| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Pediatric, Division of Endocrinology (J.N.R., A.D.R., E.F.R.), the Departments of Biology (X.L., E.F.R.) and Pharmacology (A.D.R.), and the Center for Biological Timing (A.D.R., E.F.R.), University of Virginia, Charlottesville, Virginia 22903
Address all correspondence and requests for reprints to: James N. Roemmich, Ph.D., Department of Pediatrics, Division of Endocrinology, University of Virginia Health Sciences Center, Box 386, Charlottesville, Virginia 22908. E-mail: jr5n{at}virginia.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Undernutrition affects the immature animal by delaying growth and pubertal development (1, 6). However, it is not known whether undernutrition modifies ER-ir cell number in prepubertal animals in the same manner as it does in adult animals. Immature animals may respond to underfeeding differently from adults because the hypothalamic-pituitary-gonadal axis of immature animals is more sensitive to the negative feedback of sex steroids and because the ultrastructure of the medial basal hypothalamus and its neuronal circuits may not yet be fully mature (7).
Also uncertain are the specific mechanisms involved in the induction of GnRH pulse generator activity and the onset of puberty. The increase in GnRH activity is thought to be due to a maturing of the neural systems governing the GnRH pulse generator (7, 8). Underfeeding delays the pubertal onset of LH and, by inference, GnRH pulsatility and/or release (1, 6, 9). Thus, understanding the influence of underfeeding on the neural components of the GnRH pulse generator may help elucidate the mechanisms involved in pulse generator function and maturation. Furthermore, describing the site-specific changes in ER immunoreactivity during underfeeding and refeeding will help identify those brain regions important for controlling pulse generator maturation and pubertal progression (4). Thus, we investigated whether the responses to underfeeding in the prepubertal mouse are due to changes in ER-ir cell numbers in brain areas known to regulate GnRH.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Underfeeding
For all three experiments, weanling mice were underfed by giving
them only enough ground food so that their body weight was held at or
below their weanling body weight for the prescribed number of days. For
the females this was 0.45–0.75 g food; males received between
0.65–0.95 g food. Food consumption was measured by allowing animals to
eat from feeders designed to prevent food from being hoarded,
scattered, or contaminated with feces and bedding. The underfed animals
were fed at 0900 h each day and perfused before feeding on the
last day.
Immunocytochemistry
Animals were deeply anesthetized with sodium pentobarbital and
perfused through the left ventricle with 0.9% heparin-saline followed
by modified Zamboni’s fixative. Brains were removed and cryoprotected
overnight in 20% sucrose in 0.1 M sodium phosphate buffer
at 4 C and then quickly frozen. Serial coronal sections (30 µm) were
collected on a freezing cryostat into four wells containing antifreeze.
Sections were stored at 4 C until used for immunocytochemical analysis.
Equal numbers of brains from the treatment conditions were developed to
reduce interrun or development-related variability. One quarter of the
sections from each brain were rinsed and then incubated for 48 h
at 4 C in primary antiserum. ER-ir cells were detected with the rat
monoclonal antiserum H222 (provided by Abbott Laboratories, North
Chicago, IL; working dilution, 1:1000). Incubation in the primary was
followed by rinsing in Tris-buffered saline (pH 7.6) and a 60-min
incubation at room temperature in biotinylated donkey antirat IgG
(Vector Laboratories, Inc., Burlingame, CA; working dilution,
1:500). Then another series of rinses and 1-h incubation in
avidin-biotin-peroxidase complex (ABC; Vector Elite Kit, Vector
Laboratories; dilution, 1:1000) was performed. To increase cell
permeability, during the preincubation Triton X was added to the
carrier solution for the primary and secondary antibodies and to the
ABC. To enhance staining intensity, the reaction was double bridged by
placing the tissue back into the biotinylated donkey antirat IgG for
1 h and rinsing for 30 min in Tris-buffered saline, followed by a
45-min incubation in fresh ABC. Immunoreactivity was visualized with
diaminobenzadine and nickel to produce a purple staining of the
nucleus. The tissue sections were developed in diaminobenzadine for
5–15 min (median, 10 min).
Cell counting
The numbers of detectable ER-ir cells in representative brain
sections were counted. The areas examined were the mPOA, VMN, and
arcuate nucleus (ARH). All counts for a specific area were completed by
the same person, who was blind to the experimental condition of the
material. Detectable ER-ir cells were counted unilaterally in all three
brain areas. A single representative section was counted for each area.
The mPOA ER-ir cells were counted in the anterior portion of the mPOA
at first appearance of the crossed anterior commissure and anterior to
the merging of the lateral ventricles. The atlas coordinates were
Bregma 0.02 and -5 mm from the skull (10) (Fig. 1
). The VMN and ARH were counted where
the optic tract projected superiorly and posterior to the formation of
the infundibulum. This corresponded to atlas coordinates Bregma -1.58
and -5.5 mm (10) (Fig. 2). An Olympus
light microscope equipped with a video camera (Olympus Corp., Lake
Purchase, NY) connected to a personal computer and color monitor were
used for cell counting. Images were projected onto a monitor, and using
Mocha computer analysis software (Jandel Scientific, San Rafael, CA),
each immunoreactive cell was marked on the screen.
|
|
Procedures
Exp 1: long term underfeeding. The purpose of this
experiment was to investigate whether underfeeding female and male mice
could alter the neural development of detectable ER-ir cells compared
with that in ad libitum fed counterparts who were killed at
the very end of the prepubertal state. Weanling male and female CF-1
mice were each randomly assigned to one of three groups. Weanling mice
(nine males and seven females) were killed at 16 days. The female and
male mice in the control groups were allowed to eat ad
libitum for 7 and 17 days, respectively, and then killed. Underfed
male mice (n = 9) and female mice (n = 7) were given only
enough food (for 17 and 7 days, respectively) to maintain their
weanling body weights. The feeding treatment of the male mice continued
for a longer time than that of the female mice because male mice begin
puberty at an older age. Food consumption and body weights were
recorded daily.
Exp 2: time course of changes in ER immunoreactivity with underfeeding. The purpose of this experiment was to determine how quickly detectable ER-ir cell numbers are altered in response to underfeeding. Weanling male and female CF-1 mice were each randomly assigned to groups. One group was killed at weaning (day 16; seven males and eight females). The control groups were allowed to eat ad libitum for 24 h (five females) or 48 h (five males and nine females) and then were killed. Other groups of mice were underfed for 24 h (seven males and eight females) or 48 h (seven males and nine females) and killed. Data from a pilot study demonstrated that underfeeding prepubertal male mice for 24 h did not reduce ER-ir cell numbers. Thus, ad libitum fed males were not studied at 24 h. Food consumption and body weights were recorded daily.
Exp 3: time course of changes in ER immunoreactivity with refeeding. The purpose of this experiment was to investigate how quickly refeeding could increase detectable ER-ir cell numbers in males and females that had been underfed for 48 h. Weanling CF-1 mice were randomly assigned to groups. The mice in control groups were allowed to eat ad libitum for 72 h (six males and five females) or 96 h (four of each sex) and then killed. The underfed-refed groups were underfed for 48 h and then refed for 24 h (three males and eight females) or 48 h (three males and five females). Food consumption and body weights were recorded daily.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and 4). The detectable ER-ir cell number in
the ARH was unchanged in the underfed females. Underfeeding had little
effect on the detectable ER-ir cell number of the males (Fig. 3). In
females, there was a significant relationship (Fig. 5) between the percent body weight change
the last 24 h before perfusion and the ER-ir cell number in both
the mPOA (r = 0.69; P < 0.005) and VMN (r =
0.56; P = 0.01). Relationships between the percent body
weight change 24 h before perfusion and ER-ir cell number of male
mice were not significant (data not shown).
|
|
|
). The detectable
ER-ir cell number in the male mice was not significantly changed in the
mPOA (P = 0.10), VMN (P = 0.50), or ARH
(P = 0.32).
|
). The
P values for the change in detectable ER-ir cell number of
the male mice for the mPOA, VMN, and ARH were 0.49, 0.16, and 0.06,
respectively.
|
) and in the
mPOA (r = 0.66; P < 0.001), VMN (r = 0.33;
P = 0.06), and ARH (r = 0.45, P <
0.01) of male mice (Fig. 9).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the female mice, responses of ER immunoreactivity to underfeeding varied in different brain areas. When underfed for 7 days, detectable ER immunoreactivity was reduced in the mPOA and VMN, but not in the ARH. Underfeeding for 48 h reduced detectable ER-ir cell number in all three neural locations. In the female mouse, there appears to be a transient reduction in ARH ER-ir cells, followed by a rebound during long term underfeeding. The purpose of such a transition is unknown. However, the ARH, mPOA, and VMN consist of many cells that contain both ER and neuropeptides involved in regulating feeding behavior and control of the GnRH pulse generator (11). Perhaps the transient change is related to development of one or more of these neurotransmitter systems. Future investigations should use double staining techniques to determine whether underfeeding selectively reduces ER immunoreactivity in specific populations of neurons that coexpress such neurotransmitters.
The VMN is an important site for estrogen-mediated effects on estrous behavior, and restriction of metabolic fuels reduces detectable ER immunoreactivity in the VMN (4, 12). The present study demonstrates that the VMN is sensitive to alterations in metabolic fuels before sexual maturity and the onset of estrous behavior.
Estrogen also acts within the mPOA to facilitate estrous behavior (4, 13) and gonadotropin release (14). The reduction in ER-ir cell number in the mPOA of the prepubertal female mice is opposite to the effect in adult Syrian hamsters, in which underfeeding increases detectable ER-ir cells (4). The increase in detectable ER immunoreactivity in underfed adult hamsters may increase the positive feedback effects of estradiol so that when food becomes available, an enhanced LH secretion quickly normalizes the estrous cycle (15). However, the purpose of the reduction in ER immunoreactivity in the prepubertal mouse is unknown. Underfeeding impairs the hypothalamic release of GnRH, but not the assayable pool of brain GnRH (16). Thus, one possibility is a reduction in estrogen-stimulated neuropeptide synthesis and secretion. For instance, reductions in ER-ir cell number in dopamine-containing cells may diminish dopamine-stimulated GnRH release during the peripubertal period. We have found that underfeeding for 7 days reduces the total number of cells that coexpress ER immunoreactivity and tyrosine hydroxylase immunoreactivity by 57% in female mice (17). Whether ER-ir cell number is altered in other neurotransmitter-containing neurons requires further investigation.
In addition, although estrogen levels may be low in prepubescent mice, it does not mean that estrogen is biologically nonfunctional. Estrogen levels are probably low due to an exquisitely sensitive feedback between the gonad and the hypothalamus via estrogen and its receptor. Estrogen may cause the production and release of various neuropeptides involved in the organization and maturation of the GnRH pulse generator. A reduction in ER number may reduce this maturational effect and stall puberty.
Another important neural center for the control of GnRH pulsatility is the paraventricular nucleus (PVN). Investigations in rats (18, 19) have shown that estrogen feedback at the PVN is required to reduce LH pulsatility during fasting. In adult hamsters, food restriction increases detectable ER immunoreactivity in the posterior parvocellular PVN (20). However, in our study, ER-ir staining in the PVN of both ad libitum and underfed animals was either absent or very faint, sometimes barely distinguishable from background. The, ER-ir staining in other neural areas (e.g. ARH) within the same section was always robust and readily visible. Thus, we are confident that the weak ER-ir signal in the PVN was not caused by our methodology. The lack of detectable ER immunoreactivity in the PVN suggests that the central processing of metabolic information may differ in the prepubertal mouse.
The antibody used to stain for ER immunoreactivity in the present study (H222) is affected by estradiol concentrations, such that when the ER is bound to its ligand, H222 does not bind to the ER (21). However, in our underfed prepubertal animals, estradiol concentrations would be very low. Also, the results of a reduction in ER immunoreactivity in food-restricted animals is opposite what would be expected if estradiol levels influenced the detectable ER-ir cell number.
Differences in the effects of underfeeding on the number of detectable ER-ir cells between the present and previous studies may be related not only to species and maturational differences of the animal models, but also to differences in the feeding paradigm. Previous investigations have studied nutritionally induced changes in detectable neural ER-ir cell number by food deprivation (4, 5, 18) or chemical blockade of carbohydrate and fat metabolism (4, 5). We have demonstrated that chronic and acute underfeeding (not food deprivation) also produces alterations in detectable neural ER-ir cell number. Underfeeding is probably a more common situation in rodents in the wild as well as in the human. In addition, food deprivation does not induce neurogastric signals that independently modulate LH secretion in adult rats (22) and monkeys (23).
During the first experiment, several underfed females inadvertently given a small excess of food gained body weight during the last days of the underfeeding protocol. These females had ER-ir cell numbers similar to those of ad libitum fed animals (data not shown). Thus, we initiated studies of the time course of the underfeeding- and refeeding-induced alterations in ER immunoreactivity. Alterations in metabolic fuel availability induced changes in ER immunoreactivity within 24–48 h. In addition, in 7 of 11 underfed mice that were refed for 24 h, the refeeding increased detectable ER-ir cell numbers before their body weights recovered to normal. These data suggest that the availability of metabolic fuels, not body weight or body fat stores, is influencing ER immunoreactivity in prepubertal female mice.
Our study is also unique because nutritional alterations in the ER-ir cell number of males was investigated. Although the number of detectable ER-ir cells was not significantly affected by underfeeding in the male, there was a trend for a reduction in the short term experiments. However, the percent body weight change the last 24 h before perfusion was directly related to the number of detectable ER-ir cells during short term underfeeding. Thus, changes in the detectable ER-ir cell number in male mice may be subtle, and correlation analyses may be a more sensitive way of assessing the metabolic milieu at the time of perfusion. However, correlation analyses between the percent body weight loss and detectable ER-ir cell numbers in the ARH were not significant for the males after long term underfeeding or in the females after short or long term underfeeding, suggesting that ER immunoreactivity in the ARH may be less sensitive to the magnitude of weight loss.
Interestingly, detectable ER-ir cell number in the mPOA and VMN of the males were also not related to the body weight loss the last 24 h before perfusion in the first study when mice were chronically held at their weanling body weight. These results suggest that detectable ER-ir cell number may be slightly and transiently altered by underfeeding in the prepubertal male, but eventually normalized. Conversely, detectable ER-ir cell number in females was directly related to the percent body weight loss in both the short term (2 day) and long term (7 day) experiments.
One evolutionary hypothesis for the sexual dimorphism in the detectable ER-ir cell number response to underfeeding is that the GnRH pulse generator of an adult female mouse is more sensitive to food availability than that of an adult male mouse because the female must commit energy reserves to the pre- and postnatal development of the offspring (6). A logical extension of this hypothesis is that underfeeding may block puberty in the female mouse so that the undernourished animal does not attain a reproductively mature state. In both mature and juvenile female mice, quick neural responses to alterations in fuel availability allow the animals to adjust to the existing environmental conditions. For the weanling male mouse, the ability to continue to undergo pubertal development despite suboptimal food resources is essential because of the length of time (60 days) necessary to produce mature sperm. In the wild, the male mouse will often encounter a combination of short life expectancy and the requirement of finding a territory in the face of low food availability. If pubertal and spermatic development are halted until the male located a suitable territory with adequate nourishment, the reproductive rate could be very low (6). However, this evolutionary hypothesis is not universal. For instance, the hypothalamic-pituitary axis of the male rhesus monkey is very sensitive to food deprivation (24).
In conclusion, reduced nutritional status (metabolic fuel availability) generally reduces detectable ER-ir cell numbers in the mPOA, VMN, and ARH of prepubertal female mice. Each of these neural areas has been implicated in controlling the onset of GnRH pulse generator activity (11). In the female, detectable ER-ir cell numbers respond to changes in fuel availability within 24–48 h. Changes in detectable ER-ir cell number probably represent an attempt to change neural sensitivity to sex steroids. Future research should determine in which populations of cells ER-ir cell numbers are being reduced, as this will help determine which neuropeptides may be involved in the control of GnRH release and the triggering of puberty. In contrast, food restriction has little effect on detectable ER-ir cell numbers in male mice. The dimorphic response of neural ER immunoreactivity to underfeeding supports the hypothesis that there is an evolutionary adaptation that allows males to continue sexual maturation despite inadequate metabolic fuels because of the time required to produce mature sperm (6).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 8, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. A S Reyes, H. Tsukamura, H. I'Anson, M. A. C Estacio, K. Hirunagi, and K.-I. Maeda Temporal expression of estrogen receptor {alpha} in the hypothalamus and medulla oblongata during fasting: a role of noradrenergic neurons. J. Endocrinol., September 1, 2006; 190(3): 593 - 600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. McManus, R. L. Goodman, N. V. Llanza, M. Valent, A. B. Dobbins, J. M. Connors, and S. M. Hileman Inhibition of Luteinizing Hormone Secretion by Localized Administration of Estrogen, but not Dihydrotestosterone, Is Enhanced in the Ventromedial Hypothalamus During Feed Restriction in the Young Wether Biol Reprod, October 1, 2005; 73(4): 781 - 789. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. N. Wade and J. E. Jones Neuroendocrinology of nutritional infertility Am J Physiol Regulatory Integrative Comp Physiol, December 1, 2004; 287(6): R1277 - R1296. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Kudwa, J.-A. Gustafsson, and E. F. Rissman Estrogen Receptor {beta} Modulates Estradiol Induction of Progestin Receptor Immunoreactivity in Male, But Not in Female, Mouse Medial Preoptic Area Endocrinology, October 1, 2004; 145(10): 4500 - 4506. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-S. Parent, G. Teilmann, A. Juul, N. E. Skakkebaek, J. Toppari, and J.-P. Bourguignon The Timing of Normal Puberty and the Age Limits of Sexual Precocity: Variations around the World, Secular Trends, and Changes after Migration Endocr. Rev., October 1, 2003; 24(5): 668 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Hill and J. E. Levine Abnormal Response of the Neuropeptide Y-Deficient Mouse Reproductive Axis to Food Deprivation But Not Lactation Endocrinology, May 1, 2003; 144(5): 1780 - 1786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Panicker, R. A. Mangels, J. B. Powers, G. N. Wade, and J. E. Schneider AP lesions block suppression of estrous behavior, but not estrous cyclicity, in food-deprived Syrian hamsters Am J Physiol Regulatory Integrative Comp Physiol, July 1, 1998; 275(1): R158 - R164. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
0.05.
