| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Laboratory of Neuroendocrinology, The Babraham Institute, Cambridge, United Kingdom CB2 4AT; and Institut National de la Recherche Agronomique, Physiologie de la Reproduction des Mammiferes Domestiques (A.C., B.M.), Nouzilly 37380, France
Address all correspondence and requests for reprints to: Dr. Donal C. Skinner, Station de Physiologie de la Reproduction, Nouzilly 37380, France. E-mail: skinner{at}tours.inra.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The applicability of the ventricular approach, however, relies on an assumption: that CSF GnRH reliably transmits information about secretory events occurring within the hypophyseal portal system. Thus, although the release of GnRH into the CSF shows both dynamic and longer term changes that appear to be related to LH secretion, its relationship with hypophyseal portal secretory events remains speculative. In this respect, portal GnRH pulses could occur without detectable CSF GnRH pulses. Moreover, the CSF GnRH surge could decrease after or before the portal GnRH surge, or the CSF GnRH surge could begin later than the portal surge and thus be an unreliable temporal indicator of neuroendocrine events at the hypothalamic level. A further caveat that remains to be addressed is which of the two approaches provides the most accurate estimate of GnRH input to the pituitary. In this respect, previous studies have suggested that most, but not all, LH pulses are coupled with detectable portal GnRH pulses (6). This may result from only a few portal vessels being lesioned during an experiment; thus, portal GnRH values are only a measure of the decapeptide in these vessels and not of the complete vasculature supply to the pituitary gland (7). On the other hand, GnRH in CSF is a measure of the total release into this compartment and may reflect release from areas other than just the median eminence. In this respect, it is plausible to speculate that the organum vasculosum of the lamina terminalis, which surrounds the anterior region of the third ventricle and has an abundance of GnRH neuronal perikaryia and axons (8), could contribute to the CSF pool of GnRH.
The purpose of this study, therefore, was to characterize simultaneously the release of GnRH into the hypophyseal portal blood and CSF of the third ventricle and to determine the temporal relationship between the GnRH patterns in these two compartments and their relationship to peripheral LH secretion.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Surgery
CSF surgery was performed first because the portal collection
apparatus obscures the third ventricle in x-rays. Accordingly, guide
cannulas (17 gauge, 50 mm, stainless steel luer-lock needle) were
introduced stereotaxically into the third ventricle using a slight
modification of a method described previously (1). Briefly, under
halothane anesthesia, the ewe’s head was positioned in a stereotaxic
frame, and 1 ml radioopaque liquid was injected into the right lateral
ventricle. The cannula tip was positioned 1 mm in front of the vertical
and 1 mm below the horizontal tangents of the intermediate mass;
correct placement was verified by the flow of CSF and additional
x-rays. The cannula was plugged and fixed in place with acrylic dental
cement, and a stainless-steel helmet was placed around it for
protection. Before portal surgery, all ewes were checked to ensure that
CSF flowed freely back up the third ventricular guide cannulas. The
portal blood collection apparatus was then implanted 2 weeks before an
experiment using a procedure described in detail previously (7). Both
surgeries were initially conducted under one anesthesia, but after
problems during sampling we subsequently allowed at least 1 month to
elapse between operations.
Experimental approach
A LH surge was induced in all ewes using a well characterized
follicular phase model in which a 10-mm SILASTIC brand (Dow Corning,
Midland, MI) 17ß-estradiol implant is inserted sc just after
ovariectomy (9). An intravaginal progesterone implant (CIDR, InterAg,
Hamilton, New Zealand) was inserted for 10 days and then removed to
simulate luteolysis. Four 30-mm estradiol implants were inserted sc
24 h later. For the pulse part of this study, integrated CSF
samples and portal and jugular blood samples (10 min) were collected
for the 6 h preceding insertion of the four estradiol implants.
For the surge study, instantaneous jugular blood and integrated CSF and
portal blood samples were collected hourly for 24–36 h, starting
10–12 h after insertion of the four estradiol implants.
Portal blood flow was established and maintained as described previously (7). A SILASTIC catheter was then inserted through the third ventricular guide cannula so that the distal end projected 3 mm beyond the tip of the cannula, and CSF was collected (30 µl/min) directly into methanol. GnRH concentrations in CSF and portal blood were estimated using the RIA and primary antibody (BDS-037) described by Caraty et al. (10). GnRH for radioiodination and standards was purchased from Peninsular Laboratories Europe (St. Helens, UK). All CSF and portal samples from an individual ewe were measured in duplicate in the same assay. The intraassay coefficient of variation and assay sensitivity averaged 14% and 0.2 pg/tube (portal, 0.6 pg/ml; CSF, 1.5 pg/ml), respectively (five assays).
Concentrations of LH were estimated in duplicate 100-µl jugular plasma samples using the RIA method of Niswender et al. (11), and all samples from an individual ewe were measured in the same assay. The primary antiserum was purchased from Dr. G. D. Niswender (CSU204, Colorado State University, Fort Collins, CO), iodination grade ovine LH was obtained from Biogenesis (Bournemouth, UK), and NIH LH standard S11 was used for reference preparations. The intraassay coefficient of variation averaged less than 10%, and assay sensitivity was 0.2 ng/ml (five assays).
Analysis
The onsets of the LH and GnRH surges were defined as the first LH
or GnRH sample to exceed the presurge baseline by 2 SD of
this baseline, after which hormone concentrations did not return to
baseline levels within 2 h. The presurge baseline and
SD were calculated from the samples collected for the first
4 h of the experiment. The amplitude of the surge was taken as the
peak level after estradiol insertion minus the presurge baseline. As
hormone concentrations were still above the presurge baseline at the
end of the experiment, to obtain an estimate of the durations of the LH
and GnRH surges, the period during which samples exceeded the
half-maximal concentration was calculated. Data were analyzed by
Student’s paired t test and are presented as the mean
± SEM.
Pulses were detected using Munro software (12), as described previously (1). Using this software, the number of pulses and the mean pulse amplitude were calculated for each ewe. Pulse duration was taken as the period during which concentrations exceeded the predicted baseline for that pulse as calculated by Munro. CSF GnRH and LH pulses were regarded as causally related when one or more fractions of a CSF GnRH pulse overlapped one or more samples from a LH pulse. Comparisons between parameters were made using Student’s t test for paired data, and results are presented as the mean ± SEM. In addition, to determine whether there was any relationship among portal GnRH, CSF GnRH, and jugular LH pulses, correlations were determined among these variables using all detected pulses.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Estrogen-induced surges
A GnRH surge was detectable in both the portal and CSF
compartments in five sheep (Fig. 1
). The
time of the onset of the portal GnRH surge (15.3 ± 0.5 h
after estradiol insertion) occurred simultaneously with the onset of
the CSF GnRH (15.9 ± 0.2 h) and jugular LH surges (15.6
± 0.4 h). GnRH concentrations rose rapidly during the surge in
both portal blood and CSF to maximum values (n = 4) of 86.9
± 48.0 and 61.1 ± 23.7 pg/ml, respectively. This represented
46 ± 18-fold (portal) and 23 ± 12-fold (CSF) increases
above basal levels. There was no correlation between portal and CSF
GnRH levels between ewes, but levels were correlated (r2 =
0.77) within ewes. GnRH concentrations were above half-maximal levels
for longer (portal, 7.3 ± 1.4 h; CSF, 7.3 ± 0.4
h) than LH surge levels (3.3 ± 0.4 h).
|
). A total of 20 LH (0.83 ± 0.07
pulses/6 h), 18 portal GnRH (0.75 ± 0.08 pulses/6 h), and 19
(0.79 ± 0.04 pulses/6 h) CSF GnRH pulses were detected in this
study. In one ewe (Fig. 2, ewe 86), LH and CSF GnRH pulses were
detected without a corresponding portal GnRH pulse, although there was
a notable, but slight, increase in portal GnRH at this time; when the
portal blood of this ewe was assayed for LH, a clear small pulse was
detectable (data not shown). Two CSF GnRH pulses were not related to
any portal GnRH or jugular LH pulse and were excluded from analysis of
the temporal characteristics (Fig. 2, ewe I1). Two portal GnRH and
jugular LH pulses had no corresponding CSF GnRH pulse. A LH pulse was
detected at the start of the experiment (Fig. 2; ewe 86) in one ewe
and, as it was likely that the CSF and portal GnRH pulses had occurred
before sample collection began, this pulse was excluded from
analysis.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present study establishes unequivocally that the measurement of GnRH in CSF is an accurate marker of hypothalamic GnRH secretion at the time of the preovulatory LH surge. Not only were both the onset and the duration of the CSF and portal GnRH surges similar but, during a surge, the pattern of GnRH released into the portal blood was similar to that in the CSF. There was, however, no correlation between the absolute concentrations of GnRH in CSF and portal blood between ewes; CSF concentrations were higher than portal levels in two ewes, lower in two, and similar in one. For CSF, the concentration of GnRH is probably dependent on the precise location within the third ventricle of the SILASTIC catheter, as suggested in the monkey (4). In portal blood, the GnRH concentration may be dependent on the site and extent of the lesion (13).
Whether the CSF approach is suitable for investigating GnRH pulsatility is more debatable. In portal blood, 94% of the GnRH pulses were associated with LH pulses, and there was a strong correlation between the amplitudes of portal GnRH pulses and their LH counterparts. In contrast, the relationship between CSF GnRH and LH pulses was less precise (79%) than that found in an earlier study (87%) on ovariectomized ewes (1). This may have been a result of the smaller LH pulses (amplitude, 2.8 vs. 5.4 ng/ml) detected in the present study, which was probably due to both the 10-mm estradiol implant and the collection of portal blood that would have removed some of the GnRH destined for the pituitary gland (14). Both of these factors may also have resulted in smaller CSF GnRH pulses detected in this study than reported previously (amplitude, 4.1 vs. 6.3 pg/ml); if some CSF GnRH originates from the portal vasculature through retrograde blood flow (15), then the collection of portal blood would reduce the amount of GnRH entering the third ventricle. A further noteworthy point is the difference in the duration of the portal and CSF GnRH pulses. Portal CSF GnRH pulses rarely exceeded one collection sample, thereby giving a discrete and exact signal. In contrast, CSF GnRH pulses extended over several samples, and the mean duration was identical to that reported previously (1). As the volume of portal blood collected during an experiment is not restrictive, samples can be collected frequently (every 30 sec) (16), providing more dynamic information. In this respect, the present study probably overestimates the duration of portal GnRH pulses, as other studies have shown that these range from 2–10 min long (16, 17). As the volume of CSF samples during pulse analysis is low (300 µl/10-min sample), it will not be possible to significantly enhance the resolution of CSF GnRH pulses. The reliability of CSF GnRH data may also be questionable in high frequency or low pulse amplitude conditions when CSF GnRH pulses are likely to merge. Indeed, this potential problem is evident in the present study, when pulses of GnRH occurred before all GnRH from the previous pulse had cleared from the CSF. In high pulse frequency and/or low pulse amplitude conditions, detection of CSF GnRH pulses would be dependent on the pulse amplitude; low amplitude pulses would be more likely to evade detection and may have been the reason two portal GnRH and LH pulses did not have corresponding CSF GnRH pulses in this study. The detection of two CSF GnRH pulses without corresponding portal GnRH or jugular LH pulses is also noteworthy, and a possible nonmedian eminence origin (e.g. organum vasculosum of the lamina terminalis) of these pulses cannot be excluded. Indeed, the source of CSF GnRH has not been established, although our study suggests that most CSF GnRH is probably of median eminence origin. Presumably some GnRH is released from terminals in the median eminence during a secretory episode diffuses or is transported by tanycytes (18) into the third ventricle to produce the characteristic broad pulses. If, on the other hand, GnRH is released directly into the ventricles from axon terminals, then the broad CSF GnRH pulses may be due to slower clearance from and recirculation within the third ventricle. In contrast, portal blood passes through the median eminence region rapidly, and collected blood samples reflect GnRH levels during their one and only pass down the portal vessels (7).
The CSF approach shall enable further inroads to be made in establishing how neurotransmitters and steroids control the preovulatory release of GnRH that were not possible with the portal method for three main reasons. First, the finding that blood was present in the CSF of three ewes despite the care taken to make no lesion in the third ventricle upholds the view that even the smallest cerebral lesion introduced by a microdialysis probe or push-pull apparatus during a combined study with portal cannulation might cause considerable local hemorrhaging and undermine the results obtained. As no heparin is used during CSF collection, this problem is resolved. Second, no lesion is made, and thus, animals may be reused repeatedly and serve as their own controls in long term experiments. This advantage is reinforced by the fact that the site of CSF collection within the third ventricle remains constant between experiments. To date, CSF has been collected on more than five separate occasions from individuals over 2 yr. Although it is possible to collect portal blood from individuals on more than one occasion in short term experiments, levels of GnRH are invariably lower during the second collection, possibly due to necrosis and collapse of portal vessels resulting from the previous lesion (7). Third, samples can be collected for an indefinite period during each experiment; collections have been made for over 50 h without evidence of discomfort. In contrast, during long term portal blood collections, the extensive heparinization may cause bleeding at the surgical site and even internal hemorrhaging (7).
Improvements in the sensitivity of the analysis of CSF samples (e.g. improving GnRH extraction) may make pulsatile CSF GnRH analysis more precise, but apart from marking the occurrence of a GnRH pulse, it is unlikely that we will ever be able to use this method to gain further insights into the dynamics of GnRH secretion, such as the shape of a GnRH pulse (17). Thus, if heparinization of the ewes is not a complication for a particular experiment, then it is clear that due to the precise relationship between portal GnRH and LH pulses and the lack of such precision between CSF GnRH and LH pulses, the portal method should be the approach of choice when investigating the rapid events occurring at the time of a LH pulse. However, if heparinization is a limiting factor, then the CSF GnRH approach should be used, but the shortcomings of this method must be considered, and where possible, a high amplitude LH pulse model should be employed.
What, if any, is the physiological function of CSF GnRH? One hypothesis is that CSF GnRH may regulate gonadotropin secretion through an inhibitory ultrashort loop feedback system to modulate the activity of the GnRH pulse generator (19, 20). If CSF GnRH was part of such an ultrashort loop feedback system during the preovulatory surge, then we might expect the CSF GnRH surge to terminate before or after the portal GnRH surge. It is clear from the present study that portal and CSF GnRH surges are coincident. It is also unlikely that such a mechanism operates during a pulse, as CSF GnRH pulses tended to peak after their portal counterparts. Alternatively, CSF GnRH could stimulate the pituitary by passing from the ventricles to the hypophysial portal system, possibly via tanycytic transport (18, 21, 22). As portal GnRH pulses were abrupt discrete episodes, whereas the coincident CSF GnRH pulses were substantially longer, the present study suggests strongly that CSF GnRH does not have a pituitary action at the time of a pulse; whether CSF contributes to portal GnRH levels during the surge cannot be assessed with the current data. A third hypothesis is that GnRH in the CSF at the time of the preovulatory surge may play an important role in the regulation of reproductive behavior (23, 24). As GnRH has been found to influence sexual behavior in a wide range of species from monkeys (25) and horses (26) to voles (27) and, more recently, in some nonmammalian species (28), this hypothesis warrants serious consideration.
In summary, these data demonstrate that CSF GnRH concentrations provide an accurate index of GnRH neurosecretory events that occur within the hypophyseal portal system at the time of the preovulatory LH surge. Furthermore, in the present steroidal environment, in which LH pulses were easy to define, there was a good, but not perfect, relationship between CSF GnRH pulses and portal GnRH and LH pulses; when pulse frequency increases, this relationship is likely to be impaired. Measurement of GnRH in CSF may be used, therefore, under certain circumstances as an index of portal GnRH secretion and, when used in conjunction with other techniques, will allow advances to be made in our understanding of how the neuroendocrine reproductive axis is controlled.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received June 5, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Caraty and D. C. Skinner Gonadotropin-Releasing Hormone in Third Ventricular Cerebrospinal Fluid: Endogenous Distribution and Exogenous Uptake Endocrinology, October 1, 2008; 149(10): 5227 - 5234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Ruf, M.-J. Park, F. Hayot, G. Lin, B. Roysam, Y. Ge, and S. C. Sealfon Mixed Analog/Digital Gonadotrope Biosynthetic Response to Gonadotropin-releasing Hormone J. Biol. Chem., October 13, 2006; 281(41): 30967 - 30978. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. Foradori, R. L. Goodman, V. L. Adams, M. Valent, and M. N. Lehman Progesterone Increases Dynorphin A Concentrations in Cerebrospinal Fluid and Preprodynorphin Messenger Ribonucleic Acid Levels in a Subset of Dynorphin Neurons in the Sheep Endocrinology, April 1, 2005; 146(4): 1835 - 1842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Wu, S. K. Mani, M. J. Glucksman, and J. L. Roberts Stimulation of Luteinizing Hormone-Releasing Hormone (LHRH) Gene Expression in GT1-7 Cells by Its Metabolite, LHRH-(1-5) Endocrinology, January 1, 2005; 146(1): 280 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.A. Zieba, M. Amstalden, S. Morton, M.N. Maciel, D.H. Keisler, and G.L. Williams Regulatory Roles of Leptin at the Hypothalamic-Hypophyseal Axis Before and after Sexual Maturation in Cattle Biol Reprod, September 1, 2004; 71(3): 804 - 812. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Amstalden, D.A. Zieba, J.F. Edwards, P.G. Harms, T.H. Welsh Jr., R.L. Stanko, and G.L. Williams Leptin Acts at the Bovine Adenohypophysis to Enhance Basal and Gonadotropin-Releasing Hormone-Mediated Release of Luteinizing Hormone: Differential Effects Are Dependent upon Nutritional History Biol Reprod, November 1, 2003; 69(5): 1539 - 1544. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Yoshioka, C. Suzuki, S. Arai, S. Iwamura, and H. Hirose Gonadotropin-Releasing Hormone in Third Ventricular Cerebrospinal Fluid of the Heifer During the Estrous Cycle Biol Reprod, February 1, 2001; 64(2): 563 - 570. [Abstract] [Full Text]
|
|
|
|
|
|
D. C. Skinner, T. G. Harris, and N. P. Evans Duration and Amplitude of the Luteal Phase Progesterone Increment Times the Estradiol-Induced Luteinizing Hormone Surge in Ewes Biol Reprod, October 1, 2000; 63(4): 1135 - 1142. [Abstract] [Full Text]
|
|
|
|
|
|
D. R. Mackey, J. M. Sreenan, J. F. Roche, and M. G. Diskin Effect of Acute Nutritional Restriction on Incidence of Anovulation and Periovulatory Estradiol and Gonadotropin Concentrations in Beef Heifers Biol Reprod, December 1, 1999; 61(6): 1601 - 1607. [Abstract] [Full Text]
|
|
|
|
|
|
A. Caraty and D. C. Skinner Progesterone Priming Is Essential for the Full Expression of the Positive Feedback Effect of Estradiol in Inducing the Preovulatory Gonadotropin-Releasing Hormone Surge in the Ewe Endocrinology, January 1, 1999; 140(1): 165 - 170. [Abstract] [Full Text]
|
|
|
|
|
|
O.S. Gazal, L.S. Leshin, R.L. Stanko, M.G. Thomas, D.H. Keisler, L.L. Anderson, and G.L. Williams Gonadotropin-Releasing Hormone Secretion into Third-Ventricle Cerebrospinal Fluid of Cattle: Correspondence with the Tonic and Surge Release of Luteinizing Hormone and Its Tonic Inhibition by Suckling and Neuropeptide Y Biol Reprod, July 1, 1998; 59(3): 676 - 683. [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 |
