Phyllis M. Wise*, Kathryn Scarbrought, Nancy G. Weiland$, and Gregg H. Larsont

Transcription

1 Diurnal Pattern of Proopiomelanocortin Gene Expression in the Arcuate Nucleus of Proestrous, Ovariectomized, and Steroid-Treated Rats: A Possible Role in Cyclic Luteinizing Hormone Secretion Phyllis M. Wise*, Kathryn Scarbrought, Nancy G. Weiland$, and Gregg H. Larsont Department of Physiology University of Maryland School of Medicine Baltimore, Maryland Opiate peptides are thought to modulate the pattern of LH release in female rats. We tested the hypothesis that changes in proopiomelanocortin (POMC) gene expression occur in proestrous (PRO) and ovariectomized (OVX) steroid-treated rats which may explain their unique patterns of LH secretion. Using in situ hybridization, we examined whether diurnal changes in POMC gene expression occur in the arcuate nucleus. Four groups of rats were used in this study. 1) PRO rats were used after exhibiting at least two consecutive 4-day estrous cycles; 2) OVX rats were killed 9 days after ovariectomy; 3) estradiol (E 2 )-treated rats were OVX for 7 days and then treated for 2 days; and 4) E 2 -progesterone (P 4 )- treated rats were treated with E 2 as described above, and on day 9 at 1030 h, P 4 was administered. Rats were killed at 2300, 0300, 1000, 1300, 1500, 1800, or 2300 h, beginning on the evening of diestrous day 2 or day 8 after ovariectomy. POMC gene expression exhibited a diurnal rhythm on PRO. Levels of mrna rose during the morning, peaked between h, and decreased by 2300 h. In E 2 -treated rats, which exhibited a LH surge similar in timing to the PRO surge, POMC mrna levels exhibited a diurnal rhythm strikingly similar to that observed in PRO animals. OVX abolished the rhythm; however, average POMC mrna levels across the 24-h period were not significantly different from those in PRO or E 2 -treated rats. P 4 treatment increased POMC mrna levels by 2300 h compared to those in all other experimental groups. These results support the concept that the pattern of POMC gene expression influences the pattern of secretion of LH. Furthermore, they demonstrate that 1) E /90/ S02.00/0 Molecular Endocrinology Copyright 1990 by The Endocrine Society imposes a diurnal rhythm on POMC gene expression, and 2) P 4 alters this rhythm and enhances POMC mrna levels. Finally, these results emphasize the importance of considering the presence of diurnal rhythms in gene expression when designing experiments and interpreting data relative to the effects of endocrine manipulations. (Molecular Endocrinology 4: , 1990) INTRODUCTION Opiate peptides modulate the pattern of LH secretion (for reviews, see Refs. 1-3). Considerable experimental data demonstrate that endogenous opiate peptides suppress the release of GnRH (4) and thereby contribute to the maintenance of basal secretion of LH in intact rats. This concept gains support from the observation that opiate antagonists stimulate LH release in intact male (5) and female (6) rats. Conversely, decreased release of opiate peptides is thought to allow increased LH secretion during preovulatory (6, 7) or steroid-induced (8-10) LH surges or after ovariectomy (11). Thus, administration of morphine suppresses LH surges (6, 12) and castration-induced LH release (13). The ability of estrogen and progesterone (P 4 ) to exert positive and negative feedback effects on LH may depend partially on their effects on opiate activity (9, 14). Steroids may regulate opiate peptide activity by several mechanisms, including influencing the density of opiate receptors (15), regulating proopiomelanocortin (POMC) gene expression (16-19), altering opiate peptide concentrations in tissue (16, 20), and modulating opiate release into hypophysial portal circulation (21). Opiates may regulate GnRH neurons directly (22) or influence them 886

2 Diurnal Pattern of POMC Gene Expression in Arcurate Nucleus 887 indirectly through neurotransmitters (10, 23-25) that, in turn, influence GnRH release. The purpose of this study was to test the hypothesis that changes in POMC gene expression may contribute to the occurrence of cyclic release of LH in proestrous (PRO) and steroid-treated rats. We reasoned that if POMC gene expression influences cyclic LH secretion, we may observe a unique diurnal pattern of POMC gene expression when LH surges occur which is different than that under circumstances in which no cyclic LH release occurs. Using in situ hybridization methodology, we characterized the level of POMC mrna within individual cells of the arcuate nucleus of PRO rats at multiple times over the course of 24 h. In addition, we quantified POMC gene expression in ovariectomized (OVX) rats to assess the diurnal pattern of POMC gene expression in rats in which LH is elevated during the entire day and no LH surges occur. Finally, we measured POMC mrna in estradiol (E 2 )- and E 2 - primed P 4 -treated (E 2 P 4 ) rats, which exhibit preovulatory-like LH surges, to determine whether the diurnal pattern of POMC gene expression is similar in these two endocrine models compared to that in intact PRO animals. RESULTS LH concentrations in PRO, OVX, E 2, and E 2 P 4 rats are shown in Fig. 1. PRO rats exhibited a LH surge during the afternoon which peaked between h. In OVX rats, LH concentrations were elevated in the morning compared to those in all other groups (P < 0.05); however, no surge occurred during the afternoon. E 2 rats exhibited a LH surge similar in timing and blunted in amplitude compared to that in proestrous rats. P 4 administration at 1030 h to E 2 rats caused an earlier onset of the LH surge that afternoon and amplified the magnitude of the surge (P < 0.05). In a preliminary experiment we tested the effect of increasing concentrations of 35 S-labeled POMC crna on the hybridization signal (Fig. 2). Increasing concentrations of probe resulted in increasing signal, as measured by two methods of quantitation: 1) mean area of computer-enhanced grains within a window drawn around individual labeled cells (Fig. 2, top panel), and 2) mean gray level within the same window (bottom panel). Saturation was achieved around 0.7 ng/m\ probe regardless of the index used to quantitate the hybridization signal. We used the area covered by computer-enhanced grains as a method of choice to quantitate levels of mrna, because this end point is more accurate and sensitive to small changes at the lowest signal levels. At the lowest levels of hybridization, the stain taken up by the cell contributes to the gray level (see Fig. 2, bottom panel, 0.2 Mg/ml); therefore, measurement of gray level does not allow differentiation of small differences when the hybridization signal is low Time of Day (h X 10" 2 ) Fig. 1. Serum Concentrations of LH (Mean ± SE) in PRO, OVX, OVX E 2 -Treated, and E 2 -Primed P 4 -Treated Rats PRO rats exhibit a LH surge during the afternoon; LH concentrations are higher at 1500 and 1800 h than at other times of day (by one-way ANOVA and Duncan's multiple range test, P < 0.05). In OVX rats, LH concentrations are elevated in the morning compared to those in all other groups (by twoway ANOVA, P < 0.05); however, no surge occurs during the afternoon (by one-way ANOVA, P > 0.05). E 2 rats exhibit a LH surge similar in timing and blunted in amplitude compared to that in PRO rats; concentrations are higher at 1500 and 1800 h than at other times of the day. P 4 administration at 1030 h to E 2 -treated rats causes an earlier onset of the LH surge that afternoon and amplifies its magnitude; LH concentrations are higher at h than during the morning. No SES around the means are shown when they are smaller than the symbols used to depict the means. Figure 3 shows frequency distributions of the area of enhanced grains over cells in the arcuate nucleus from a representative brain section hybridized with a low concentration of crna (0.2 fig/m\) and a brain section from the same animal that was hybridized with a high concentration of crna (0.7 ng/rr\\). These frequency distributions reveal the striking heterogeneity of mrna levels exhibited by individual cells within one section of the arcuate nucleus. Nevertheless, the variance between animals, depicted by the SE around the means in Fig. 2, shows that the variation in the mean hybridization signal from approximately 50 cells/section is relatively small. The cover figure is a composite photograph of a brain section, exhibiting POMC mrna-containing cells limited to the arcuate nucleus {top panel), a cluster of heavily labeled cells (lower left panel), and some lightly labeled cells (lower right panel). On PRO, POMC gene expression exhibited a diurnal rhythm (Fig. 4). Levels of mrna rose during the morning, peaked between h, and decreased by 2300 h. Ovariectomy abolished the diurnal rhythm in POMC gene expression observed in PRO rats, but did not alter

3 MOL ENDO Vol 4 No. 6 SATURATION ANALYSIS 30 Frequency Distributions of Hybridization Signal from Individual Cells 25 S Qo.2 //g/ml probe Ho.7 //g/ml probe < >250 Area of Enhanced Grains (urn 2 ) Fig. 3. Frequency Distributions of the Area of Enhanced Grains over Cells in the Arcuate Nucleus from a Representative Brain Section Hybridized with a Low Concentration of crna (0.2 Mg/ml) and a Brain Section from the Same Animal that Was Hybridized with a High Concentration of crna (0.7 M9/ml) Note the striking heterogeneity of mrna levels exhibited by individual cells within one section of the arcuate nucleus. C_ Concentration of crna (//g/ml) Fig. 2. The Effect of Increasing Concentrations of POMC crna on the Hybridization Signal Similar results are obtained using two methods of quantification. Top panel, Mean area of computer-enhanced grains within a designated window drawn over an individual cell. Bottom panel, Mean gray level within the same window. the average level of POMC mrna (PRO, 395 ± 14.1 Mm 2 ; OVX, 400 ± 14.7 Mm 2 ; Fig. 5). In E 2 rats, which exhibited an LH surge similar in timing to the PRO surge, POMC mrna levels exhibited a diurnal rhythm strikingly similar to that observed in PRO animals; POMC mrna levels rose significantly between 2300 h on day 8 and 0300 h on day 9 and fell during the afternoon, such that values at 1800 h were lower than peak values at 0300 h (Fig. 5). Average POMC mrna levels across the 24-h period were not significantly different in E 2 compared to OVX rats (E 2, 380 ± 13.0 Mm 2 ; OVX, 400 ± 14.7 Mm 2 ). In E 2 rats, P 4 treatment increased POMC mrna levels compared to those in all other experimental groups, including E 2 animals (Fig. 6). POMC mrna levels at 2300 h were significantly higher than those at 1800 h in P 4 animals as well as greater than those in any other experimental group at 2300 h Time of Day (h X 10~ 2 ) Fig. 4. POMC mrna Levels in PRO Rats POMC gene expression exhibits a diurnal rhythm; levels rise significantly between 2300 h on diestrous day 2 and 0300 h on PRO and peak at 1000 h, then decrease during the afternoon, such that values at 2300 h on PRO night are significantly lower than the peak and are the same as on the previous night at 2300 h (by one-way ANOVA and Duncan's multiple range test, P < 0.05). DISCUSSION We have clearly demonstrated that the level of POMC mrna within the arcuate nucleus of PRO rats exhibits a diurnal rhythm; levels are low during the night of diestrous day 2, rise dramatically during the early morning, and return to low levels late in the evening of PRO. We interpret these data to suggest that opiate tone increases during the evening of diestrous day 2, peaks

4 Diurnal Pattern of POMC Gene Expression in Arcurate Nucleus Time of Day (h X 10" 2 ) Fig. 5. POMC mrna Levels in OVX and E 2 Rats OVX abolishes the diurnal rhythm in POMC gene expression observed in PRO rats (by one-way ANOVA, P > 0.05). In E 2 rats POMC mrna levels exhibit a diurnal rhythm similar to that observed in PRO animals; POMC mrna levels rise between 2300 h on day 8 to 0300 h on day 9 and fall during the day, such that values at 1800 h are lower than peak values at 0300 h (by one-way ANOVA and Duncan's multiple range test, P < 0.05). (X cz CJ ^- o a_ CVI =4 ins, co c_ O) o CO cu c_ fo Time of Day (h X 10" 2 ) Fig. 6. POMC mrna Levels in E 2 and E 2 P 4 Rats P 4 increases POMC mrna levels compared to values in all other experimental groups, including E 2 animals (by two-way ANOVA, P < 0.05). In E 2 P 4 animals POMC mrna levels increase significantly between h (by one-way AN- OVA and Duncan's multiple range, P < 0.05). At 2300 h, they are higher than those in any other experimental group at that time (by one-way ANOVA and Duncan's multiple range test, P < 0.05). early during the morning of PRO, and decreases before the initiation of the preovulatory surge of LH. To accept this interpretation, one must accept several assumptions that are based upon previous findings in other systems. First, we assume that there is a delay between changes in the rates of transcription or stabilization of mrna and detectable changes in the level of mrna, as has been shown in other systems (26, 27). Second, we assume that changes in the release of a peptide are linked to changes in the rate of transcription that, in turn, lead to changes in the level of mrna. This is supported by information gained from combining studies on the release of inhibin (28) and on changes in inhibin mrna levels (29). The length of the delay between peptide secretion, alterations in the rate of gene transcription and/or stability, and a detectable change in the pool of mrna depends upon the magnitude of the change in the rate of transcription or stability, the size of the pool, and the half-life of the mrna (30). Since these variables are not known for POMC mrna in the arcuate nucleus, we can only estimate when changes in rates of transcription have occurred that lead to changes in PRO levels of mrna. In the neurointermediate lobe, changes in the rate of transcription of the POMC gene result in changes in the level of mrna approximately 8 h later (31). If this temporal relationship applies to the arcuate nucleus, it allows us to estimate that low rates of transcription occur during the evening of diestrous day 2 through the early morning of PRO and decrease during late PRO morning or early afternoon. To determine whether a proestrous-like diurnal rhythm in POMC gene expression is critical to the occurrence of LH surges, we induced LH surges in E 2 - treated rats and assessed the pattern of POMC gene expression. Our results indicate that a diurnal rhythm in levels of POMC mrna accompanies E 2 -induced LH surges. Both the timing of the rise and fall in POMC mrna and the amplitude of the rhythm are strikingly similar to those observed in PRO rats. This diurnal rhythm was absent in OVX rats. Together, the correlation of the presence of a rhythm in animals that exhibit a surge and its absence in nonsurging rats lend credence to the concept that both the rise and/or the rhythm of POMC gene expression contribute to LH surges. In OVX rats the overall mean concentration of POMC mrna is not different from that in intact PRO or E 2 females; however, no diurnal rhythm is detectable in POMC gene expression. Thus, both exogenously administered E 2 and the endogenously secreted preovulatory secretion of E 2 impose a diurnal rhythm on POMC gene expression. These data are in accordance with the findings that /3-endorphin concentrations were not different in OVX compared with cycling rats killed on various days of the estrous cycle (20). However, in some regards the current data were unexpected based upon the negative feedback effects of E 2 on LH. Specifically, estrogen treatment suppresses castration-induced hypersecretion of LH, an effect that can be reversed by naloxone (9, 10), suggesting that E 2 may suppress LH through enhancing opiate activity. Steroids have been associated with decreased /?- endorphin concentrations in the medial basal hypothalamus (16, 20) and increased concentrations in portal plasma (14, 21) as well as decreased (16, 19) or in-

5 MOL ENDO Vol 4 No. 6 creased (18) POMC mrna in the arcuate nucleus. It is interesting to consider the possibility that the apparent effect of castration and steroid treatment on POMC mrna may depend upon the time of day when the tissue was collected. Several studies, including the present one, demonstrate that opiate activity, as assessed by POMC mrna levels (31) or peptide concentrations (32, 33), exhibit a diurnal rhythm in intact or steroid-treated rats. Assuming that opiate tone exhibits a diurnal rhythm in intact males (33), the results of Blum et al. (16) and Chowen-Breed et al. (18) may differ due to differences in the time of tissue collection. In this regard, it is intriguing to note that POMC mrna levels in intact PRO and steroid-treated rats tend to be higher than those in OVX animals in the morning, but lower than those in OVX animals in the afternoon. In addition, other factors, such as the length of castration at the time of steroid treatment (18, 20), the length and dose of steroid treatment (14, 20, 34), and the anatomical region analyzed (17), may influence the measured levels of POMC mrna. P 4 treatment of E 2 rats stimulated a rapid increase in POMC mrna levels: 1) animals treated only with E 2, POMC mrna decreased by 1800 h; P 4 blocked this decrease; and 2) at 2300 h, POMC mrna levels were significantly higher in P 4 -treated rats than in all other experimental groups. We treated E 2 P 4 rats with P 4 at 1030 h. In contrast, PRO rats began to secrete endogenous P 4 in response to the LH surge approximately 6-10 h later or around h. In addition, the dose of P 4 that we administered results in significantly higher plasma P 4 than is normally secreted endogenously (35). Therefore, we would predict that in intact cycling rats, a P 4 -induced rise in POMC mrna would occur later (perhaps early on estrus) and to a lesser extent than in E 2 P 4 rats. The rise in POMC mrna may inhibit LH surges on estrus and may mediate the inhibitory effects of P 4 on subsequent days (36). The present results demonstrating that P 4 enhances POMC gene expression corroborate and extend the previous findings that P 4 increases opiate activity (14, 19,20,37), as assessed by a variety of indices of opiate activity. Together these results strongly suggest that increased POMC gene expression may not play a role in P 4 's positive feedback effects on LH that afternoon, but may contribute to P 4 's ability to inhibit LH surges on subsequent days (36) and may prevent LH surges in cycling rats in estrus (38, 39). The P 4 -induced increase in POMC gene expression, presumably reflecting increased rates of transcription and/or stability, at the time P 4 enhances the amplitude of the LH surge and advances the timing of the onset of the surge is unexpected based upon the findings of Kalra and his colleagues (7-9). The use of the E 2 P 4 -treated animal model reveals two additional important clues regarding the role of changing POMC gene expression in regulating cyclic LH secretion. First, since LH surges occur under this steroid treatment paradigm, and POMC mrna levels do not decrease, we must conclude that the afternoon fall in POMC mrna levels, which we observed in PRO and E 2 rats may be a normal component of the LH surge; however, this aspect of the diurnal rhythm is not essential to cyclic LH release. Second, a comparison of the experimental groups included in this study reveals that in all animals destined to surge during the afternoon (PRO, E 2, and E 2 P 4 ), we observed an increase in POMC mrna levels between h during the morning before the surge. In contrast, in rats that will not surge during the day or during the subsequent day (OVX), no such increase occurred during the early morning hours. Thus, our data suggest that the suppression of POMC mrna during the evening before the afternoon of a LH surge may constitute an important component of the trigger for LH surges. Other lines of evidence corroborate the concept that E 2 induces an important neurochemical signal that occurs several hours before the onset of LH release. Kalra (40) found that if rats are OVX after 0300 h on PRO, the preovulatory surge will occur. However, if rats are OVX before 2300 h on diestrous day 2, the LH surge will not occur. Thus, an important E 2 -induced event appears to occur during the interval between 2300 and 0300 h. It is tempting to speculate that suppressed opiate activity may contribute to that event. In summary, we have demonstrated that POMC mrna levels in the arcuate nucleus exhibit a diurnal rhythm in PRO rats and that E 2 treatment imposes a similar rhythm on OVX rats. P 4 exerts a dramatic and rapid effect on POMC gene expression. Together these observations support the concept that changing POMC gene expression influences the secretion of LH. Furthermore, they emphasize the importance of considering the presence of diurnal rhythms in gene expression when designing experiments and interpreting data relative to the effects of endocrine manipulations. MATERIALS AND METHODS Animals Female Sprague-Dawley rats, 2 months of age, weighing between g, were purchased from the supplier (Zivic- Miller, Allison Park, PA) and maintained in a temperature- and light-controlled environment (14-h light, 10-h dark cycle; lights on at 0400 h). Animals were given food and water ad libitum, and estrous cyclicity was monitored by daily vaginal lavage. Four groups of rats were used in this study. 1) PRO rats were used after exhibiting at least two consecutive 4-day estrous cycles. 2) OVX rats were killed 9 days after ovariectomy. 3) E 2 rats were OVX for 7 days and then treated for 2 days with a Silastic capsule containing E 2 (180 ng/m\\ 20 mm in length). 4) E 2 P 4 rats were treated with E 2 as described above, and on day 9 at 1030 h, P 4 (3 mg/rat) was injected sc. Rats (n = 6-7/treatment/time) were killed at 2300, 0300, 1000, 1300, 1500,1800, or 2300 h, beginning on the evening of diestrous day 2 or day 8 after ovariectomy. Trunk blood was collected, and serum was separated and frozen until it was radioimmunoassayed for LH. Brains were removed, frozen on dry ice, and stored at -70 C until they were sectioned (20 ^m) in a cryostat. Eight coronal sections of the arcuate nucleus [A4620-A4230 in Konig and Klippel (41)] from each rat were used in this study. Brain sections were thaw-mounted onto gelatin-coated

6 Diurnal Pattern of POMC Gene Expression in Arcurate Nucleus 891 slides and stored at -70 C until hybridization histochemistry was performed. In Situ Hybridization In situ hybridization was performed according to the method of Rogers et al. (42) with several modifications. Brain sections were defrosted, dried rapidly, then fixed in RNase-free phosphate-buffered 4% paraformaldehyde for 5 min. Proteinase-K treatment was omitted, since in a preliminary experiment we found that this treatment decreased the hybridization signal strength and increased its variability (data not shown). Slides were washed sequentially in phosphate buffer (0.1 M), diethylpyrocarbonate-treated water, and 80 HIM triethanolamine buffer (ph 8.0), acetylated using 0.25% acetic anhydride in triethanolamine buffer, and set in 2x sodium chloride-sodium citrate buffer (SSC; 1X SSC = 150 ITIM NaCI, 15 ITIM trisodium citrate) until probe was applied. An 35 S-labeled riboprobe (SA, 1.5 x 10 8 dprn/^g) complementary to a fragment of the mouse pituitary POMC precursor containing 98 basepairs of the 5' noncoding sequence, the entire 705-bp coding sequence, and 105 bp of the 3' noncoding sequence was transcribed and hydrolyzed to fragments with a mass average of approximately 150 bases according to the methods of Chowen-Breed et al. (18). For the transcription reaction, we used 50 HM total a- thio-utp, of which 3.5 ^M was 35 S-labeled. Hybridization buffer containing 0.6 ng/m\ crna and 500 Mg/ml unlabeled trna was applied to the sections, and slides were coverslipped with parafilm, and incubated overnight at 45 C in humidified petri dishes. After hybridization, coverslips were floated off, and slides were rinsed in 4x SSC containing 4 ITIM DTT. All subsequent buffers contained 2 ITIM DTT; slides were treated for 30 min with RNase-A (20 nglm\) at 37 C and sequentially washed for min in RNase buffer (37 C), 2x SSC, and 0.1 x SSC at 60 C. Slides were dehydrated, dipped in Kodak NTB-3 emulsion (Eastman Kodak, Rochester, NY) which had been previously diluted 1:1 with distilled water, exposed, and developed using conventional photographic methods. Sections were counterstained with % Toluidine blue, and coverslips were applied. In preliminary experiments, saturation analysis was performed by hybridizing brain sections with varying concentrations of probe (Fig. 2). We quantitated the hybridization signal on duplicate slides at each probe concentration from four animals. Six concentrations of crna ( Mg/ml) were used to construct the saturation curve. In addition, varying lengths of exposure time were tested to determine the optimal length of exposure to emulsion (data not shown). Image Analysis The level of POMC mrna in individual cells was assessed using a Bioquant IV Image Analysis System (R & M Biometrics, Bethesda, MD). The amount of mrna was measured using two indices (Fig. 2). First, we measured the gray level (which is related to light transmission) within a window drawn around an individual cell that was covered by silver grains. Second, we quantified the area covered by computer-enhanced silver grains within the same window. In this case, a gray level threshold was chosen such that the majority of the silver grains were computer enhanced and measured, but the histologically stained nucleus was below the threshold and, therefore, not included in the measurement. A labeled cell was included in the analysis only if 1) its nucleus was visible within a cluster of silver grains, 2) the cluster did not overlap with the cluster of another labeled cell, and 3) the area covered by enhanced grains was at least 5 times the background in other areas of the brain. Approximately 50 cells/brain section met these criteria regardless of the experimental treatment or time of day. Statistical Analysis Multiple statistical tests were applied to determine the effects of hormone treatment and time of day. Serum LH concentrations and POMC gene expression were measured at seven times of day in PRO, OVX, and E 2 -treated rats. To determine the effect of time of day and hormone treatment among these groups, two-way analysis of variance (ANOVA) was applied. To determine whether a diurnal rhythm exists within each of the first three experimental groups, one-way ANOVA was performed using all time points. If a significant effect of time of day was detected, Duncan's multiple range test was applied to evaluate which times of day were different from each other. In E 2 P 4 -treated rats, P 4 was administered at 1030 h, and the level of POMC mrna was quantified only at the four times of day after P 4 treatment. Therefore, to determine the effect of treatment, two-way ANOVA was performed using the data from h from all four experimental groups. If an effect of treatment was detected, Duncan's multiple range test was used to assess which groups were different from each other. In addition, a one-way ANOVA was used to determine the effect of time of day on POMC mrna in P 4 -treated rats. P < 0.05 was considered significant. Acknowledgments We wish to thank Ms. Klara Roth and Ms. Katherine Rosewell for their superb technical assistance. In addition, we thank Dr. Robert Steiner and members of his laboratory for teaching us the in situ hybridization methodology and for providing the linearized POMC cdna. We are grateful to Dr. Anne Hirshfield, whose expertise was essential for photographing the brain section and POMC-labeled cells. Received January 3,1990. Revision received February 28, Accepted February 28,1990. Address requests for reprints to: Dr. Phyllis M. Wise, Department of Physiology, University of Maryland School of Medicine, 655 West Baltimore Street, Baltimore, Maryland This work was supported by NIH Grants AG and HD-15955(toP.M.W.). * NIH MERIT Awardee. ttrainee on NIH Training Grant HD i Partially supported by NIH Grant AG REFERENCES 1. Kalra SP 1986 Neural circuitry involved in the control of LHRH secretion: a model for preovulatory LH release. In: Ganong WF, Martini L (eds) Frontiers in Neuroendocrinology. Raven Press, New York, 1986, vol 9: Kalra SP, Kalra PS 1984 Opioid-adrenergic-steroid connection in regulation of luteinizing hormone secretion in the rat. Neuroendocrinology 38: Meites J, Van Vugt DA, Forman LJ, Sylvester Jr PW, leiri T, Sonntag W 1983 Evidence that endogenous opiates are involved in control of gonadotropin secretion. In: Bhatnagar AS (ed) The Anterior Pituitary Gland. Raven Press, New York, 1983, pp Ching M 1983 Morphine suppresses the proestrous surge of GnRH in pituitary portal plasma of rats. Endocrinology 112: Cicero TJ, Schainker BA, Meyer ER 1979 Endogenous opioids participate in the regulation of the hypothalamicpituitary-luteinizing hormone axis and testosterone's negative feedback control of luteinizing hormone. Endocrinology 104:

7 MOL ENDO Vol 4 No leiri T, Chen HT, Campbell GA, Meites J 1980 Effects of naloxone and morphine on the proestrous surge of prolactin and gonadotropins in the rat. Endocrinology 106: Allen LG, Kalra SP 1986 Evidence that a decrease in opioid tone may evoke preovulatory luteinizing hormone release in the rat. Endocrinology 118: Gabriel SM, Berglund LA, Simpkins JW 1986 A decline in endogenous opioid influence during the steroid-induced hypersecretion of luteinizing hormone in the rat. Endocrinology 118: Gabriel SM, Simpkins JW, Kalra SP 1983 Modulation of endogenous opioid influence on luteinizing hormone secretion by progesterone and estrogen. Endocrinology 113: Akabori A, Barraclough CA1986 Gonadotropin responses to naloxone may depend upon spontaneous activity in noradrenergic neurons at the time of treatment. Brain Res 362: Leadem CA, Kalra SP 1985 Effects of endogenous opioid peptides and opiates on luteinizing hormone and prolactin secretion in ovariectomized rats. Neuroendocrinology 41: Pang CN, Zimmermann E, Sawyer CH 1977 Morphine inhibition of the preovulatory surges of plasma luteinizing hormone and follicle stimulating hormone. Endocrinology 101: Cicero TJ, Badger TM, Wilcox CE, Bell RD, Meyer ER 1977 Morphine decreases luteinizing hormone by an action on the hypothalamic-pituitary axis. J Pharmacol Exp Ther 203: Wardlaw SL, Wehrenberg WB, Ferin M, Antunes JL, Frantz AG 1982 Effect of sex steroids on beta-endorphin in hypophyseal portal blood. J Clin Endocrinol Metab 55: Weiland NG, Wise PM 1990 Estrogen and progesterone regulate opiate receptor densities in multiple brain regions. Endocrinology 126: Blum M, Roberts JL, Wardlaw SL 1989 Androgen regulation of proopiomelanocortin gene expression and peptide content in the basal hypothalamus. Endocrinology 124: Chowen-Breed JA, Clifton DK, Steiner RA 1989 Regional specificity of testosterone regulation of proopiomelanocortin gene expression in the arcuate nucleus of the male rat brain. Endocrinology 124: Chowen-Breed J, Fraser HM, Vician L, Damassa DA, Clifton DK, Steiner RA 1989 Testosterone regulation of proopiomelanocortin messenger ribonucleic acid in the arcuate nucleus of the male rat. Endocrinology 124: Wilcox JN, Roberts JL 1985 Estrogen decreases rat hypothalamic proopiomelanocortin messenger ribonucleic acid levels. Endocrinology 117: Wardlaw SL, Thoron L, Frantz AG 1982 Effects of sex steroids on brain beta-endorphin. Brain Res 245: Sarkar DK, Yen SSC 1985 Changes in beta-endorphinlike immunoreactivity in pituitary portal blood during the estrous cycle and after ovariectomy in rats. Endocrinology 116: Thind KK, Goldsmith PC 1988 Infundibular gonadotropinreleasing hormone neurons are inhibited by direct opioid and autoregulatory synapses in juvenile monkeys. Neuroendocrinology 47: Adler BA, Crowley WR 1984 Modulation of luteinizing hormone release and catecholamine activity by opiates in the female rat. Neuroendocrinology 38: Johnson MD, Crowley WR 1984 Effects of opiate antagonists on serotonin turnover and on luteinizing hormone and prolactin secretion in estrogen- or morphine-treated rats. Neuroendocrinology 38: Van Vugt DA, Aylsworth CF, Sylvester PW, Leung FC, Meites J 1981 Evidence for hypothalamic noradrenergic involvement in naloxone-induced stimulation of luteinizing hormone release. Neuroendocrinology 33: Maurer RA 1982 Relationship between estradiol, ergocryptine, and thyroid hormone: effects on prolactin synthesis and prolactin messenger ribonucleic acid levels. Endocrinology 110: Maurer RA 1982 Estradiol regulates the transcription of the prolactin gene. J Biol Chem 257: DePaolo LV, Shander D, Wise PM, Barraclough CA, Channing CP1979 Identification of inhibin-like activity in ovarian venous plasma of rats during the estrous cycle. Endocrinology 105: Woodruff TK, D'Agostino J, Schwartz NB, Mayo KE 1988 Dynamic changes in inhibin messenger RNAs in rat ovarian follicles during the reproductive cycle. Science 239: Hargrove JL, Schmidt FH 1989 The role mrna and protein stability in gene expression. FASEB J 3: Millington WR, Blum M, Knight R, Mueller GP, Roberts JL, O'Donohue TL 1986 A diurnal rhythm in proopiomelanocortin messenger ribonucleic acid that varies concomitantly with the content and secretion of beta-endorphin in the intermediate lobe of the rat pituitary. Endocrinology 118: Barden N, Merand Y, Rouleau D, Garon M, Dupont A 1981 Changes in the beta-endorphin content of discrete hypothalamic nuclei during the estrous cycle of the rat. Brain Res 204: Kerdelhue B, Karteszi M, Pasqualini C, Reinberg A, Mezey E, Palkovits M 1983 Circadian variations in beta-endorphin concentrations in pituitary and in some brain nuclei of the adult male rat. Brain Res 261: Wardlaw SL, Wang PJ, Frantz AG 1985 Regulation of beta-endorphin and ACTH in brain by estradiol. Life Sci 37: DePaolo LV, Barraclough CA 1979 Dose dependent effects of progesterone on the facilitation and inhibition of spontaneous gonadotropin surges in estrogen treated ovariectomized rats. Biol Reprod 21: DePaolo LV, Barraclough CA 1979 Interactions of estradiol and progesterone on pituitary gonadotropin secretion: possible sites and mechanisms of action. Biol Reprod 20: Wehrenberg WB, Wardlaw SL, Frantz AG, Ferin M 1982 Beta-endorphin in hypophyseal portal blood: variations throughout the menstrual cycle. Endocrinology 111: Lustig RH, Pfaff DW, Fishman J 1988 Induction of LH hypersecretion in cyclic rats during the afternoon of oestrus by oestrogen in conjunction with progesterone antagonism or opioidergic blockade. J Endocrinol 117: Lustig RH, Pfaff DW, Fishman J 1988 Opioidergic modulation of the oestradiol-induced LH surge in the rat: roles of ovarian steroids. J Endocrinol 116: Kalra SP 1975 Observations on facilitation of the preovulatory rise of LH by estrogen. Endocrinology 96: Konig JFR, Klippel RA 1963 The Rat Brain. A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem. Krieger, Huntington, pp Rogers KV, Vician L, Steiner RA, Clifton DK 1987 Reduced preprosomatostatin messenger ribonucleic acid in the periventricular nucleus of hypophysectomized rats determined by quantitative in situ hybridization. Endocrinology 121:90-93