| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a INRA, Physiologie de la Reproduction et des Comportements, 37380, Nouzilly, France
b Department of Physiology, West Virginia University, Morgantown, West Virginia 26506
ABSTRACT
Seasonal anestrus in ewes results from an increase in response to the negative feedback action of estradiol (E2). This increase in the inhibitory effects of E2 is controlled by photoperiod and appears to be mediated, in part, by dopaminergic neurons in the retrochiasmatic area of the hypothalamus (A15 group). This study was designed to test the hypothesis that E2 increases multiunit electrical activity (MUA) in the A15 during inhibitory long days. MUA was monitored in the retrochiasmatic area of 14 ovariectomized ewes from 4 h before to 24 h after insertion of an E2-containing implant subcutaneously. In six of these ewes, MUA activity was also monitored before and after insertion of blank implants. Three of the 14 ewes were excluded from analysis because E2 failed to inhibit LH. When MUA was recorded within the A15, E2 produced a gradual increase in MUA that was sustained for 24 h. Blank implants failed to increase MUA in the A15 area, and E2 did not alter MUA if recording electrodes were outside the A15. These data demonstrate that E2 increases MUA in the A15 region of ewes and are consistent with the hypothesis that these neurons mediate E2 negative feedback during long photoperiods.
dopamine, estradiol, hypothalamus, LH, seasonal reproduction
INTRODUCTION
Ovarian function in most breeds of sheep is controlled by photoperiod so that under natural conditions, ewes are fertile only during the fall and winter [1]. This control reflects a complex interaction between changes in the external environment and endogenous timing mechanisms within the animal [2, 3]. This interaction ensures that under normal conditions, ewes exposed to long days (LD) are anovulatory, and ewes exposed to short days (SD) are fertile. These photoperiodically controlled changes in ovarian function are caused by changes in the responsiveness of the hypothalamus to the negative feedback action of estradiol (E2) [1, 4]. During the breeding season, physiological levels of E2 inhibit LH and GnRH pulse amplitude but not frequency [5–7]. In contrast, during anestrus, E2 exerts a strong inhibition of the frequency of GnRH pulses [8], and the resultant decrease in LH pulse frequency [6, 7] disrupts normal ovarian function, causing anovulation [1].
Studies over the last decade have provided strong evidence that a group of dopaminergic (DA) neurons, with perikarya in the retrochismatic area of the hypothalamus (the A15 group), plays an important role in these seasonal changes in response to E2 negative feedback. Neurotoxic [9] and radiofrequency [10] lesions of the A15 decreased the inhibitory effects of E2 in ovariectomized ewes maintained on LD. These lesions also blocked the ability of a DA antagonist to stimulate LH release [10], suggesting that an inhibitory DA system had been disrupted. The lesions did not affect LH secretion in the absence of E2 or the ability of E2 to inhibit LH pulse amplitude in the breeding season [9, 10]. Taken together, these observations support the hypothesis that the increase in response to E2 negative feedback occurs in anestrous ewes because the LD photoperiod activates E2-sensitive DA neurons in the A15, which in turn inhibit GnRH pulse frequency. According to this proposal, in anestrus, E2 stimulates DA release from these neurons and thereby inhibits GnRH pulse frequency, but this steroid does not inhibit GnRH pulse frequency in the breeding season when the A15 DA system is inactive.
There is also some evidence that E2 stimulates the activity of the A15 DA neurons. Specifically, E2 treatment during LD increased the in vivo bioactivity of tyrosine hydroxylase (TH) in the A15 region [11] and induced the expression of the early immediate gene product, Fos, in A15 DA perikarya [12]. During SD, the same E2 treatment failed to increase Fos expression in these neurons [12], suggesting that there may be seasonal differences in their response to E2. In contrast, E2 increased TH mRNA levels during SD but not during LD [13]. One problem in interpreting these results is that indirect indices of neuronal activity have been used. Therefore, in this study, we have examined the effects of E2 on electrical activity of neurons in the A15 region of ewes exposed to LD. Because E2 takes 18–24 h to exert its inhibitory effects on LH pulse frequency [14], we used chronically implanted electrodes to monitor multiunit electrical activity (MUA) in conscious ewes.
MATERIALS AND METHODS
Animals
Adult Romanov ewes were maintained in light-controlled rooms; fed a maintenance diet of hay, straw, and corn; and given free access to water. They were exposed to alternating periods of SD (8L:16D) and LD (16L:8D) lasting at least 75 days, as described elsewhere [11, 13]. Ovariectomies were performed on anesthetized ewes via a midventral incision at least 5 mo before electrical recordings. All ewes received 2-cm-long Silastic implants containing E2 for at least 2 mo before the start of the experiment to ensure responsiveness to this steroid. All work with animals was done in accordance with the international guiding principles for biomedical research involving animals and under a permit from the French Ministry of Agriculture.
Recording Electrodes
The design of the recording electrodes and the procedure for MUA recording were adapted from previous work in the monkey [15], goat [16], and ewe [17]. Four Teflon-insulated tungsten wires (75-µm diameter) were glued (with epoxy) inside a single 22-gauge stainless steel tube, with about 4 mm of each wire extending beyond the ventral end of the tube. The dorsal end of each wire was then soldered to the appropriate pin of an integrated circuit (IC) socket, and these connections were sealed in dental acrylic. The 22-gauge tube was inserted into an 18-gauge tube that extended beyond the recording electrodes and was held in place by a spacer during surgical implantation [15–17]. Electrodes were unilaterally implanted into the A15 area with a stereotaxic apparatus and gas anesthesia as described elsewhere [9, 11]. Placement of electrodes was aided by lateral and frontal X-rays after injection of radio-opaque dye into the ventricular system. When the electrodes were 5 mm dorsal to the A15, the spacer was removed, and the electrodes lowered beyond the 18-gauge tube into the A15 area. The electrodes were then cemented in place and the animal allowed to recover for at least 2 wk.
Electrical Recordings
The day before the start of MUA recordings, two ewes were placed in individual stalls (they could stand up and lie down but not turn around), and each had a jugular vein catheterized (for blood collection). The two stalls were within a single Faraday cage in a room that had been electrically shielded. Two ewes were always within the Faraday cage during recordings. Animals had free access to food and water throughout the experiment.
The differential electrical activity between two of the four implanted electrodes was monitored with a preamplifier (operational amplifier IC, Texas Instruments, Dallas, TX) connected to the IC socket on the ewe's head and grounded to the skin in the neck. With four implanted electrodes, there were six different pairs of electrodes whose differential electrical activity could be recorded. All six pairs were tested before recording, and a single pair was chosen for each animal on the basis of the robustness of the signal. The differential signal from the preamplifier was further amplified and filtered (pass bandwidth, 0.3–3 kHz) with a Grass P511 amplifier (Grass Instruments, Quincy, MA). The signal was then converted from analog to digital and analyzed using commercially available software (Brain Wave System, Discovery version 3.1, Longmont, CO). Spikes were identified by threshold criteria that were selected at the beginning of each recording to give approximately 100 spikes/sec. The averaged activity during each minute was recorded and stored for later analysis.
Experimental Protocol
Seventeen E2-treated ovariectomized ewes were switched from SD to LD, and recordings were then done after an average of 84 days of exposure to LD (range: 71 to 98 days). The E2 implant was removed approximately 2 wk before the start of recordings and at least 2 wk after surgical implantation of electrodes, and the ewes were placed in stalls and prepared for blood collection and recording the day before recording. Ewes continued to be exposed to LD while in the recording room. At 0800 h the next day, MUA recording and blood sampling (2 ml every 10 min from the jugular catheter) were begun. At 1200 h (0 h of the experiment), an implant was inserted s.c. in the front axillary region after injection of a local anesthetic (Xylocaine), and blood samples were collected for another 4 h. Multiunit electrical activity recording was continued for 24 h after implant insertion (so that data were collected continuously for 28 h), and frequent blood samples were also collected for 4 h on Day 1 (20–24 h). MUA and LH pulses were monitored in all 17 ewes before and after insertion of an E2 implant. To control for E2 treatment, MUA and LH pulses were also monitored in 6 of these 17 ewes before and after insertion of blank implants. For these 6 ewes, the order of treatment (blank vs. E2 implant) was randomized using a crossover design, with approximately 3 wk between treatments.
Identification of Electrode Placement
After completion of the experiment, the ewes were anesthetized with pentobarbital, and direct current was passed between the two recording electrodes (30 sec, 2 mA, two times with inverted polarities) to produce an electrolytic lesion. Approximately a week later, ewes were killed with an overdose of pentobarbital, the heads perfused with 4% paraformaldehyde, and the appropriate block of tissue collected and placed in fixative. Frozen coronal sections (40-µm thick) were cut, and every fifth section was stained with cresyl violet and examined to determine location of the lesion. Anteroposterior location was determined relative to the anterior commissure (A32, from the stereotaxic atlas of Richard [18]), and medial lateral location was measured directly on the section. The electrodes in three ewes were not in the retrochiasmatic area (two were in the optic chiasm; one was 3 mm dorsal); data from these animals were not analyzed.
Analyses
Luteinizing hormone concentrations were measured in duplicate with a previously described radioimmunoassay [19]. Luteinizing hormone pulses were identified by using the Monro algorithm (Baxter parameters used were B1 = 0.055, B2 = 0.06877, and B3 = 0.00063), and interpulse intervals (IPI) for each 4-h period were calculated. Multiunit electrical activity was averaged for 1-h bins throughout the duration of recording, and these values were normalized to the mean of pretreatment values (first 4 h of each data collection). Changes in MUA (1-h bins) and IPI (4-h bins) were statistically assessed by one-way and two-way ANOVA with repeated measures; P < 0.05 was considered statistically significant.
RESULTS
Effects of Implants on LH Pulse Frequency
Treatment with E2 significantly increased the interval between LH pulses on Day 1 (Fig. 1) of treatment (from 45.6 ± 2.5 to 68.1 ± 6.2 min, P < 0.01), but the E2 implants had no effect on pulsatile LH secretion during the 4 h immediately after insertion (47.5 ± 2.1 min). Blank implants had no effect on pulsatile LH secretion in the six ewes that received this treatment (IPI on Day 0: 48.6 ± 4.6 min; on Day 1: 49.9 ± 3.8 min). Although E2 significantly decreased pulse frequency, this effect was not observed in three of the E2-treated ewes, and consequently, MUA readings from these animals were not included in subsequent analyses.
|
Sites of Recordings
The medial-lateral and anterior-posterior positions of the electrodes in 11 ewes in which E2 inhibited LH pulse frequency (IPI on Day 1 was greater than 110% of that on Day 0) are illustrated in Figure 2. In the dorsal-ventral axis, all electrodes were at the level of the base of the third ventricle. In six ewes, the electrodes were either in or on the edge of the A15 area, whereas electrodes were clearly outside the A15 in five animals; five of the ewes with electrodes in the A15 were treated with both blank and E2 implants.
|
Electrical Activity
Multiunit electrical activity within an animal occasionally showed abrupt increases or decreases, lasting only 2–3 min (Fig. 3), that in most ewes (see below) were not correlated with pulsatile LH secretion or behavioral changes (e.g., chewing, lying down). There were usually fewer than 10 such episodes over the 28-h sampling period. In five ewes, an abrupt increase or decrease in electrical activity was observed so that MUA reached a new plateau that was then maintained for several hours; the data presented in Figure 3 is a particularly strong example of this variability. A third pattern observed in the MUA of four animals was suggestive of a diurnal variation in that there was a clear slope evident at the same time on Day 0 and Day 1 of sampling (Fig. 1). This pattern was evident in two of the six ewes receiving both blank and E2-filled implants and, in these two animals, occurred during both recording periods (Fig. 1). The stress of implant insertion had no obvious effect on MUA.
|
Insertion of E2 implants produced a consistent increase in MUA if the recording electrodes were in the A15 area (Figs. 1–4). This increase began shortly after implant insertion and developed gradually over the next 8 h; the level reached at 8 h was then sustained throughout the 24 h of observation (Figs. 3 and 4). In contrast, blank implants produced no consistent change in MUA in the A15 area (Figs. 3 and 4). The sharp fall in mean MUA in controls between 2 and 4 h after implant insertion is due to data from two ewes (e.g., Fig. 3) and was not statistically significant. Statistical analysis of the effects of blank vs. E2 treatment on MUA in the A15 indicated a significant effect of E2 (P < 0.04) and a significant treatment x time interaction (P < 0.01). Although E2 treatment increased MUA within the A15, it had no effect on MUA in the five animals in which the electrodes were outside this area (Fig. 5). Statistical comparison of MUA in these two groups of ewes before and after E2 treatment indicated a significant effect of electrode placement (P < 0.05) and a significant placement x time interaction (P < 0.001).
|
Unexpected Observation
In two ewes treated with blank and E2-containing implants, bursts in MUA that correlated with pulsatile LH secretion were observed (Fig. 6). In these ewes, every identified LH pulse was coincident with a spike in MUA, but some increases in MUA occurred without concurrent LH pulses. However, even in these cases, there were usually simultaneous increments in LH that did not meet the criteria for an LH pulse. This activity was observed before and after insertion of E2 or blank implants, although the frequency of both spikes of MUA and pulses of LH secretion was decreased on Day 1 during E2 treatment (Fig. 6). The electrodes in these two ewes were not in the central portion of the A15 region; one was posterior to this area and the other was on the medial edge of it (Fig. 2).
|
DISCUSSION
The results of this experiment demonstrate that E2 treatment of ovariectomized ewes maintained on inhibitory photoperiod increases the spontaneous electrical activity of neurons within the A15 region. In addition, this stimulatory action of E2 was not observed if electrodes were near, but not in, the A15 area, so it is unlikely to represent increased activity of axons passing through the retrochiasmatic area. It is not clear whether the E2-induced increase in MUA reflects changes in DA or other neurons in the A15 area. But the positive correlation between electrode placement within the A15 and changes in MUA in response to E2 (Fig. 2) supports an action on A15 DA perikarya. These data are thus consistent with the hypothesis that A15 DA neurons play a role in mediating E2 negative feedback in anestrous ewes.
The stimulatory effect of E2 on MUA observed in this study was not seen in a previous experiment in which the effect of E2 on electrical activity in anestrous ewes was examined [17]. However, the electrodes in that study were placed in various sites in the hypothalamus, and only four ewes were recorded from the lateral retrochiasmatic area. Moreover, the importance of the A15 to E2 negative feedback was not understood at the time this study was done, so we do not know how close the recording electrodes were to the A15. On the basis of the results of the present work, it is clear that electrodes in the retrochiasmatic area can miss E2-induced changes in MUA if they are outside the A15.
The results of this experiment are consistent with those of two other studies suggesting that E2 can stimulate A15 DA neurons. Thus, E2 treatment of ewes on LD increased Fos expression in A15 perikarya [12] and the in vivo bioactivity of TH [11] in the A15 region. Because E2 did not stimulate TH mRNA levels in such animals [13], the increase in TH bioactivity probably reflects changes in phosphorylation of this enzyme [20]. On the other hand, other possible measurements of DA activity in this area are not increased by E2 during anestrus. For example, in vivo E2 treatment did not increase retrochiasmatic TH bioactivity when it was assessed with an in vitro assay [21]. In addition, E2 was more effective in increasing TH mRNA in the A15 of ewes exposed to short days than in those exposed to long days [13]. The observation that E2 stimulates MUA in the A15, as well as Fos expression and in vivo TH activity, strengthens the conclusion that A15 neurons are responsive to this steroid.
Although there is strong evidence that E2 increases the activity of A15 neurons, the mechanisms underlying this effect are unclear. Previous work has demonstrated that E2 acts within the retrochiasmatic area to inhibit LH pulse frequency [22], but this is most likely either a nongenomic or an indirect action because A15 DA perikarya do not appear to contain either alpha [23, 24] or beta [25] receptors for E2. There is also suggestive evidence that E2 may act in the preoptic area to stimulate a neural system that projects to the A15 area [26]. The effects of E2 on MUA occurred fairly rapidly, with an increase clearly evident within a few hours after insertion of the E2 implants (Figs. 2 and 3). The relatively rapid onset of E2 action is consistent with a nongenomic action [27] but does not rule out an action via a classical nuclear receptor [28]. It is interesting to note that the E2-induced increase in MUA occurs much more rapidly than the increase in Fos expression, which was not evident 6 h after the start of E2 treatment [12], and the E2-induced decrease in LH pulse frequency, which takes approximately 24 h [14]. If MUA reflects changes in A15 DA neurons, then this lag time between the increase in MUA and the decrease in LH secretion raises the possibility that some processing of the DA signal occurs before it exerts its inhibitory effects on GnRH release.
In two ewes, brief increases in MUA were observed that correlated with the episodic release of LH. Similar data have previously been reported for monkeys [15, 29], rats [30], and goats [31], but this is the first such observation in sheep. In the rat and goat, this electrical manifestation of the GnRH pulse generator occurs primarily in the median eminence [30] and medial basal hypothalamus [31], respectively. However, in the monkey, spikes in MUA have been observed in the retrochiasmatic area, although usually they are recorded from more caudal sites [32]. In the sheep, there are few GnRH perikarya in the area where these spikes in MUA were recorded [33, 34]. However, GnRH axons run through this area on their way to the median eminence, so the activity observed in these two ewes could have been recorded from these axons. It should be noted that previous attempts to correlate spikes in MUA with GnRH neurons in the monkey have not been entirely successful [32].
In these two ewes, spikes in MUA continued during E2 treatment and occurred less frequently on Day 1 than during the pretreatment period on Day 0 (Fig. 6). This contrasts with the ovariectomized monkey, in which E2 infusions completely blocked spikes of MUA [35]. However, similar E2 treatments in ovariectomized goats [36] and rats [37] decreased the frequency of MUA spikes without completely abolishing them. These conflicting data may reflect differences in dose and duration of the E2 treatment. In the monkey, continuous exposure to low E2 concentrations decreased the frequency of MUA spikes after 1 wk and only abolished them after 3 wk of treatment. In the anestrous ewe, chronic E2 treatments completely block pulsatile LH secretion [6, 12] and would be expected to do the same to the spikes in MUA associated with LH pulses. Although there are a number of studies that examined the effects of ovarian steroids on these spikes of MUA activity [29, 35–39], to our knowledge, this is the first report of changes in MUA baseline associated with the negative feedback actions of E2. In this regard, these previous studies specifically selected electrical activity thought to reflect a component of the GnRH pulse generator, whereas this study targeted an area that was not thought to be part of this pulse generator but provides afferent input to it.
In conclusion, the data reported in this study demonstrate that E2 increases MUA in the A15 area of ewes maintained on inhibitory photoperiod. They thus are consistent with the hypothesis that A15 DA perikarya play a role in E2 negative feedback in anestrous ewes. However, the mechanisms by which E2 increases the activity of these neurons remain to be elucidated.
|
ACKNOWLEDGMENTS
We thank M.L. Goubillon and J.C. Thalabard for their expert advice and help in setting up the software used to analyze MUA.
FOOTNOTES
First decision: 26 April 2000.
1 R.L.G. supported by NIH-HD17864 and NIH-FO6-TWO2219. 
2 Correspondence: Robert L. Goodman, Department of Physiology, West Virginia University Health Sciences Center, Morgantown, WV 26506-9229. FAX: 304 293 3850; bgoodman{at}hsc.wvu.edu
Accepted: June 8, 2000.
Received: March 30, 2000.
REFERENCES
containing neurons in the preoptic area of the ewe depends on reproductive status. Endocrinology 2000; 141:190–196.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]
|
|
|
|
 |
|
 C. J McManus, M. Valent, S. L Hardy, and R. L Goodman Does nitric oxide act in the ventromedial preoptic area to mediate oestrogen negative feedback in the seasonally anoestrous ewe? Reproduction, July 1, 2007; 134(1): 137 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. L. Adams, R. L. Goodman, A. K. Salm, L. M. Coolen, F. J. Karsch, and M. N. Lehman Morphological Plasticity in the Neural Circuitry Responsible for Seasonal Breeding in the Ewe Endocrinology, October 1, 2006; 147(10): 4843 - 4851. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Anderson, S. L. Hardy, M. Valent, H. J. Billings, J. M. Connors, and R. L. Goodman Evidence that Thyroid Hormones Act in the Ventromedial Preoptic Area and the Premammillary Region of the Brain to Allow the Termination of the Breeding Season in the Ewe Endocrinology, July 1, 2003; 144(7): 2892 - 2901. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
