| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
-Aminobutyric AcidB Agonist Reverses the Negative Feedback Effect of Testosterone on Gonadotropin-Releasing Hormone and Luteinizing Hormone Secretion in the Male Sheep1
Department of Veterinary Biosciences, University of Illinois, Urbana, Illinois 61802
Address all correspondence and requests for reprints to: Dr. Gary L. Jackson, Department of Veterinary Biosciences, 2001 South Lincoln Avenue, Urbana, Illinois 61802. E-mail: g-jackson{at}uiuc.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-aminobutyric
acid (GABA) as a candidate. GABA is widely distributed throughout the hypothalamus. GABAergic neurons concentrate steroids (7, 8, 9, 10) and synapse with GnRH-secreting cell bodies in the preoptic area (11). Within the medial basal hypothalamus (MBH) there are high affinity uptake sites for GABA, a dense plexus of GABAergic neurons, and messenger RNAs coding for multiple GABA receptor subunits (12, 13, 14, 15, 16). Application of drugs that alter GABA metabolism and of GABA antagonists and agonists, both in vivo and in vitro, disrupt GnRH or LH secretory patterns (17, 18, 19, 20). Furthermore, that observation that T increases GABA turnover coincident with suppressed LH release (21, 22) is consistent with the hypothesis that GABA at least in part mediates the action of T.
However, the specific role of GABA in regulating GnRH remains unclear. Within the hypothalamus, GABA probably acts via one or both of the two major receptor types, GABAA and GABAB. The postsynaptic GABAA receptor, activation of which usually results in decreased neuronal responsiveness, probably mediates the inhibitory action of GABA. The GABAB receptors, originally thought to be strictly or largely presynaptic autoreceptors, have been less intensely investigated (23). However, observations by Scott et al. (24) that localized microinjection of baclofen, a GABAB agonist, into the medial preoptic area of estrogen-treated ewes during the anestrous period transiently elevated LH concentrations in some individuals suggested a possible role for this receptor type. This observation was extended by Ferreira et al. (17), who found that application of baclofen into the MBH of castrated rams by microdialysis resulted in prolonged and robust increases in LH pulse amplitude, with no observable effect on pulse frequency. This unexpected observation raised several questions. 1) Would baclofen increase LH in T-treated castrated and intact rams? 2) Was the increased LH secretion due to increased secretion of GnRH? 3) Was the increased secretion of LH paralleled by increased secretion of FSH? The experiments reported here were conducted to answer these questions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Surgery
Guide cannula placement. Surgery for bilateral placement of
the guide cannulae was carried out under aseptic conditions, using
modifications of procedures previously described (17).
Anesthesia was induced with sodium thiopental and was maintained with
3–4% halothane. The animal’s head was secured firmly in a
stereotaxic instrument (Kopf Instruments, Tujunga, CA). After an
incision, a circular piece of skull (2.5 cm in diameter) was removed,
and the sagittal sinus was doubly ligated. The sinus was then
retracted, and 0.15 cc of radiopaque dye (Renografin-76, E.R. Squibb
and Sons, Princeton, NJ) was injected into the third ventricle. Lateral
radiographs that outline the ventricle were used to aid in the
placement of guide cannulae. The final placements were made using an x,
y, z manipulator and additional radiographs. Twenty-gauge stainless
steel guide cannulae 62 mm long with stylets extending an additional 1
mm (Exp 1) or 3 mm (Exp 2) were then placed bilaterally into the MBH.
For Exp 1, the tips of the stylets were placed 2.8–3.0 mm above the
floor of the ventricle, 1.0 mm anterior to the most anterior portion of
the posterior wall of the infundibular recess, and 2.25 mm lateral to
midline. Given that the microdialysis probes extended 3 mm beyond the
guide tubes, the dorsoventral target of the probe tip was 0.8–1.0 mm
above the floor of the ventricle. For Exp 2, the tips of the stylets
were placed at this location. The cannulae and a protective cap were
anchored to the skull with dental acrylic and screws, and the incision
was then closed.
In all studies, butorphanol (Torbugesic, Fort Dodge Laboratories, Fort Dodge, IA) was provided as analgesic for 2 days postsurgery, and Liquamycin, LA-200 (0.1 cc/kg; Pfizer, Inc., New York, NY) antibiotic was provided for 10 days.
Portal cannula placement. For Exp 2, approximately 3 weeks after guide cannula placement the animals were reanesthetized, and a portal cannula device was installed and maintained as previously described (25, 26, 27). The only change was that the cannula device was flushed daily with a saline-heparin solution (125 U/ml) instead of saline solution.
The animals were then placed in adjacent individual pens. Approximately 7 days later the dual drug perfusion-portal blood sampling procedure was conducted.
Experimental design and protocol
Exp 1: effects of T and baclofen. The objective of this
experiment was to determine whether T affected the ability of baclofen
to increase the level of circulating LH. The experiment was a factorial
with two factors: steroid (T vs. no T) and drug [baclofen
vs. artificial cerebrospinal fluid (aCSF)]. A cross-over
design was used so that within steroid groups each animal was treated
with both aCSF-aCSF and aCSF-baclofen in random order. There were seven
animals each within the T and no T groups. All animals had been
castrated at least 60 days before receiving bilateral guide cannulas. T
treatment was achieved by placing SILASTIC implants (Dow Corning, Inc., Midland, MI; 3.32 mm i.d. x 4.61 mm o.d. x 10 cm)
containing packed crystalline T sc over the dorsal rib area at a dose
of 1.70–2.0 cm/kg BW (25). The implants were left in
place for the duration of the study.
Starting approximately 4 weeks after implantation of the guide cannulas and 1 day before drug dialysis, groups of three animals had a catheter inserted into the jugular vein for the purpose of blood collection. Each animal was placed into an individual metabolism pen and allowed 12–16 h to acclimate. On the day of the experiment, the stylets were removed from the guide cannulas and replaced by microdialysis probes. The probes were anchored in place by paper tape. The probes were connected to an infusion pump (Sage Instruments, Boston, MA) by means of microline tubing (Cole Parmer, Chicago, IL). Each animal then received a 4-h period of control dialysis of aCSF at a flow rate of 2 µl/min. During the subsequent 4 h, either dialysis of aCSF was continued or the solution was switched to one containing baclofen (1 mM; Sigma, St. Louis, MO). The probes were not changed. Due to dead space that was filled with aCSF within the probe assembly, it required approximately 30 min for the baclofen to move from the inlet to the probe tip and into the surrounding tissue. During the entire 8-h dialysis period, jugular blood samples were collected at 10-min intervals. Blood samples were collected in glass tubes containing 100 µl heparin (125 U/ml) and were centrifuged within 1 h of collection. Plasma was stored at -20 C until assayed for LH. After each dialysis session, the microdialysis probes were replaced by sterilized stylets, and the animals were returned to group pens after receiving Liquamycin LA-200. Subsequent dialysis sessions were conducted at 1-week intervals.
Exp 2: effects on GnRH secretion. The objectives of this experiment were to determine whether 1) baclofen would elevate LH in intact males; and 2) the baclofen-induced elevation in LH was accompanied by coincident elevation of GnRH secretion.
Adult males, approximately 1–2 yr old were subjected to placement of bilateral guide tubes as described for Exp 1. Three weeks later each animal was fitted with a portal-cannulation device. Then 7–10 days later they were subjected to simultaneous bilateral microdialysis and collection of portal blood using procedures described previously (25, 26, 27).
Approximately 12–14 h before this procedure three catheter assemblies were inserted into the jugular veins. Two were used to administer heparin and one to collect blood. Pairs of animals were placed into adjacent individual pens and allowed to acclimate. On the morning of the procedure, the stylets were removed, dialysis probes were inserted, and the dialysis of aCSF was started. Approximately 10–15 min later heparinization of the animal was started, and portal sampling was conducted as described previously (25, 26, 27). To maintain stable blood flow heparin was infused at a rate of 200 IU/kg·h and was supplemented with periodic injection as necessary.
Microdialysis of aCSF was continued until baseline portal and jugular blood samples had been collected for 3 h. Then the microdialysate was changed to a 1-mM baclofen solution, and microdialysis was continued for 4 h. Given that in previous experiments we repeatedly had found no effect of aCSF during the second half of perfusion, that treatment was not included in this experiment. During the entire procedure jugular blood samples were collected continuously into 9-min fractions, whereas portal samples were collected into 4.5 or 3 min (animal 9915) fractions. The animals were killed after the dialysis-portal collection procedure.
Dialysis probe and dialysis buffer
The microdialysis probe had a nitrocellulose hollow fiber
dialysis membrane with a molecular mass cut-off of 6 kDa (Spectra/Por,
Spectrum, Gardena, CA). The probe was of a concentric design adapted
for use in sheep with modifications previously described
(17). It was constructed in our laboratory from 24-gauge
stainless steel tubing through which fused silica tubing passed
(Polymicro Technologies, Phoenix, AZ) and exited from the microline
inlet. The silica tubing extended 1.75 mm from the stainless guide
around which the dialysis membrane (length, 2.5 mm) was fitted. The
distal end of the dialysis membrane was sealed with epoxy (Devcon
Corp., Danvers, MA). The final length of dialysis membrane in
direct contact with brain tissue was 2 mm. It should be noted that the
maximal dorsoventral dimension of the ventromedial nucleus is
approximately 3.5 mm in sheep (28). The aCSF consisted of
127.6 mM NaCl, 2.5 mM KCl, 0.69 mM
CaCl2, 1 mM
MgSO4, 2.3 mM
NaH2PO4, and 9.7
mM Na2HPO4 (pH
7.4).
Hormone assay
Plasma samples were assayed in duplicate for LH using a
previously described RIA validated for use in our laboratory
(25). The sensitivity was 0.5 ng/ml NIH LH S2O at 90%
binding. The intraassay coefficient of variation was 3.4%, and the
interassay coefficient of variation averaged 10.0% for three pooled
sera. Values for mean LH, LH interpulse interval, and LH pulse
amplitude were determined using the Pulsar algorithm (29).
The G values were set as G(1) = 2.2 and G(2, 3, 4, 5) = 2.0 for LH
and G(1) = 3.4, G(2) = 3.3, and G(3, 4) = 3.2 for
GnRH. FSH was assayed using the assay kit provided by the NIDDK as
previously validated in our laboratory (30). FSH RP-2 was
used as the standard. The assay sensitivity was 0.5 ng/ml. Intra- and
interassay coefficients of variation for four pooled sera averaged
3.1% and 12.4%.
T also was assayed using a previously described RIA validated for use in our laboratory (25). The sensitivity was approximately 0.4 ng/ml, and the intra- and interassay coefficients of variation were 6.3% and 7.6%, respectively.
GnRH was assayed using antibody BDS-037 in a previously validated procedure (31). The sensitivity was 2.0 pg/ml; intra- and interassay coefficients of variation were 8.7% and 15.4%, respectively. With this assay there was no detectable background for GnRH in peripheral blood.
Histology
At the end of the experiment, the animals were killed. The
brains were removed after perfusion with saline via carotid artery,
followed by 4% formalin fixative, after which the hypothalami were
isolated and immersed in fixative. The sections collected were
histologically processed and stained with Luxol fast blue to localize
probe placement. Evaluation of probe placement was made with the aid of
diagrams from Lehman et al. (32).
Analysis of data
For Exp 1, data across treatments were analyzed by two-way ANOVA
with repeated measures for one factor (drug). As pulses of LH were not
detectable in some T-treated animals, the across-treatment analysis was
confined to changes in mean LH and FSH concentrations. This analysis
was based on the 
, i.e. change in values during the
control or first half of each infusion vs. those during the
experimental or last 3 h of the second half of infusion. Thus,
(drug period minus control period) in response to aCSF-aCSF was
compared with that in response to aCSF-baclofen across both T and no T
groups. Analysis of the effects of baclofen on pulse parameters within
the castrated (no T) group was performed using one-way ANOVA with
repeated measures. For Exp 2, analysis of interpulse interval could not
be performed due to absence of pulses during aCSF infusion. Thus,
analysis of the effect of baclofen on GnRH pulse frequency was
performed using the nonparametric Wilcoxon sign-rank test. Analysis of
effects of baclofen on mean GnRH and LH was performed using one-way
ANOVA for repeated measures after transformation of the data following
tests for homogeneity of variance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) either in or near the ventromedial
nuclei.
|
.
The main effect of baclofen compared with that of aCSF was highly
significant (P < 0.01). The interaction between
steroid and baclofen was nonsignificant, i.e. baclofen
increased mean LH concentrations in both castrated and castrated
T-treated animals (Fig. 3). Notably, in
some cases (e.g. Fig. 2, sheep 9904 and 9906), the rise of
LH in T-treated animals occurred as an apparent increase in basal
concentrations unaccompanied by the appearance of distinct LH
pulses.
|
|
Although T reduced mean FSH concentrations (10.31 ± 1.39 vs. 3.99 ± 1.27 ng/ml; P < 0.01), baclofen had no significant effect on FSH in either castrated or T-treated animals (data not shown).
GnRH in intact rams
This experiment was conducted to determine whether the
baclofen-induced increases in LH concentrations were accompanied by
increased GnRH secretion. Representative results obtained from four of
seven rams are shown in Fig. 4. In all
animals no or few GnRH and LH pulses were detected before baclofen
infusion. However, baclofen treatment induced robust GnRH pulses within
a few minutes in six of the seven animals. Overall, mean GnRH pulse
frequency changed from 0.10 ± 0.06 to 1.57 ± 0.48/h
(P < 0.03) during baclofen infusion. LH pulses first
appeared either at the same time as GnRH pulses (Fig. 4, sheep 9916) or
only after a variable delay (Fig. 4, sheep 9915). Subsequently, GnRH
and LH pulses were wholly coincident, although in cases where GnRH
pulses were very frequent, the LH pulses became obscured. Baclofen
treatment increased the mean GnRH concentration from 3.41 ± 0.63
to 8.50 ± 2.92 pg/ml (P < 0.05) and the mean LH
concentration from 0.79 ± 0.09 to 5.00 ± 2.03 ng/ml
(P < 0.01).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The noted effect of baclofen in castrated animals confirms previous observations (17) of increased pulse amplitude without an effect on pulse frequency. A similar effect was observed in some (sheep 9914), but not all, T-treated animals. Another response pattern in T-treated animals was a clear increase in mean circulating LH, unaccompanied by LH pulses. An explanation for this nonpulsatile release was not readily apparent; thus, it was important to determine the effect of baclofen on GnRH release. This was accomplished in Exp 2, albeit in intact males. Although there are limitations in making inferences from data obtained in the intact male to the other two models, the results support the conclusion that the effects of baclofen on LH are mediated entirely or wholly through increased secretion of GnRH.
These results of Exp 2 are useful for several reasons. First, they show that the effect of baclofen on GnRH secretion is rapid and robust. This observation suggests the availability of a readily releasable pool of GnRH and that baclofen was acting relatively close to the infusion site. Second, they help explain the nonpulsatile release of LH in response to baclofen shown by some T-treated animals in Exp 1. In some animals (e.g. sheep 9915), the anterior pituitary of either T-treated or intact males was not readily responsive to the initial pulse(s) of GnRH, but became so after repeated priming pulses. Whether this initial unresponsiveness was due to secondary effects of prolonged suppression of GnRH by T, direct effects of T on the pituitary, or both is not clear. However, the results do demonstrate that a full response can be achieved relatively rapidly and again illustrate the well known priming action of GnRH.
A third aspect of baclofen’s action in the intact male is that in some animals there was a clear increase in both the frequency and the amplitude of measurable GnRH pulses. Although the increase in pulse frequency may simply have reflected an increase in amplitude, it seems unlikely for at least two reasons. The first is that T decreases GnRH pulse frequency, with little effect on amplitude (25). The second is that GnRH pulse frequency in some intact males during baclofen treatment approached that in castrated males (1.6 vs. 1.7 pulses/h). Accordingly, the data can be interpreted as suggesting that GABAB receptors also regulate the pulse generator. In castrated animals that effect may not be apparent because the pulses already are occurring at a near-maximal rate. In contrast, in the intact or T-treated animal this effect of baclofen is detectable because T suppresses the pulse rate, and baclofen reverses this action of T. These observations also indicate that baclofen probably acts downstream from the site of action of T.
The specific sites and mechanisms by which baclofen exerts these effects is unknown. A priori there are at least three possible general modes of action: direct stimulation of the GnRH neuron, direct disinhibition of stimulatory interneurons, and inhibition of inhibitory interneurons. The first two seem relatively unlikely. Although baclofen alters the firing rate of GT-1 cells in vitro, the effect is inhibitory (33). There are several potential stimulatory systems; however, the net effect of baclofen is to reduce neuronal activity (23) by either direct inhibition or disinhibition of GABA-secreting neurons. Thus, it appears improbable that stimulatory systems are directly activated by baclofen, although there is evidence that activation of GAGAB receptors facilitates activation of adenyl cyclase by neurotransmitters such as norepinephrine (34). That action potentially could increase the effect of a stimulatory neurotransmitter.
On the other hand, the wide distribution of GABAB receptors and the presence of several putative inhibitory neuronal systems that affect GnRH secretion lend credence to the third possibility. The anatomical distribution of GABAB receptors within the brain has not been described in the sheep, but in the rat their wide distribution includes several hypothalamic nuclei (35, 36, 37, 38, 39). GABAB receptors are located on cell bodies, axons, and dendrites (40); thus, they have the potential for both pre- and postsynaptic effects. Although two morphological studies found a relative paucity of GABAB receptor protein on GABA-synthesizing neurons (37, 38), many physiological data indicate that presynaptic GABAB receptors act as autoreceptors to suppress GABA release (41, 42). Of equal or greater interest is the clear presence of heterologous GABAB receptors on non-GABAergic neurons, e.g. glutaminergic or dopaminergic (40, 41). Thus, GABAB receptors may have an important role in regulating the secretion of several neurotransmitters or neuromodulators. Although critical double labeling studies have not been reported, there is support for the concept that activation of GABAB receptors affects the release or action of a variety of neurotransmitter systems, including the glutaminergic and dopaminergic systems (43, 44, 45, 46, 47, 48). For example, baclofen reduced the release of dopamine from striatal slice preparations and other experimental systems (45, 46) and inhibited the effect of heroin on self-administrative behavior (49).
Many data implicate the GABAergic, dopaminergic, and opiatergic systems as exerting an inhibitory influence on GnRH secretion in sheep and other species (17, 18, 19, 50, 51, 52); thus, each is a potential target for the action of baclofen. Inhibition of any of these systems, either directly or indirectly, has the potential to increase GnRH and LH release.
The failure of baclofen treatment, and subsequently increased GnRH secretion, to effect equally rapid increases in FSH and LH concentrations is consistent with previous observations made in the ram, namely, that postcastration LH concentrations rise within a few hours, whereas the rise of FSH starts several days later (53) and that administration of exogenous GnRH increases LH much more rapidly than it does FSH (54). Our results simply underscore, and clearly demonstrate, the relative unresponsiveness of the FSH secretory system to GnRH.
Finally, it is not yet obvious how these observations relate to the normal regulation of GnRH pulse frequency and amplitude. The most parsimonious interpretation would be that the GABAergic system normally constrains the GnRH secretory system by acting at distinct sites to regulate both frequency and amplitude and that localized activation of GABAB receptors is an integral part of a multineuronal control system. However, the available data are neither sufficiently consistent nor complete to provide a clear understanding of these relationships.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received April 24, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß messenger ribonucleic acids in adult
gonadotropin-releasing hormone neurons. Endocrinology 140:51951 subunit is expressed by
distinct neuronal populations. Mol Brain Res 15:121[Medline]
-aminobutyric acid receptor subtypes in the arcuate-ventromedial
region of the castrated ram. Endocrinology 137:3453[Abstract]
-aminobutyric acid (GABA) is mediated by GABA-A receptors but not
GABAB receptors. Endocrinology 132:1789[Abstract]
-aminobutyric acid turnover: effect of castration on
-aminobutyric acid turnover in microdissected brain regions of the
male rat. J Neurochem 60:2254
-aminobutyric acid may
be involved in seasonal changes in the negative-feedback effects of
estrogen on gonadotropin-releasing hormone secretion and plasma
luteinizing hormone levels in the ewe. Endocrinology 133:2904[Abstract]
-aminobutyric acid type B
receptor 1 and 1ß mRNA variants in GABA and non-GABAergic neurons
in the rat brain. J Comp Neurol 416:475[CrossRef][Medline]
-aminobutyric acid subtype
1 receptor mRNA in the rat brain. J Comp Neurol 407:166[CrossRef][Medline]
-hydroxybutyric acid (GABA) are specifically
induced by activation of GABAB receptors. Naunyn
Schmiedberg Arch Pharmacol 357:11
This article has been cited by other articles:
 |
|
 |
|
  A. L. Bogusz, S. L. Hardy, M. N. Lehman, J. M. Connors, S. M. Hileman, J. H. Sliwowska, H. J. Billings, C. J. McManus, M. Valent, S. R. Singh, et al. Evidence that {gamma}-Aminobutyric Acid Is Part of the Neural Circuit Mediating Estradiol Negative Feedback in Anestrous Ewes Endocrinology, June 1, 2008; 149(6): 2762 - 2772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Jackson and D. Kuehl Effects of Applying Gamma-Aminobutyric AcidB Drugs into the Medial Basal Hypothalamus on Basal Luteinizing Hormone Concentrations and on Luteinizing Hormone Surges in the Female Sheep Biol Reprod, February 1, 2004; 70(2): 334 - 339. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. T. Jansen, C. Cutter, S. Hardy, M. N. Lehman, and R. L. Goodman Seasonal Plasticity within the Gonadotropin-Releasing Hormone (GnRH) System of the Ewe: Changes in Identified GnRH Inputs and Glial Association Endocrinology, August 1, 2003; 144(8): 3663 - 3676. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. L. Jackson and D. Kuehl The GABAB Antagonist CGP 52432 Attenuates the Stimulatory Effect of the GABAB Agonist SKF 97541 on Luteinizing Hormone Secretion in the Male Sheep Experimental Biology and Medicine, May 1, 2002; 227(5): 315 - 320. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-K. Han, I. M. Abraham, and A. E. Herbison Effect of GABA on GnRH Neurons Switches from Depolarization to Hyperpolarization at Puberty in the Female Mouse Endocrinology, April 1, 2002; 143(4): 1459 - 1466. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Terasawa and D. L. Fernandez Neurobiological Mechanisms of the Onset of Puberty in Primates Endocr. Rev., February 1, 2001; 22(1): 111 - 151. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
,
Location in castrated animals; •, location in T-treated animals. arc,
Arcuate nucleus; cp, cerebral peduncle; dm, dorsomedial nucleus; III,
third ventricle; fx, fornix; me, median eminence; mt, mammillothalamic
tract; ot, optic tract; vm, ventromedial nucleus.
) and castrated T-treated (n = 7;
) male sheep. Each bar represents the mean ±
SEM. The main effect of baclofen was significantly
different (P < 0.01) from that of aCSF, and there
was no interaction between steroid and drug.