, N.G., Karow, W.G., Gandelman, R.,
1978. Prenatal exposure to prednisone in humans and animals
retards intrauterine growth. Science 202, 436–438.
406
Reul, J.M., Van der Bosch, F.R., de Kloet, E.R., 1987. Relative
occupation of type-I and type-II corticosteroid receptors in
rat brain following stress and dexamethasone treatment:
functional implications. J. Endocrinol. 115, 459–467.
Schneider, M.L., Roughton, E.C., Koehler, A.J., Lubach, G.R.,
1999. Growth and development following prenatal stress
exposure in primates: an examination of ontogenetic
vulnerability. Child Dev. 70, 263–274.
Seidell, J.C., 2000. Obesity, insulin resistance and diabetes—a
worldwide epidemic. Br. J. Nutr. 55, 58.
Shepherd, P.R., Crowther, N.J., Desai, M., Hales, C.N., Ozanne,
S., 1997. Altered adipocyte properties in the offspring of
protein malnourished rats. Br. J. Nutr. 78, 121–129.
Siebe, H., Baude, G., Lichtenstein, I., Wang, D., Buhler, H.,
Hoyer, G.A., Hierholzer, K., 1993. Metabolism of dexamethasone: sites and activity in mammalian tissues. Renal. Physiol.
Biochem. 16, 79–88.
Skyler, J.S., Oddo, C., 2002. Diabetes trends in the USA. Diabetes
Metab. Res. Rev. 18, 521–526.
Tobey, T.A., Mondon, C.E., Zavaroni, I., Reaven, G.M., 1982.
Mechanism of insulin resistance in fructose-fed rats. Metabolism 31, 608–612.
Trautman, P.D., Meyer-Bahlburg, H.F., Postelnek, J., New, M.I.,
1995. Effects of early prenatal dexamethasone on the cognitive
and behavioral development of young children: results of a pilot
study. Psychoneuroendocrinology 20, 439–449.
A.P. D’mello, Y. Liu
Vallee, M., Mayo, W., Maccari, S., Le Moal, M., Simon, H., 1996.
Long-term effects of prenatal stress and handling on
metabolic parameters: relationship to corticosterone
secretion response. Brain Res. 712, 287–292.
Vernikos, J., Dallman, M.F., Bonner, C., Katzen, A., Shinsako, J.,
1982. Pituitary adrenal function in rats chronically exposed to
cold. Endocrinology 110, 413–420.
Von Hoersten, S., Dimitoijevic, M., Markovic, B.M., Jankovic,
B.D., 1995. Effort of early experience on behavior and
immune response in the rat. Physiol. Behav. 54, 931–940.
Waddell, B.J., Atkinson, H.A., 1994. Production rate, metabolic
clearance rate, and uterine extraction of corticosterone
during rat pregnancy. J. Endocrinol. 143, 183–190.
Wadhwa, P.D., Garite, T.J., Porto, M., Glynn, L., Chicz-DeMet,
A., Dunkel-Schetter, C., Sandman, C.A., 2004. Placental
corticotropin-releasing hormone (CRH), spontaneous preterm
birth, and fetal growth restriction: a prospective investigation. Am. J. Obstet. Gynecol. 191, 1063–1069.
Ward, I.L., Weisz, J., 1984. Differential effects of maternal
stress on circulating levels of corticosterone, progesterone,
and testosterone, in male and female rat fetuses and their
mother. Endocrinology 1635, 1644.
Zorrilla, E.P., 1997. Multiparous species present problems
(and possibilities) to developmentalists. Dev. Psychobiol.
30, 141–150.