| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Article |
a Reproductive Sciences Program
b and Departments of Obstetrics and Gynecology,
c Biology,
d and Physiology, University of Michigan, Ann Arbor, Michigan 48109-0404
e Laboratory of Animal Reproduction, Nagoya University, Nagoya 464–01, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The temporal relationship between insulin action and reproductive function is also a fundamental question pertaining to the metabolic control of the timing of puberty. Delayed puberty has long been associated with poorly controlled diabetes [6], yet virtually nothing is known about the regulation of LH secretion by insulin in the developing individual. In view of the hypothesis that changes in glucose availability are important to the timing of sexual maturation, a greater understanding of the regulatory role of insulin upon LH secretion in the immature individual becomes paramount. To begin to test the hypothesis that GnRH secretion is modulated by insulin, the present study examined LH secretion in the immature, diabetic sheep. Our aim was to disassociate the acute effects of hypoinsulinemia, from the metabolic and physiologic consequences of chronic insulin deficiency. Our approach was to monitor LH secretion in growing diabetic lambs, both in the presence and in the absence of exogenous insulin. We first determined whether chronic insulin supplementation could sustain growth and LH secretion during development. Secondly, we determined if short-term insulin withdrawal and re-supplementation could influence LH secretion. Lastly, we determined if the effects of hypoinsulinemia upon LH secretion are influenced by gonadal steroids. The site(s) of insulin action in this animal model is investigated in a companion paper [7].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Spring-born (April) male lambs of predominantly Suffolk breeding were used. They were born at the University of Michigan Reproductive Sciences Program Sheep Research Facility, Ann Arbor, MI. To remove sex steroid negative feedback control of LH secretion, the testes were removed at 7 days of age by bilateral orchidectomy under local anesthesia (Lidocaine; Butler Co., Columbus, OH). After weaning (8 wk of age), the lambs were fed ad libitum (Kent Lamb Pellets and hay). This diet was maintained throughout the course of the study. Water and supplemental vitamins and minerals were provided at all times. All procedures were approved by the University Committee for the Use and Care of Animals at the University of Michigan.
Streptozotocin/Insulin Treatments
To create a diabetic metabolic state, each lamb received two doses of Streptozotocin (STZ; Sigma Chemical Co., St. Louis, MO; 100 mg/kg, i.v.) three days apart at 12 wk of age. This treatment paradigm was adapted from the protocol established in the pregnant ewe [8]. The diabetic condition (hyperglycemia, absence of glucose-induced insulin release, prolonged glucose clearance) was confirmed by a glucose-tolerance test performed before and 2 wk after STZ treatment (Fig. 1). Beginning 2 wk after STZ treatment, beef/pork insulin (1.0 U/kg, Lente Iletin 1; Eli Lilly & Co., Indianapolis, IN) was continually replaced (every 12 h, by subcutaneous [s.c.] injection) except for brief periods of withdrawal (see Experimental Designs, below). We chose this heterospecific insulin based upon our previous findings [9], which determined this inexpensive source of insulin to be biologically active (reliably suppressed blood glucose) in non-diabetic sheep.
|
Experimental Designs
Chronic insulin treatment To determine if daily insulin supplementation can sustain growth, as well as LH secretion, in STZ-induced diabetic lambs (n = 6), we monitored body weights, LH secretory patterns, and glucose and insulin concentrations before and after STZ treatment. At 12 wk of age (28 ± 1.1 kg BW), jugular blood was collected every 10 min for 6 h. Upon completion of sampling, the lambs were rendered diabetic by STZ treatment (see protocol above). Beginning at 14 wk of age (after confirmation of the diabetic model), insulin (1.0 U/kg) was administered every 12 h. To approximate euglycemic conditions, the daily insulin doses were adjusted (increased or decreased) according to glucose values determined by a portable glucometer (Lifescan; Johnson & Johnson, Milpitas, CA). After 4 wk of insulin supplementation (18 wk of age, 41 ± 1.4 kg BW), jugular blood was again collected every 10 min for 6 h.
Acute insulin withdrawal and re-supplementation To determine if LH secretion is impaired after insulin withdrawal, we monitored LH secretory patterns, together with body weight, glucose, insulin, non-esterified fatty acid (NEFA), cortisol, and urinary ketone concentrations in nine diabetic lambs (original six lambs plus three additional ones) before and after acute (24 h) or chronic (96 h) periods of insulin withdrawal. In addition, to determine if LH secretion is restored after insulin re-supplementation, we monitored LH secretory patterns, together with body weight, glucose, insulin, NEFA, cortisol, and urinary ketone concentrations after acute or chronic periods of insulin re-supplementation. At 26 wk of age (55 ± 1.7 kg BW), jugular blood was collected every 10 min for 6 h on a random day of insulin supplementation. This same sampling paradigm was repeated after 24 h of insulin withdrawal, after 96 h of insulin withdrawal, after 24 h of insulin re-supplementation, and after 96 h of insulin re-supplementation. Samples were pooled for glucose, insulin, NEFA, and cortisol determinations. Urine samples were collected at random for ketone determinations.
Insulin withdrawal and gonadal steroid replacement To determine if the effect of insulin withdrawal on LH secretion is influenced by gonadal steroids, estradiol was replaced at physiological levels (3–5 pg/ml, [10]). The same 9 lambs (20 wk of age, 47 ± 1.8 kg BW) were implanted (s.c.) with a small Silastic capsule (o.d. 0.46 cm, i.d. 0.34 cm; Dow Corning, Midland, MI) containing a 30-mm packed column of crystalline estradiol-17ß (Sigma Chemical Co.), and sealed with Silastic adhesive type-A (Dow Corning). This implant maintains estradiol in lambs at a constant level of approximately 3–5 pg/ml [10]; estradiol was used in male lambs because males have a similar degree of responsiveness to estradiol and testosterone with regard to the regulation of LH pulse frequency [11]. Ten days after the beginning of estradiol treatment, we compared LH secretory patterns on a day of insulin supplementation and after 96 h of insulin withdrawal.
Blood Sample Collection
All blood samples (2.5 ml) were collected through indwelling jugular catheters (Angiocath; Becton Dickinson, Mountain View, CA) and were placed into tubes containing 20 IU heparin. After centrifugation, plasma was harvested and frozen for later analysis.
Hormone Assays
LH was measured in duplicate 25- to 200-µl aliquots of plasma using a modified [12, 13] RIA developed by Niswender et al. [14]. Assay sensitivity, defined as 2 SD from the buffer control value, averaged 0.86 ± 0.06 ng/ml (n =13 assays) for 200 µl plasma expressed relative to NIH-LH-S12. Intraassay coefficient of variation (CV), determined from a serum pool that bound at 50%, averaged 7.64%; interassay CV averaged 10.97%. For a serum pool that bound at 20%, intraassay CV averaged 6.06%, and interassay CV averaged 7.48%.
Glucose was quantified from fresh whole blood using a portable glucometer (Lifescan; Johnson & Johnson, [2]).
Insulin was measured in duplicate 100-µl aliquots of plasma using a solid-phase radioimmunoassay kit (ICN Pharmaceuticals, Inc., Costa Mesa, CA) validated for use in sheep by Hileman et al. [15]. Assay sensitivity, defined as 2 SD from the buffer control value, averaged 4.13 ± 0.58 µU/ml (n = 11 assays). Intraassay coefficient of variation (CV), determined from a serum pool that bound at 60%, averaged 7.3%; interassay CV averaged 16.37%.
Non-esterified fatty acids (NEFA) were measured in plasma by spectrophotometric analysis (absorbance @ 550 nm) using a commercial kit from Wako Chemicals, Inc. (Richmond, VA), following the modified procedure of McCutcheon et al. [16].
Urine ketone concentrations were estimated using Ketostix (Bayer Corp., Elkhart, IN). The sensitivity of this colorimetric test strip for the presence of acetoacetic acid in the urine ranges from 5 mg/dl (trace) to 160 mg/dl (large).
Cortisol was measured in duplicate 25-µl aliquots of plasma using a solid-phase radioimmunoassay kit (Diagnostic Products Corp., Los Angeles, CA) validated for use in the sheep by Battaglia et al. [17]. Assay sensitivity, defined as 2 SD from the buffer control value, averaged 0.74 ± 0.22 ng/ml (n = 2 assays). Intraassay coefficient of variation (CV), determined from a serum pool that bound at 50%, averaged 8.36%; interassay CV averaged 9.84%.
Statistical Analyses
For the identification of LH pulses in the samples collected at 10-min intervals, Cluster analysis, developed by Veldhuis et al. [18] was used. The nadir and peak clusters were 1/1 points; the t statistics for significant increases and decreases were 2.6/2.6. The number of identified LH pulses per unit of time (6 h) was compared with or without insulin supplementation. The amplitude of LH pulses was similarly compared with or without insulin supplementation. Means for each period were subjected to analysis of variance with repeated measures, followed by Scheffé F tests (Statview SE + graphics; Brainpower, Inc., Calabasas, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
After STZ treatment and before insulin supplementation, body weight decreased (1.5 ± 0.34 kg/wk). Twice-daily treatment with long-acting insulin (beginning at Week 14) minimized diabetes-induced hyperglycemia, and sustained growth (2.4 ± 0.17 kg/wk before STZ treatment, 1.86 ± 0.34 kg/wk after STZ treatment, P > 0.05; Fig. 2, top). These growth rates are consistent with those found in non-diabetic individuals in our laboratory at comparable stages of development [10]. Moreover, insulin treatment maintained LH pulses at a frequency comparable to pre-diabetic conditions (5.33 ± 0.62 pulses/6 h before STZ, 6.0 ± 0.63 pulses/6 h, 6 wk after STZ, P > 0.05; Fig. 2, bottom). These LH pulse frequencies are consistent with those found in non-diabetic individuals in our laboratory at comparable stages of development [10]. LH pulse amplitude was not affected in diabetic lambs with insulin supplementation (14.9 ± 0.94 ng/ml before STZ, 13.5 ± 0.83 ng/ml after STZ, P > 0.05).
|
Acute Insulin Withdrawal and Re-Supplementation
Figure 3 presents mean levels of insulin, glucose, body weight, NEFA, urinary ketones, and cortisol. Mean LH pulse frequencies and amplitudes are shown in Figure 4, accompanied by representative profiles of plasma LH concentrations in individual lambs. Twenty-four hours of insulin withdrawal increased circulating glucose levels, increased the concentration of plasma NEFAs, and increased urinary output of ketones, but it did not reduce body weight. Under these metabolic conditions, LH pulse frequency was decreased (8.1 ± 0.26 pulses/6 h on a day of insulin supplementation versus 6.7 ± 0.24 pulses/6 h after 24 h of insulin withdrawal, P < 0.05). Moreover, 96 h of insulin withdrawal did not increase circulating glucose levels further (P > 0.05), but the chronic absence of insulin decreased body weight, increased concentrations of plasma NEFAs, elevated plasma cortisol, and markedly increased urinary output of ketones (P < 0.05). LH pulse frequency continued to decline after 96 h of insulin withdrawal (6.0 ± 0.6 pulses/6 h). These changes in metabolism and LH secretion were reversed by acute insulin re-supplementation. Twenty-four hours of insulin replacement reduced circulating glucose and NEFA concentrations, decreased plasma cortisol, and reduced urinary output of ketones (P < 0.05), but it did not increase body weight (P > 0.05). Accompanying these acute changes in metabolism, LH pulse frequency increased (7.2 ± 0.5 pulses/6 h, P < 0.05). LH pulse frequency continued to increase during the 96 h of insulin re-supplementation to levels comparable with those before insulin withdrawal (8.6 ± 0.29 pulses/6 h, P > 0.05). Moreover, 96 h of insulin re-supplementation did not increase body weight, glucose, or cortisol values further than those measured at 24 h, but it did decrease plasma NEFAs and urinary output of ketones. LH pulse amplitude did not undergo any pattern of change, and it remained essentially constant throughout the course of the experiment.
|
|
Insulin Withdrawal and Gonadal Steroid Replacement
Figure 5 depicts the patterns of LH secretion before and after insulin withdrawal in the absence of steroids and in the presence of exogenous estrogen. During insulin supplementation, estradiol treatment decreased LH pulse frequency (8.1 ± 0.26 pulses/6 h in the absence of steroid; 5.4 ± 0.48 pulses/6 h in the presence of steroid, P < 0.05; Fig. 5, +Insulin). After 96 h of insulin withdrawal, LH pulse frequency decreased (6.0 ± 0.6 pulses/6 h in the absence of steroid, P < 0.05; 2.9 ± 0.87 pulses/6 h in the presence of steroid, P < 0.05; Fig. 5, -Insulin 96h). The decline in LH pulse frequency after 96 h of insulin withdrawal in the presence of estradiol (47%) was greater than the decline in these same individuals without steroid supplementation (26%). The amplitude of LH pulses was decreased by estradiol treatment both before and after insulin withdrawal (P < 0.05).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our present findings also determined that longer-term (96 h) insulin withdrawal exaggerates the short-term (24 h) effects of insulin withdrawal on LH pulsatility. These changes in LH occurred during a period of time (i.e., -insulin 24 h to -insulin 96 h) when insulin and glucose values remained constant, while both plasma concentrations of NEFA and cortisol, and urinary concentrations of ketones increased. This observation is consistent with a report in men whose diabetes-induced hyperglycemia did not suppress serum testosterone unless their diabetic condition was complicated by ketosis [19]. On the other hand, high levels of free fatty acids, per se, were unlikely to have contributed to LH suppression in our diabetic sheep based on the report of Estienne et al. [20], who found that increasing plasma lipid concentrations (intravenous infusion of a commercial parenteral lipid solution) did not decrease LH secretion in well-nourished, non-diabetic sheep. In addition, our finding that cortisol concentrations increased in the diabetic lamb between 24 and 96 h of insulin withdrawal, is consistent with previous results in the sheep that determined that elevated levels of cortisol (> 100 ng/ml) accompany endotoxin-induced suppression of LH secretion [17]. However, the modest increase in cortisol found in our present study (20 ng/ml) was likely insufficient to depress LH pulse frequency. Nonetheless, we cannot rule out the possibility that activation of the stress axis contributed to the decline in LH in the diabetic sheep (perhaps by an increase in corticotropin releasing hormone). Leptin levels were not determined in our current study (at present, no assay for ovine leptin is extant). However, recent work in the rodent by ourselves [21] and others [22] suggests that this adipose tissue-derived hormone is decreased in association with hypoinsulinemia (fasting and diabetes). Whether changing levels of leptin in the diabetic sheep have a correlative and/or causal relationship to changes in the frequency of LH secretion remains an interesting question.
In contrast with our conclusions derived from the diabetic sheep, Williams et al. [2] have argued that in the male monkey, insulin does not participate in the metabolic regulation of LH secretion. In their non-diabetic, monogastric animal model, 24 h of fasting reduces plasma insulin and suppresses LH; re-feeding restores insulin values in association with an increase in LH pulse frequency. Circulating glucose values remain unchanged during this dietary manipulation. However, pharmacologically blocking the feeding-induced rise in insulin did not prevent the feeding-induced increase in LH, despite a relative increase in circulating glucose. Based on these findings, they concluded that insulin is not causally associated with LH secretion. Such a conclusion is not supported by our overall results, wherein a diabetes-induced decrease in LH secretion was clearly alleviated by insulin supplementation. However, LH secretion in our diabetic model is not dependent wholly upon insulin activity, per se, because LH pulse frequency remained relatively high (hourly) even after 96 h of insulin withdrawal. One explanation for this paradoxical finding is that LH secretion may be sustained by high glucose (in the face of low insulin) through putative glucodetectors that do not rely on insulin for glucose transport and metabolism. In this regard, specialized cells exist within the hypothalamus that are exquisitely glucosensitive, even in the absence of insulin [23]. In addition, insulin is not required for glucose to cross the blood-brain barrier, or for transport into neuronal tissue [24]. On the other hand, both insulin and its receptor, as well as GLUT 4 (the insulin-dependent glucose transporter), have been identified in key brain regions known to influence GnRH secretion (hypothalamus and hindbrain) [25, 26]. Therefore, the possibility exists that glucomodulation of GnRH secretion could be both insulin-dependent as well as insulin-independent. If this were the case, low insulin (peripheral and/or central) would compromise, but not abolish, glucose-associated GnRH secretion. For example, basal levels of insulin in the presence of sufficient metabolic substrate (glucose) could promote the release of GnRH (the case of the re-fed monkey [2]). On the other hand, glucose alone, in the absence of insulin, will not sustain ultra-high frequency GnRH/LH secretion (the case of the diabetic sheep).
The present study also considered the role of sex steroids in the metabolic control of LH secretion. It has previously been established that suboptimal nutrition can impair LH secretion independent of gonadal steroids in several animal models including the sheep [5, 21]; this effect may be magnified in gonadal intact individuals or castrates with gonadal steroids replaced ([27], unpublished). Our current finding that estradiol treatment potentiates the decline in LH pulse frequency found after insulin withdrawal is consistent with such reports in both diabetic and non-diabetic animal models. For example, in both the rat [28] and the monkey [29] with normal insulin secreting capacity, short-term fasting decreases LH secretion; this effect is markedly enhanced in the presence of gonadal steroids. Moreover, in the diabetic rodent, estradiol treatment blocks the abrupt increase in mean LH normally found after gonadectomy [30]. Such steroid-dependent effects may be due to recruitment/up-regulation of estradiol receptors in important brain regions associated with GnRH secretion during periods of metabolic challenge [31, 32]. This same mechanism may also apply to our current finding that estradiol treatment suppresses LH pulse frequency even during insulin supplementation, when the diabetic condition is improved but not completely normalized.
We have previously determined that reducing the availability of glucose (pharmacologic blockade of glucose metabolism) either peripherally or centrally profoundly and transiently suppresses LH release [33]. However, our present results do not determine any causal relationship between LH secretion and glucose availability, nor do they uncover the site of insulin action. In this latter regard, we now have evidence in the diabetic sheep which suggests that insulin acts centrally to stimulate GnRH release [7]. On the other hand, we cannot conclude from our present findings, nor from those of Tanaka, et al. [7], whether insulin affects LH secretion directly (independent of changes in glucose uptake and metabolism), or indirectly (by promoting glucose availability at key neuronal centers known to influence GnRH secretion [34].
To date, many putative metabolic signals have been proposed to link somatic metabolism with reproductive function. Our present study provides evidence that in the diabetic lamb, insulin modulates LH pulse frequency, and that such effects are potentiated by, but not dependent upon, gonadal steroids. In a broader context, these results are consistent with the hypothesis that insulin is a metabolic cue interfacing energy metabolism with GnRH secretion.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by grants from the NIH (HD-18394 and HD-18258). A preliminary report of this work was presented at the 28th Annual Meeting of the Society for Neuroscience, Los Angeles, CA, November, 1998 (Abstract #110.2). 
2 Correspondence: Douglas L. Foster, Room 1138, 300 North Ingalls Building, University of Michigan, Ann Arbor, MI 48109-0404. FAX: 734 936 8620; dlfoster{at}umich.edu
3 Current address: University of Wisconsin Medical School, Dept. of Biomolecular Chemistry, 1300 University Ave., Madison, WI 53706.
Accepted: January 18, 2000.
Received: August 11, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. R. Sessions, S. E. Reedy, M. M. Vick, B. A. Murphy, and B. P. Fitzgerald Development of a model for inducing transient insulin resistance in the mare: Preliminary implications regarding the estrous cycle J Anim Sci, August 1, 2004; 82(8): 2321 - 2328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ohkura, T. Ichimaru, F. Itoh, S. Matsuyama, and H. Okamura Further Evidence for the Role of Glucose as a Metabolic Regulator of Hypothalamic Gonadotropin-Releasing Hormone Pulse Generator Activity in Goats Endocrinology, July 1, 2004; 145(7): 3239 - 3246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S T Butler, S H Pelton, and W R Butler Insulin increases 17{beta}-estradiol production by the dominant follicle of the first postpartum follicle wave in dairy cows Reproduction, May 1, 2004; 127(5): 537 - 545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. C. Lopez-Alvarenga, T. Zarinan, A. Olivares, J. Gonzalez-Barranco, J. D. Veldhuis, and A. Ulloa-Aguirre Poorly Controlled Type I Diabetes Mellitus in Young Men Selectively Suppresses Luteinizing Hormone Secretory Burst Mass J. Clin. Endocrinol. Metab., December 1, 2002; 87(12): 5507 - 5515. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ramanathan, K. Shirota, S. Morita, T. Nishimura, Y. Huang, X. Zheng, and S. Hunyor Left ventricular oxygen utilization efficiency is impaired in chronic streptozotocin-diabetic sheep Cardiovasc Res, September 1, 2002; 55(4): 749 - 756. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Sisk, H. N. Richardson, P. E. Chappell, and J. E. Levine In Vivo Gonadotropin-Releasing Hormone Secretion in Female Rats during Peripubertal Development and on Proestrus Endocrinology, July 1, 2001; 142(7): 2929 - 2936. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ohkura, T. Tanaka, S. Nagatani, D. C. Bucholtz, H. Tsukamura, K.-I. Maeda, and D. L. Foster Central, But Not Peripheral, Glucose-Sensing Mechanisms Mediate Glucoprivic Suppression of Pulsatile Luteinizing Hormone Secretion in the Sheep Endocrinology, December 1, 2000; 141(12): 4472 - 4480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tanaka, S. Nagatani, D. C. Bucholtz, S. Ohkura, H. Tsukamura, K.-I. Maeda, and D. L. Foster Central Action of Insulin Regulates Pulsatile Luteinizing Hormone Secretion in the Diabetic Sheep Model Biol Reprod, May 1, 2000; 62(5): 1256 - 1261. [Abstract] [Full Text]
|
|
|
|
 |
|
 E. W. C. M. Van Dam, F. Roelfsema, J. D. Veldhuis, F. M. Helmerhorst, M. Frolich, A. E. Meinders, H. M. J. Krans, and H. Pijl Increase in daily LH secretion in response to short-term calorie restriction in obese women with PCOS Am J Physiol Endocrinol Metab, April 1, 2002; 282(4): E865 - E872. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
