BIOLOGY OF REPRODUCTION 57, 107-111 (1997)
The Estradiol-lnduced Luteinizing Hormone Surge in the Ewe Is Not Associated with Increased Gonadotropin-Releasing Hormone Messenger Ribonucleic Acid Levels' Harveen Dhillon, Anita M. Dunn, Eduardo Esquivel, Debora L. Hamernik, and Mark E. Wise 2 Departments of Animal Sciences and Physiology, University of Arizona, Tucson, Arizona 85724 ABSTRACT This experiment was undertaken to determine whether the estrogen-induced LH and GnRH surge in the ewe is associated with activation of a specific subpopulation of neurons in the mid-brain of the ewe as indicated by a change in GnRH mRNA levels. Fifteen ovariectomized ewes were assigned to treatment groups 3-4 wk after ovariectomy. One group of ewes served as controls (n = 2); 50 t±g estradiol-171 (E2) was administered to the remaining ewes. Blood samples were collected from all ewes before treatment (2-h period at 10-min intervals) and continued at 30-min intervals until tissue was collected. At 6, 12, 18, and 24 h after E2 (n = 3 for each time point), brains were collected and processed for localization and measurement of GnRH mRNA by in situ hybridization histochemistry. Serum was analyzed for LH concentrations. Serum LH was pulsatile in controls and decreased at 6 h after E2, and by 12 h the LH surge was initiated. LH levels peaked at 18 h after E2 and returned to basal levels 24 h after E2 treatment. A cRNA probe corresponding to the GnRH-associated peptide region of ovine GnRH prepropeptide mRNA was used to identify GnRH mRNA. Associated with the onset and peak of the LH surge were decreased levels (p < 0.1) of GnRH mRNA in neurons of the preoptic area (POA). Neither the number nor mRNA content of GnRH neurons in the diagonal band of Broca, septal area, or medial basal hypothalamus (MBH) changed during the LH surge. In contrast to E2induced increases in GnRH secretion during the LH surge, our data indicate that E2 decreases steady-state amounts of GnRH mRNA and that GnRH neurons in the POA are influenced to the greatest extent during the E2-induced GnRH surge. INTRODUCTION In ovariectomized (OVX) ewes, an injection of estradiol17 (E2) initially inhibits tonic secretion of LH and then induces a large release of LH similar to the preovulatory surge seen in intact ewes [1]. This biphasic pattern of LH secretion, in which a decrease is followed by a preovulatory-like surge (positive feedback), is also seen in GnRH secretion into the hypophysial portal blood of OVX ewes [2]. The stimulatory action of E2 on LH secretion is due to a large surge in the secretion of GnRH which appears to be caused by an increase in GnRH pulse frequency. Unlike the pulsatile mode of GnRH secretion, which is more susceptible to steroid-induced suppression during the anestrous season, the preovulatory-like surge of GnRH induced by E2 does not vary with season [3]. Caraty et al. [2] have suggested the presence of two populations of GnRH neurons: one population that controls the preovulatory surge of GnRH and LH, and another population that controls tonic secretion of GnRH and LH. VarAccepted February 20, 1997.
Received September 23, 1996. 'Supported by USDA 9337203. Correspondence: Mark E.Wise, Department of Animal Sciences, 228 Shantz, University of Arizona, Tucson, AZ 85721-0038. FAX: (520) 6212
9435; e-mail:
[email protected]
107
ious studies done in the rat, hamster, and rabbit have shown the presence of morphologically distinct tonic and surge groups of GnRH neurons [4-7]. There is increasing evidence that steady-state levels of GnRH mRNA can be used as an index for the rate of biosynthesis and release of neuropeptides [8]. With respect to GnRH, levels of GnRH prepropeptide are usually correlated with the cellular levels of GnRH mRNA [9], suggesting that steady-state levels of GnRH mRNA reflect rates of GnRH biosynthesis and degradation. To examine the relationship between GnRH mRNA levels and GnRH secretion, we used the model of OVX ewes treated with E2. In the OVX ewes, a surge of GnRH and LH is consistently induced within 12-18 h after administration of E2 [2]. To determine whether a specific population of GnRH neurons were influenced and their GnRH mRNA levels altered during the GnRH surge, in situ hybridization histochemistry was used. Our objective was to examine the relationship between steady-state amounts of GnRH mRNA and GnRH secretion (reflected as changes in LH secretion) and to determine whether a specific population of GnRH neurons was influenced indirectly by E2 during the preovulatory GnRH surge. MATERIALS AND METHODS Animals and Treatments Fifteen western range ewes were ovariectomized during the breeding season. Three to four weeks after ovariectomy, ewes were cannulated in one external jugular vein with polyethylene tubing (Intramedic; Clay Adams, Parsippany, NJ). Ewes were divided randomly into treatment groups that consisted of animals given no E2 (controls), or those given E2 (Sigma Chemical Co., St. Louis, MO). Estrogen treatment consisted of simultaneous injections administered both i.m. and i.v. as a bolus at a dosage of 25 g i.v. and 25 g i.m. This regimen of E2 has previously been shown to induce a synchronized surge of GnRH in ovariectomized ewes [2]. Blood samples were collected from all ewes at 10-min intervals for 2 h to establish pretreatment LH concentrations. In ewes receiving E2 injections, blood samples were taken at 30-min intervals after E 2 until ewes were killed. Ewes were divided into five groups on the basis of the time exposed to E2 (i.e., zero hour = control, 6, 12, 18, and 24 h post-E 2 treatment; n = 3/time point). Tissue from one control was damaged during the freezing process and not used for in situ hybridization. After the final blood sample was taken, ewes were anesthetized with sodium pentobarbital and exsanguinated. Brains were quickly removed (within 5 min of anesthesia) and halved mid-sagittally. The hypothalamus was excised with boundaries approximately 1 cm anterior to the septal/preoptic area, just posterior to the mammillary body, dorsal to the anterior commissure, and 1.0-1.5 cm lateral to the mid-line. The tissue was cooled to -30 0 C in isopentane and stored at -80°C. All
108
DHILLON ET AL.
procedures were carried out in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the University of Arizona Institutional Animal Care and Use Committee. Serial sections 12 tpm thick were cut using a cryostat (Model 1800 Reichert Jung Cryocut; Cambridge Instruments, Heidelberg, Germany) at -16°C. One hundred to 150 sections were cut per animal from the medial surface extending through the entire block of excised hypothalamic tissue. The tissues were thaw-mounted on gelatin-subbed microscope slides and stored at - 80°C until used for in situ hybridization. Serum concentrations of LH were measured by a previously validated RIA [10, 11] with 125I-labeled NIADDK ovine LH-I-3 as a tracer. The lower limit of detection was 0.2 ng/ml as determined by inhibition of total binding to 90%. The intra- and interassay coefficients of variation were less than 9% and 12%, respectively.
chloride, 0.015 M sodium citrate)/l mM dithiothreitol (DTT) for 15 min at room temperature, incubated for 30 min at 37°C in 20 pug/ml RNase A, and rinsed twice (30 min each) at room temperature in single-strength SSC/l mM DTT, once in 0.5-strength SSC/1 mM DTT, and once in 0.1-strength SSC/1 mM DTT Two 30-min high-stringency washes were then carried out at 420 C in 0.1-strength SSC/1 mM DTT/50% formamide. The slides were cooled to room temperature in 0.1-strength SSC/1 mM DTT, dehydrated through a series of ethanol baths with 300 mM ammonium acetate, dipped in 100% ethanol, and allowed to dry. Visualization. Sections containing hybridized RNA probe were dipped in autoradiographic emulsion (NTB 3 nuclear emulsion; Eastman Kodak, Rochester, NY). Slides were stained in toluidine blue dye and developed using Kodak photographic developer and fixer at 1-wk intervals to optimize exposure time. The optimal exposure time was approximately 5 wk after dipping in emulsion.
Complementary RNA Probe Preparation Nucleotide sequence of a partial ovine hypothalamic GnRH cDNA was determined after reverse transcription and amplification of ovine hypothalamic total RNA by polymerase chain reaction (PCR). The primers used were inferred from rat hypothalamic GnRH cDNA [12]. The initial upstream (5'-GAAGGCTGCTCCAGCCAGCAC-3') and downstream primers (5'-CTTCTTCTGCCCAGCTTCCTC3') represent base pairs (bp) 87-107 and 285-305, respectively, of the rat GnRH cDNA [12]. The resulting PCR product was subcloned into the plasmid pCRII (T/A cloning kit; Invitrogen, San Diego, CA) before sequencing. To ensure maximal hybridization efficiency with ovine GnRH mRNA, homologous primers were constructed, and the template cDNA was liberalized and transcribed in the presence of 3 5S-labeled ao-UTP, T3 RNA polymerase, to yield a 3 5S-labeled antisense RNA probe complementary to a 177-bp fragment corresponding to the GnRH-associated peptide region of GnRH mRNA. A probe using T7 RNA polymerase was generated to use as a sense probe (i.e., negative control). Tissue Processing and In Situ Hybridization Tissue preparation. Every tenth section from the medial face of each tissue block was processed for in situ hybridization. Sections from each animal were processed using methods previously described (as modified from [13]). Briefly, sections were fixed with 4% formaldehyde-PBS, pH 7.4, and rinsed twice in PBS. Fixed sections were treated with 0.25% acetic anhydride in 0.1 M triethanolamine HC1, pH 8.0, and delipidated through a series of ethanol baths and in chloroform. The slides were then left upright to dry at room temperature. Purified water treated with 0.1% diethyl pyrocarbon for 24 h and autoclaved was used for all buffers to minimize RNAase activity. Hybridization. For hybridization, labeled probe (2 x 106 dpm per 100 pl1) was applied to each tissue section in 60 pxl hybridization buffer containing 50% formamide, 300 mM NaCl, 20 mM Tris-HCI (pH 7.5), 1 mM EDTA, 10% dextran sulfate, single-strength Denhardt's solution, 100 pxg/ml salmon sperm DNA, 250 ,ag/ml yeast total DNA, and 250 pig/ml yeast transfer RNA. Sections were covered with parafilm cover slips and allowed to hybridize to radioactive probe overnight at 52°C in humid chambers. After hybridization, the sections were washed in four changes of 4-strength SSC (single-strength SSC = 0.15 M sodium
Tissue Mapping Tissues were systematically mapped for GnRH-positive neurons. A neuron was identified as positive or "labeled" if the density of silver grains localized over the cell soma (verified at 45x brightfield illumination) exceeded five times background. Total number of positive neurons was determined for each tissue section studied per animal (1015/animal). The tissue was divided into various anatomical sections including the septal region, medial basal hypothalamus (MBH), preoptic area (POA), and diagonal band of Broca (dbB). The total number of neurons containing GnRH mRNA was also determined in each region of the hypothalamus. Image Analysis The autoradiographic signal (silver grain density) over individual cells was analyzed on a Macintosh computer using the public domain NIH Image program (written by Wayne Rashband at the US National Institutes of Health and available on the Internet by anonymous ftp from zippy.mimh.nih.gov or on floppy disk from NTIS, 5285, Port Royal Rd., Springfield, VA 22161; part number PB93504868). Integrated density, the sum of gray values of each pixel in the selection with background subtracted, was determined in 5 neurons per region per animal. Data Analysis Mean serum concentrations of LH were determined at each time point (10-min interval before E2 , 30-min intervals after E2). Differences in the number of neurons containing GnRH mRNA observed in each experimental group were analyzed using one-way analysis of variance (ANOVA). The number of neurons containing GnRH mRNA in the POA, septal region, dbB, and MBH were analyzed between treatment groups (time points) using a one-way ANOVA. Total GnRH mRNA containing neurons was determined, and the percentage of the total number of GnRH neurons in each region was calculated for each animal and analyzed by one-way ANOVA. Total grain density was compared for neurons in each brain region using a one-way ANOVA. When significant treatment effects were observed (p < 0.1), differences in means were determined using LSD.
LOCALIZATION OF GnRH NEURONS IN THE EWE
E
I -J
HOURS POST-ESTRADIOL FIG. 1. Changes in serum LH levels in response to E2 treatment. Arrow indicates time of E2 treatment. Samples before E2 treatment taken at 10-min intervals; samples after E2 treatment taken at 30-min intervals. For time 0, 6, 12, 18, and 24 h, n = 14, 12, 9, 6, and 3, respectively.
RESULTS Mean LH release during the experimental period is shown in Figure 1. In the 2 h preceding E 2 administration, mean LH concentrations were variable. Plasma LH concentrations decreased to baseline levels (< 0.5 ng LH/ml serum) after the animals were given E2 and remained low through 6 h after E 2 treatment. At 12 h post-E 2, the plasma LH levels started to rise and remained considerably elevated at 18 h. At 24 h after E2 treatment, the LH-surge subsided, with LH levels decreasing to near preinjection levels. Sequence analysis revealed that the partial ovine hypothalamic GnRH cDNA shared 100%, 78%, 77%, and 40.7% homology with bovine (unpublished results), human [12], rat [12], and salmon [14] GnRH cDNA sequence. The nucleotide sequence of the partial ovine GnRH cDNA determined after PCR amplification and sequencing was as follows: TGGTCCTATGGGCTGCGCCCTGGAGGAAAGAGAAATGCTAAGAACGTGATTGATTCTTTC60 CAAGAGATAGCCAAGGAGGTCGATCAGCCAGTAGAACCTAAGTGCTGTGGGTGCATTGTT 120 CACCAGTCCCATTCTCCTCTGAGGGACCTGAAGGCAGCTCTGGAAAGTCTGATTGAA 177
109
The probe used in these hybridization studies yielded virtually no background signal (Fig. 2). When specific regions of the hypothalamus were examined, the number of neurons containing GnRH mRNA tended to be lower in the POA at the 12-h time point (p < 0.1) than those observed at 0, 6, and 18 h after E2 administration (Fig. 3). The total number of neurons containing GnRH mRNA neurons was not influenced by E2 treatment in any other region. Total numbers of neurons containing GnRH mRNA observed at 0, 6, 12, 18, and 24 h post-E 2 ranged from 31 to 51 for each treatment group. No significant change (p > 0.1) in regional distribution of neurons containing GnRH mRNA was observed over the time course of the experiment. The highest number of neurons containing GnRH mRNA was observed in the POA, followed by the septal area and dbB, and the lowest number was observed in the MBH. The number of neurons seen in the dbB, septal area, and MBH remained constant at all time points examined. There was no significant change in the percentage of neurons containing GnRH mRNA in each region when expressed as a percentage of the total number of neurons containing GnRH mRNA (Fig. 4). No labeled neurons were detected on slides hybridized with the GnRH sense probe. A significant treatment effect (p < 0.08) was observed in the total density of silver grains in neurons located in the POA of the hypothalamus (Fig. 5). Mean comparison tests revealed that the total density declined (p < 0.1) from that of the controls until 18 h after E2 treatment and thereafter increased back to levels similar to those at time 0 at the 24 h time point, coincident with the return of LH to basal levels. The lowest grain density was observed at 18 h after treatment with E2. DISCUSSION The main objective of the current study was to examine and identify changes in GnRH mRNA content of populations of neurons in the mid-brain associated with the estrogen-induced preovulatory-like GnRH surge. In our experiment we observed a decrease in GnRH mRNA (silver grain density) in GnRH neurons in the POA of the hypothalamus FIG. 2. Photomicrograph of neuron containing GnRH mRNA located in the POA. Original magnification x250 (reproduced at 71%).
110 FIG. 3. Changes in number of neurons containing GnRH mRNA seen in various regions of the mid-brain. A significant decrease was observed at 12 h post-E2 treatment in the POA. There were no significant changes in number of neurons containing GnRH mRNA in any other region. Septal, septal area. *Significance at p < 0.1).
DHILLON ET AL.
Z
I
E .O C
So o
dbB
Septal
POA
MBH
Hypothalamic Region during the E 2-induced LH and GnRH surges. Various studies done in the rat, hamster, and rabbit have shown the presence of anatomically distinct tonic and surge cell groups [5-7, 15]. Preliminary evidence from Goodman et al. [16] suggests the possibility of a surge-specific subpopulation of GnRH neurons in the ewe. The neurons present in the POA and the anterior hypothalamus give rise to a true pre-optico-terminalis GnRH tract, which constitutes one of the principal efferent pathways for GnRH cell bodies [17]. The peculiar arrangement of the septico-preoptico-terminalis pathway (i.e., very short with numerous radiating collaterals) appears, from a morphological viewpoint, particularly suitable to mediate acute GnRH discharge. The number of neurons in other regions of the mid-brain examined did not change before, during, or after the E2-induced preovulatory-like surge of GnRH. The only regional changes we observed were in the total number of neurons and the density of silver grains in neurons in the POA. The data support the assumption that the neurons of the POA are the subpopulation of GnRH neurons that are responsive to E and are involved in mediating the surge release of GnRH. The action of E2 on GnRH neurons in the POA would appear to be indirect as these neurons do not contain receptors for E2 [18]. We hypothesized that estrogen induced the GnRH surge by facilitating release of already existing stores of GnRH FIG. 4. Changes in neurons containing GnRH mRNA expressed as percentage of total neurons containing GnRH mRNA present. No significant change was observed in the percentage of total neurons containing GnRH mRNA after E treatment (p> 0.1).
in hypothalamic neurons and increasing the rate of GnRH gene expression to compensate for the elevated level of GnRH secretion. Data from the present experiments indicate that an increase in GnRH mRNA levels does not occur during the GnRH surge. Significant decreases in total density of silver grains, which represents steady-state amounts of GnRH mRNA, occurred in the POA until the end of the LH surge. Previous studies have reported that levels of GnRH peptide are usually correlated with cellular levels of GnRH mRNA [9], suggesting that steady-state levels of GnRH mRNA reflect rates of GnRH biosynthesis and degradation. Our data to some extent agree with this observation as the decrease in total density of silver grains in GnRH mRNA containing neurons at 18 h corresponds to a slight but nonsignificant decrease in GnRH content in the POA at the peak of the GnRH surge [19]. As with mRNA levels in this experiment, the only alteration in GnRH content during the GnRH surge occurs in the POA, with no alterations in the MBH or median eminence [19]. However, mRNA levels for GnRH in the POA and cellular content of the peptide decrease in the face of a massive release of GnRH during this period. Thus, it seems unlikely that during the E2-induced surge of GnRH transcription and/or posttranscriptional processing of GnRH mRNA is coupled to synthesis and release of GnRH. E2 acting indirectly on the GnRH neuron would appear to have two diverse ac-
0
0) Z
C I0o
M
dbB
Septal
POA
Hypothalamic Region
MBH
LOCALIZATION OF GnRH NEURONS IN THE EWE
III
ACKNOWLEDGMENTS -,
The authors are grateful to Dr. Gordon Niswender (Colorado State University, Fort Collins, CO) for providing LH antibody and to the NIH Pituitary Hormone Distribution Program (Baltimore, MD) for RIA reagents.
7.5
0, 5.o 5.0
REFERENCES C 2.5 0 2.5
Septal POA Hypothalamic region FIG. 5. Changes in silver grain density in neurons of the POA and septal area. A significant change was observed (p < 0.1) in the total density in the neurons in the POA. *Significance at p < 0.1).
tions, to stimulate GnRH release and to reduce levels of GnRH mRNA. Our results indicate that the reduction in levels of GnRH mRNA in the POA occurs well before an increase in secretion of GnRH and any change in hypothalamic GNRH content in the POA. The concept that estrogen affects GnRH release independent of positive actions on transcription of the GnRH gene is supported by several observations. At the time GnRH secretion is inhibited by E2 during the first few hours after administration, no increase in GnRH mRNA was observed as would be expected if E2 activated GnRH gene transcription in preparation for the GnRH surge. In fact, we observed a decrease in density of silver grains in GnRH neurons at 6 h after E 2 treatment. The time points used in this study would appear adequate to detect such an increase, since the half-life of GnRH mRNA has been reported to be approximately 22 h in GT1-7 cells in vitro [20]. Additionally, GnRH mRNA levels begin to return to pretreatment levels coincident with the end of the LH (GnRH) surge. Thus, E2 most likely induces the GnRH surge independent of any direct or indirect stimulatory action on the GnRH gene itself. The negative effects of E2 on GnRH mRNA levels in the POA could be due to decreased transcription of the GnRH gene or decreased stability of GnRH mRNA. The above discussion is based on the assumption that LH release is a reflection of GnRH release. Our results show a classical inhibition of LH release for a period of 6 h after estrogen injection, corresponding to the initial inhibitory effect of estrogen on LH release. We observed the beginning of an LH surge between 6 and 12 h and a peak of a preovulatory-like LH surge between 12 and 18 h after E 2 treatment. The surge ended within 24 h after E2 . The E 2 treatment employed in the present experiment was identical to that used by others [2], in which LH and GnRH were both measured. We observed LH profiles similar to theirs; hence we could definitively use the LH levels as an accurate indication of GnRH secretion before the start and to the peak of the LH surge. In summary, the E2 -induced GnRH surge in the OVX ewe was not accompanied by increased levels of GnRH mRNA. These data suggest that neurons containing GnRH mRNA in the POA are influenced by E2 . If neurons in the POA contribute to the preovulatory surge of GnRH, it seems likely that posttranscriptional processes are enhanced to compensate for the increased release of GnRH during the GnRH surge in the ewe.
1. Pelletier J, Signoret JP. Controle de la decharge de LH dans le sang par la progesterone et le benozate d'oestradiol chez la brebis castree. C R Hebd Seances Acad Sci Ser D 1969; 269:2595-2598. 2. Caraty A, Locatelli A, Martin GB. Biphasic response in the secretion of gonadotropin releasing-hormone in ovariectomized ewes injected with estradiol. J Endocrinol 1989; 123:374-382. 3. Moenter SM, Caraty A, Karsch FJ. The estradiol-mnduced surge of gonadotropin-releasing hormone in the ewe. Endocrinology 1990; 127:1375-1384. 4. Rubin BS, King JC. The number and distribution of detectable luteinizing hormone(LH)-releasing hormone cell bodies changes in association with the preovulatory surge in the brains of young but not middle aged female rats. Endocrinology 1994; 134:467-474. 5. Pork-Kaheiskanen T, Urban JH, Turek FW, Levine JE. Gene expression in a sub-population of luteinizing-hormone-releasing hormone(LHRH) neurons prior to the preovulatory gonadotropin surge. J Neurosci 1994; 14:5548-5558. 6. Lmn WW, Ramirez VD. Effect of mating behavior on luteinizing hormone releasing hormone in female rabbits as monitored with pushpull canulae. Neuroendocrinology 1991; 53:229-235. 7. Berriman SJ, Wade GN, Blaustein JD. Expression of Fos-like proteins mingonadotropin-releasing hormone neurons of Syrian hamsters: effects of estrous cycles and metabolic fuels. Endocrinology 1992; 131: 2222-2228. 8. Uhl GR, Nishimori T. Neuropeptide gene expression and neural activity: assessing a working hypothesis in nucleus caudalis and dorsal horn neurons expressing proenkephalin and preprodynorphin. Cell Mol Neurobiol 1990; 10:73-98. 9. Roberts JL, Dutlow CM, Jakubowski M, Blum M, Miller RP. Estradiol stimulates preoptic area-anterior hypothalamic proGnRH-GAP gene expression in ovanectomized rats. Mol Brain Res 1989; 6:127-134. 10. Wise ME. Gonadotropin-releasing hormone secretion during the postpartum anestrous period of the ewe. Biol Reprod 1990; 43:719-725. 11. Niswender GD, Reichert LE Jr, Midgley AR Jr, Nalbandov AV. Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 1969; 84:1166-1173. 12. Adelman LA, Mason AJ, Haylick JS. Seeburg PH. Isolation of the gene and hypothalamic CDNA for the common precursor of gonadotropin-releasing hormone and prolactin release-inhibiting factor in human and rat. Proc Natl Acad Sci USA 1986; 83:179-183. 13. Rance NE, McMullen NT, Smialek JE, Price DL, Young III WS. Postmenopausal hypertrophy of neurons expressing the estrogen receptor gene in the human hypothalamus. J Clin Endocrinol & Metab 1990; 71:79-85. 14. Suzuki M, Mitlyodo S, Kobayashi M, Aida K, Urano A. Characterization and localization of mRNA encoding salmon-type gonadotropin-releasing hormone precursor of the masi salmon. J Mol Endocrinol 1992; 9:73-82. 15. Kaynard AH, Pau KYH, Hess DL, Spies HG. Gonadotropin-releasing hormone and norepinephrine release from the rabbit mediobasal and anterior hypothalamus during the mating induced luteinizing hormone surge. Endocrinology 1990; 127:1176-1185. 16. Goodman RG, Berriman SJ, Gu X, Lehman MN. Is a subset of gonadotropin-releasing hormone (GnRH) neurons involved in pulsatile GnRH secretion in the ewe? Soc Neurosci Abstr 1994; 20:272.10. 17. Barry J. Immunocytochemistry of luteinizing hormone releasing hormone neurons of the vertebrates. Int Rev Cytol 1979; 60:179-219. 18. Herbinson AE, Robinson JE Skinner DC. Distribution of estrogen receptor immunoreactive cells in the preoptic area of the ewe: co-localization with glutamine acid decarboxylase but not luteinizing hormone-releasing hormone. Neuroendocrinology 1992; 57:751-759. 19. Crowder ME, Nett TM. Pituitary content of gonadotropins and receptors for gonadotropin releasing-hormone (GnRH) and hypothalamic content of GnRH during the periovulatory period of the ewe. Endocrinology 1984; 114:234-239. 20. Bruder JM, Weirman ME. Evidence for transcriptional inhibition of GnRH gene expression by phorbol ester at a proximal promoter region. Mol Cell Endocrinol 1994; 99:177-182.
